19th 2 Days International Conference on “Hydraulics, Water Resources, Coastal and Environmental Engineering( HYDRO 2014 International)” December 18-20, 2014 Organized by Department of Civil Engineering, MANIT Bhopal Maulana Azad National Institute of Technology Bhopal (Madhya Pradesh) India Pin -462051 Web : www.manit.ac.in In association with International journal of scientific engineering and Technology (ISSN : 2277-1581) Website : www.ijset.com and email : [email protected]

International journal of Engineering Research ISSN:2319-6890)(online),2347-5013(print) Website : www.ijer.in and email : [email protected]

Indexing of journals : google scholar, DOAJ, endnote, OALIB and many more

S. N.

Title & Authors Names

Page

1.

Assessment Of Hydropower Potential In Nethravathi River Basin Using Swat Model M P Shobhita, Santosh Babar, H Ramesh

1

2.

Water And Sediment Yield Modeling For Micro Watershed Nagargoje Sonali R, D G Regulwar

4

3. 4.

Approaches To Hydrological Modeling Of The Heterogeneous Catchment Of The Dal Lakes Raazia, R Khosa Probability Analysis For Estimation Of Annual One Day Maximum Rainfall Of Devgarhbaria Station Of Panam Catchment Area Kapil Shah, T M V Suryanarayana

7 11

Theme: Hydraulics Of Spillway And Energy Dissipators

5. 6.

Experimental And Three Dimensional Numerical Studies For A Sluice Spillway A Kulhare, M R Bhajantri Physical Model Study For Energy Dissipation Arrangements To The Pick Up Weir Across Pachaiyar River In Tamilnadu C Prabakar, P K Suresh, T Ravindrababu , A Parthiban, A Muralitharan

15

19

Theme: Hydraulic Structures

7. 8.

Experimental Investigations For Estimation Of The Height Of Training Wall Of Convergent Stepped Spillway P J Wadhai, N V Deshpande, A D Ghar Studies For Location Of Bridges In The Vicinity Of Existing Hydraulic Structures B Raghuram Singh, R G Patil, M N Singh

23 27

9.

Study Of Sharp-Crested Triangular Weir M Shaheer Ali, Talib Mansoor

31

10.

Study Of Elliptically Shaped Sharp Crested Weirs N P Singh, R Singh

35

11.

Turbulence Characteristics Of Flow Past Submerged Vanes H Sharma , Z Ahmad

38

12.

Hydraulic Design Aspects Of Stilling Basin With Sloping Apron V S Rama Rao, K T More, M R Bhajantri, V V Bhosekar

42

13.

Hydraulic Design Of Barrage In Montane Terrains Rajendra Chalisgaonkar , Mukesh Mohan, Manish S Sant, Pratibha S Sant.

46

14.

Optimal Design Of Intake Upstream Of A Weir – A Case Study Kuldeep Malik, R G Patil, M N Singh

15.

Study Of Effect On The Stresses & Safety Of Gravity Dam With Changes In Width Parameter B S Ruprai, A D Vasudeo

50

55

Theme: Integrated Watershed Management

16. 17.

Assessment Of Environmentally Stressed Areas For Soil Conservation Measures Using Usped Model Bikram Prasad, R K Jaiswal, H L Tiwari. A Novel Optimisation Model Applied To Godavari River Basin R B Katiyar, Balaji Dhopte, Tejeswi Ramprasad, Shashank Tiwari, Anil Kumar, K R Gota

58 63

18.

The Effect Of Parthenium Hysterophorus Weed On Basin Hydrology Soham Adla,Shivam Tripathi

19.

Runoff And Sediment Yield Modeling Of An Agricultural Hilly Watershed Using Wepp Model Saroj Das, Laxmi Narayan Sethi, R K Singh

66

20.

Prioritization Of A Watershed Based On Spatially Distributed Parameters C D Mishra, R K Jaiswal, A K Nema

70

Theme: Rehabilitation Of Dams

21.

Stability Assessment Of Chang Dam After Rehabilitation R Singh, D Roy

22.

Rehabilitation And Improvement Of Sher Tank Project Vishnu Arya.

Theme: Reservoir Operation And Irrigation Management

23. 24. 25. 26. 27. 28. 29.

Water Balance Assessment Of Krishna River Basin Through System Simulation N S R Krishna Reddy, S K Jain. Minimization Of Conveyance Losses For Nashik Left Bank Canal [Nlbc] By Closed Conduit Irrigation [Cci] Gayatri R Gadekar, Sunil Kute, N J Sathe. Methods For Estimation Of Crop Evapotranspiration Using Climate Data: A Review Gopal H Bhatti, H M Patel Estimation Of Deep Percolation From Rice Paddy Field Using Lysimeter Experiments On Sandy Loam Soil Hatiye Samuel D, K S Hari Prasad, C S P Ojha, G S Kaushika. Reservoir Modelling In Bearma Basin By Using Mike Basin Shikha Sachan, T Thomas, R M Singh, Pushpendra Kumar Replacement Of Field Channels With Pressurized Irrigation Systems: In Ssp Command Area Sahita I Waikhom, Monali Patel, P G Agnihotri Reservoir Operation Based On Real Time Flow Data For Flood Control And Incremental Power Generation Rameshwar Prasad Pathak

76 81 86 92 96 99

Theme: Reservoir Sedimentation And Irrigation Management

30.

Effect Of Conservation Works On Soil Erosion-A Case Study Of Punegaon Reservoir Catchment Area M B Nakil, M Vkhire.

31.

Sediment Trap Efficiency Of Porcupine Systems For Riverbank Protection Mohd Aamir, Nayan Sharma

Sedimentation Assessment In Nath Sagar Reservoir (Jayakwadi Project) Of 32. Maharashtra By Remote Sensing Technique – A Case Study Prakash Bhamare, Manoj Bendre, Ravindra Shrigiriwar, Mahendra Nakil, Sudhir Kalvit.. Theme: Risk Reliability Analysis And Design

103 107

112

33.

Hydrological Data Modelling Using Wavelet, Neural Network And Ar Models G.Khadanga, B.Krishna.

115

34.

Improved Neuro-Wavelet Model For Reservoir Inflow Forecast B.Krishna, Y R Satyaji Rao, R.Venkata Ramana.

118

35.

Application Of Particle Swarm Optimization In Multiobjective Irrigation Planning D V Morankar, K Srinivasa Raju, A Vasan, L Ashoka Vardhan

121

36.

Artificial Neural Network Model For Design Of Air Vessel For Controlling The Water Hammer Pressures N Mowlali, E Venkata Rathnam

126

37.

Monthly Inflow Prediction Using Wavelet Neural Network Rutuja Patil, J N Patel, S M Yadav, D G Regulwar.

131

38. 39. 40.

Improving Location Specific Wave Forecast Using Using Soft Computing Techniques S N Londhe, P R Dixit, B Nair T M, A Nherakkol Discrete Wavelet Support Vector Conjuction Model For Significant Wave Height Time Series Forecasting Paresh Chandra Deka, Y N Suryadatta. Potential Impact Of Soft Computing Techniques In Water Resources Engineering Satish Kumar Jain, R K Shrivastava

134 139 143

Theme: Water And Waste Water Management

41.

Typologies For Successful Operation And Maintenance Of Horizontal SubSurface Flow Constructed Wetlands Lohith Reddy D, Dinesh Kumar, Shyam R Asolekar.

147

42.

A Mini Review On Fixed Film Reactor For Wastewater Treatment Saraswati Rana, S Suresh

155

43. 8 Technological Utilization Of Parthenium Hysterophorus-A Review S.Arisutha, R.B. Katiyar And S. Suresh

159

Theme: Water Quality Assessment And Modeling

44.

Water Quality And Flow Simulation Along River Amarsinh B. Landage..

160

45.

Assessment Of Groundwater Quality Of Bah Block, Agra, India Azmatullah Noor,Dr. Izharul Haq Farooqi

165

46.

Changing Water Quality Scenarios Of Tank Cascade System And Its Implications J Hemamalini, B V Mudgal, J D Sophia.

171

47. 48.

Booster Chlorination Strategy For Managing Chlorine Disinfection In Drinking Water Distribution System – A Review Roopali V Goyal, H M Patel Hydrogeochemical Stuidies Of Groundwater In And Around Metropolitan City Vadodara, Gujarat, India M K Sharma, C K Jain

175 180

49.

Evaluation Of Various Objectives In Multi-Objective Sensor Placements In Water Distribution Systems S Rathi, R Gupta

185

50.

Water Quality Assessment Of Dal Lake, Kashmir, J&K Shabina Masoodi.

191

51. 52.

Spatial Water Quality Analysis Of Nagalamadike Watershed Of Pavagada Taluk, Tumkur District Karanataka Using Geo Informatic Tools Nandeesha, C Ravindranath, T Gangadaraiah, S G Swamy Water Pollution In Ganga River Susmita Saha

196 203

Theme: Water Resource And Hydrology

53. 54. 55. 56. 57. 58.

59.

Flood Frequency Analysis Using A Novel Mathematical Approach Bidroha Basu,V V Srinivas Performance Comparative Of Wavelets And Savitzky-Golay Filter On Bathymetry Survey Data M.Selva Balan1 Arnab Das2 Simulation Study On Performance Of Household Rainwater Harvesting Systems P.G. Jairaj1 P. Athulya2 COMMUNITY-BASED WATER RESOURCE MANAGEMENT, STUDY AREA NAWLI VILLAGE, MEWAT DISTRICT, HARYANA Amit Kumar Dogra1 Singh2SOIL WATER SIMPLE MODEL TO Nitin ESTIMATE RETENTION LIMITS OF CHATTISGARH STATE N.G.Pandey1, B. Chakravorty1, Sanjay Kumar2 & P. Mani1 Land Cover Classification By Ls-Svm With Landsat Satellite Imagery Shilpi1 R.M. Singh2 Assessing Impacts Of Landuse/Landcover Change On Surface Runoff For Kadalundi River Basin: A Watershed Modeling Approach Sinha R. K.1, Eldho T. I.2, Ghosh S.2

209 213 219 222 227 230

234

60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72.

Impact Of Land-Use Land-Cover Changes On Runoff Generation In A Bangalore Urban Catchment R. L. Gouri1 V. V. Srinivas2 Change Detection In Land Use/Land Cover Using Remote Sensing And Gis – A Case Study For Ur Basin In Tikamgarh District S. Karwariya1* Goyal2 V. C. Goyal3 T. Thomas4 Multi ObjectiveS.Optimization Of Cropping Pattern In A Canal Command Area Paritosh Srivastava1 and Raj Mohan Sing Urban Watershed Rainfall Forecast Of Chennai City R. Venkata Ramana, B Krishna,Y. R. S. Rao and V.S.Jayakanthan Agriculture Water Consumption In Madhya Pradesh – An Analysis From Virtual Water Perspective Vivek K. Bhatt1Of Dr.Generalized J.S. Chouhan2 Development Neural Network Based Eto Models From Limited Climatic Data For Different Agro-Ecological Regions In India Sirisha Adamala1 Raghuwanshi2 Mishra3 Estimating FloodN.S. Inundation Using Ashok Hec-Ras And Regression Models R.S. Meena1 R. Jha2 and K.K. Khatua3 Analysis Of Penman – Monteith Sensitivity Method For Estimation Of Evapotranspiration Ch.V.S.S. Sudheer1 Dr.G.K.Viswanadh2 Dr.G.Venkata Ramana3 In Sediment Size Of Two SubVariability 1Asst. Professor, GRIET, Hyderabad Catchment Areas Of Ganga Basin, Western Himalayas M.Y.A. Khan* S.Study Panwar Comparative Of Double Ring And Tension Infiltrometers To Measure Infiltration Properties And Hydraulic Conductivity B. Ghosh1P. Spatial AndSreeja2 Temporal Distribution Of Rainfall Trends In Bist-Doab Region Of Punjab (1901– 2010) M. K.Radiation Nema1 S. K. Jain1 P.K. From Mishra1 Net Estimation A Remotely Sensed Data Using Sebal Model M.V.S.S.Giridhar1 and P. Suneel2 Low Flow Analysis In Bina River Basin Of Madhya Pradesh V.K. Chandola1*, Sunil Kumar Yadav1, R.V. Galkate3, Palak Mehata4

238 243 247 252 257 261 267 272 277 281 285 291 295

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Assessment of Hydropower Potential in Nethravathi River Basin Using Swat Model 1

Shobhita M. P1, Santosh Babar2, H. Ramesh3 Lecturer, Dept. Civil Engineering, JSS Academy of Technical Education, Mauritius 2 Research Scholar, Dept. of Applied Mechanics and Hydraulics, National Institute of Technology Karnataka, Surathkal, Mangalore-575025, India 3 Assistant Professor, Dept. of Applied Mechanics and Hydraulics, National Institute of Technology Karnataka, Surathkal, Mangalore-575025, India Email: [email protected], [email protected], [email protected],

Abstract: Hydropower plants have the advantage of producing renewable and clean power, the renewable and reliable energy source that serves national environmental and energy policy objectives. Therefore, the development of hydropower plant and improvements of water management have essential in contributing to sustainable growth and energy reduction in developing countries like India. The present study is concerned with the development of methodology and assessment of hydro power potential in Nethravathi River Basin with the help of Remote Sensing and GIS. The catchment area covers 3200 km2, where most of the land cover is dominated by forest. The basin was divided into six sub-basins based on hydrology and topography using GIS tools. The climate over the basin is coastal, humid tropical and receives an average annual rainfall of about 4000 mm. sub-basin discharges were estimated using SCS curve number method. To ensure the total discharge from six sub-basins computed from SCS curve number method, the flows were routed and simulated at the outlet using Soil and Water Assessment Tool (SWAT).Streamflow calibration was carried out at monthly time steps for the period of 1998–2001, and validated for 2002– 2003. Flow-duration curves (FDC) were generated for individual sub-basins. The results have shown a good agreement between observed and the simulated flows. The available discharge at 75%, 80% and 90% of time for each sub-basin were extracted from the FDC. This information was used to calculate the hydro power potential in all five subbasins at Q75, Q80 and Q90, by integrating thematic layers using ArcSWAT. Keywords: Flow Duration Curve, GIS, Hydropower, Nethravathi Basin, Remote Sensing, SWAT model 1. INTRODUCTION Energy supply is an important key parameter in the economic development of a country. Hydroelectric Power is a form of energy, a renewable resource. There are several sources of energy that is being used by human beings, such as thermal, nuclear, geothermal. One of them is hydro power which is one of the oldest and the most reliable and environment friendly source of all renewable energy. The use of fossil energy sources contributes to environmental problems such as global warming, acid rain, and desertification. Under these circumstances, demands for the development of non-fossil energy sources grow

HYDRO 2014 International

significantly. Hydropower is a renewable energy sources that do not emit the carbon dioxide and other flue gases that contaminate the environment. It has the least adverse environmental impact (i.e. greenhouse gas, SO2, NOx emission) and has the most energy payback ratio when compared among all electricity generation systems. One Gega Watt of electricity produced by small hydropower means a reduction of CO2emissions by 480 tons (Kusre et.al., 2010). Hydropower is an indigenously available, clean and renewable source of energy. The broad application of GIS and remote sensing technology for digital mapping, river morphology studies, terrain analysis, the integration of socio-economic variables and for modeling and simulation play very crucial roles in hydropower development (Pathak Mahesh., 2008). The hydropower development shows the advantages based on economic, environmental and social front as the reliable service, long life (50 to 100 years), no atmospheric pollutants, can create a new freshwater ecosystem with increased productivity, often provides flood protection and it helps in sustainable development(Nguyen Trung Dung., 2009). The Indian economy uses a variety of energy sources, both commercial and non-commercial. Fuel wood, animal waste and agricultural residue are the traditional or non commercial sources of energy that continues to meet the bulk of the rural energy requirements even today. However, the share of these fuels in the primary energy supply has declined from over 70% in the early 50's with a little over 30% as of today. The Ministry of Power has set on the objective of providing "Power for all by 2012". This will entail electrification of all villages by 2007 and of all households by 2012. It is also a known fact that electricity is one of the key infrastructure elements for the economic growth of the country. The existing power deficit and a rapid growing demand have necessitated a large scale capacity power addition programme. Severe power shortage is one of the greatest obstacles to any country‟s development. Power the most important need in the modern world. Hydropower development needs integrated approaches to analyzing natural resources, physiographic setting and the socio-economic indicators. GIS is also used to input, store, retrieve, manipulate, analyze and output geographically referenced data or geospatial data, in order to support decision making for planning and management of land use, natural resources, environment, transportation, urban facilities, and other administrative records (Dudhani S, 2006). The many studies on hydropower location and hydropower potential has been carried out using GIS and remote sensing methodology (Arun et al; 1995, Pannathat et al; 1998, Balance et al; 2000, Kupakrapinyo and Chaisomphob 2003, Santasmita Das and Paul, P.K. 2006, Choong-Sung et al; 2010, Vani 2010, Shobitha 2012).The scope of the present study is in development of SWAT (Soil and Water Assessment Tool) model which will help to evaluate the hydropower potential within watershed. 2. STUDY AREA Nethravathi River is one of the major west flowing rivers in Karnataka. The geographical location of the Nethravathi river basin lies between 12º29'11" to 13º11´11" N latitudes and 74º49´08" to 75º47´53" E longitudes as shown in figure 1. The

MANIT Bhopal

Page 1

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Nethravathi River originates in the south of Samse village, at an altitude of approximately 1200 meters from mean sea level in the Western Ghats of Karnataka. The river flows towards westward for about 103 kilometers with a drainage area of 3657 km2 (Shobitha, 2012) and empties into the Arabian Sea at Mangalore city. The river is joined by Mundaja Neriya, Shishla Uppar, Kumaradhara and Beltangady nallas from either side. Average annual rainfall in the region is about 3930 mm with 90% of the rainfall contribution from South west monsoon (June – September) alone and rest during pre and post monsoon. Nethravathi River provides water supply for Mangalore city, industries, hydropower production and agricultural activities in the basin. 3. MATERIALS AND METHODOLOGY 3.1 Data used Daily rainfall data were collected for six years from nine rain gauge stations. The Daily gridded climate record for a period of six years (1998-2003) including precipitation and temperature were obtained from IMD (India Meteorological Department), land use land cover of year 2003, soil data, DEM (topography data). 3.2 SWAT model development SWAT is a river basin scale model that operates on a daily, monthly time-step. It was developed at the University of Texas, USA. Major components of the SWAT model include hydrology, weather, erosion, soil temperature, crop growth, nutrients, pesticides, and agricultural management (Neitsch et al; 2001b).

SW = The final soil water content (mm), SW = The water t

0

content available for plant uptake, defined as the initial soil water content minus the permanent wilting point water content (mm), t = Time in days, R = Rainfall (mm), Q = Surface day

surf

runoff (mm), E = Evapotranspiration (mm), w a

= Percolation seep

(mm) and Q = Return flow (mm) gw

3.3 Model calibration and validation Understanding the model processes, checking the various components such as rainfall to runoff ratio, ET, base flow contribution, etc. are very important. To make sure all the major components are represented well for a watershed before attempting either manual or auto-calibration. The model contains both manual and auto-calibration tools. In this study, model parameters will calibrate using the observed daily flow. Sensitivity analysis will be conducted for the SWAT model to guide calibration process. Figure 2 represents the detailed methodology which was used during the research work. 3.4 Estimating power output Figure 2. Methodol ogy Flowchart

Figure 1. Location map of study area The computation of hydrologic processes operates in five phases: (1) precipitation, interception, (2) surface runoff, (3) soil and root zone infiltration, (4) evapotranspiration and soil and snow evaporation, and (5) groundwater flow. A water balance equation calculates the change in soil water content (SWt) as: (1) where:

HYDRO 2014 International

MANIT Bhopal

Page 2

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

3.4.1 Head The head is the vertical distance that waterfall. It is usually measured in meters. The higher head consumes less water to produce a given amount of power, and can use smaller, less expensive equipment. Low head refers to a change in elevation of less than 10 feet (3 meters). When determining head, both gross head and net head need to be considered. The gross head is found by considering the difference of head between weir and power house. Net head equals gross head minus losses due to friction and turbulence in the piping. Hydraulic power can be captured wherever a flow of water falls from a higher level to a lower level. The vertical fall of the water, known as the “head”, is essential for hydropower generation; fast-flowing water on its own does not contain sufficient energy for useful power production except on a very large scale. Hence two quantities are required: a flow rate of water (Q), and head (H). It is generally better to have more head than more flow, since this keeps the equipment smaller. The Gross head (H) is the maximum available vertical fall in the water, from the upstream level to the downstream level. The actual head seen by a turbine will be slightly less than the gross head due to losses incurred when transferring the water into and away from the machine. This reduced head is known as the net head as shown in figure 3. 3.4.2 Identification of sites having a suitable head  The possible potential sites for power houses along streams based on the gross head were located at the intersection points of contour lines and streams.  For this purpose a set of contour lines with intervals of 6, 10 and 20 m were generated from ASTER DEM.  The flow accumulation map has been created by using the flow direction map. The flow accumulation function calculates accumulated flow, as the accumulated weight of all cells flowing into each down slope cell in the output raster.  Suitable sites were identified by using the DEM and the flow accumulation map.  The flow accumulation map was used to locate weir and powerhouse on the high flow accumulated stream.

variability is through flow-duration curves. The flow-duration curve of a stream is based on daily mean discharges (not peak flows) and shows the percentage of time that a given daily mean discharge is equalled or exceeded. A flow-duration curve is a plot of discharge against the percent of time the flow has equalled or exceeded. Flow-duration curves are extremely useful in evaluating various dependable flows in the planning of water resources engineering projects, the characteristic of the hydropower potential of a river. The stream flow data are arranged in a descending order of discharges, using class intervals. The data can be daily, weekly or monthly values. Chiang et al. (2002) stated that monthly stream flow data satisfy the basic data requirement for water resource projects. If N numbers of data points are used in this listing, the plotting position of any discharge Q is (2) Where, Pp = percentage of probability of the flow magnitude being equalled or exceeded, m = the order number of the discharge, N = Total count (Number of data) 3.4.4 Estimating power output The dependable flows (Q90, Q80,Q75) and head substitute in the power equation to determine the power output from each subbasin. P=ηρQgH (3) Where, P = mechanical power produced at the turbine shaft (Watts), η = hydraulic efficiency of the turbine, ρ = density of water (1000 kg/m3), g = acceleration due to gravity (9.81 m/s2), Q = volume flow rate passing through the turbine (m3/s), H = head of water across the turbine (m). 4. RESULTS AND DISCUSSIONS From figure 3 represents sub-basin wise monthly average flow during the months of June to January. It can be observed that sub-basin 3 has the highest amount of discharge compared to the rest of the sub-basins. This is due to the presence of C group of HSG soil, which are moderately high runoff potential soil type and very high rainfall. Sub-basin 4 has the lowest amount of discharge. This is due to the presence of A group of HSG, which are low runoff potential soil type

Figure 3. Measurement of head (Sale Michael et al., 2006) 3.4.3 Flow-Duration Curves (FDC) It is well known that the stream flow varies over a water year. One of the popular methods of studying the stream flow

HYDRO 2014 International

Figure 4. Monthly average flow (cms) Sub-basin wise

MANIT Bhopal

Page 3

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

4.1 Flow duration curves (FDC) The stream flow data are arranged in a descending order of discharges, using class intervals. The data used can be daily, weekly or monthly values. If N numbers of data points are used in this listing, the plotting position of any discharges Q in equation 2. Table 1 gives the dependable flows of all sub-basins, which are derived from flow duration curves. Table.1 shows the flow quantiles derived from the flow duration curves for 90%, 80% and 75% for each sub-basins. These flow quantiles were used for power estimation. Table 1. Flow quantiles from each sub-basin Discharge Q in Cumecs

Subbasin No

Q90

Q80

Q75

1

380

540

625

2

364

508

580

3

389

564

651

190 272 313 4 4.2 Estimation of power (P) The flow quantiles estimated from FDC and hydraulic head determined from DEM are substituted in equation 3 to estimate power in watts for each sub-basin. The turbine efficiency (η) was taken as 85% for Kaplan turbine. The value of power in mega watts is tabulated in Arc GIS, as shown in Table 2. Table 2. Sub-basin wise power potential Sub-basin No

Power in Mega Watts Q90

Q80

Q75

1 2

63.37 60.70

90.06 84.72

104.23 96.73

3

64.87

94.06

108.57

4

31.69

45.36

52.20

5. CONCLUSIONS  In the study area it was found from the hydrographs that the months June, July, August and September had shown a greater amount of discharge and so shows maximum hydropower potential.  Sub-basin - 3 leads to the highest amount of average monthly discharge during June to January months about 3000 m3 /s when compared to the remaining sub-basins. This is mainly due to presence of C group of HSG soil, which are moderately high runoff potential soil type and very high rainfall.  Using the flow duration curve, it is possible to estimate the percentage of time that a specified flow is equalled to or exceeded, once we know the amount of discharge that will be available for 90% of the time from the flow duration curve, we can calculate the power that can be produced.  SWAT model was calibrated and validated, the R2 (coefficient of correlation) in calibration equal to 0.91 and in validation equal to 0.92. The discharge of June, July and gradually reduces towards August, September and October, with

HYDRO 2014 International

little discharge during the months of November and December and January. 6. REFERENCES i. Arun K.S (1995) ―GIS in small hydro planning resource management‖ Department of Earth Sciences, University of Roorkee, Alternate Hydro Energy Centre, University of Roorkee. ii. Ballance, D, Stephenson, R.A, Chapman, Muller, J (2000) ―A geographic information systems analysis of hydro power potential in South Africa‖ Journal of Hydro-informatics iii. Choong S.Y, Jin, H.L, Myung, P.S (2010) ―Site location analysis for small hydropower using geo-spatial information system‖ Journal of Renewable Energy, 35: 852–861. iv. Dudhani S. (2006). ―Small hydropower and GIS for sustainable growth in energy sector‖ Map India 2006. v. Kupakrapinyo C, Chaisomphob T (2003) ―Preliminary Feasibility Study on Run-of-River Type Hydropower Project in Thailand: Case Study in Maehongson Province‖ Proceedings of the 2nd Regional Conference on Energy Technology towards a Clean Environment 12-14 February 2003, Phuket, Thailand. vi. Kurse, B.C., Baruah, D.C., Bordoloi, P.K., Patra, S.C (2010) ―Assessment of hydropower potential using GIS and hydrological modeling technique in Kopili River basin in Assam India‖ Journal of Applied Energy, 87: 298–309. vii. Mahesh, P. (2008) ―Application of GIS and Remote Sensing for Hydropower Development in Nepal‖ Hydro Nepal Issue No. 3, 1-4 viii. Michael, S. (2006) ―Hydropower Summary‖ Department of Energy Biennial report. ix. Nguyen, T.D. (2009) ―Sustainable hydropower development‖ DAAD, Germany. x. Neitsch, S.L., Arnold, J., Kiniry, J.R., Srinivasan, R., Williams, J.R. (2001b). ―SWAT theoretical documentation version 2009.‖ xi. Pannathat R, Taweep C, Thawilwadee B. (2009) ―Application of GIS to site selection of small run-of-river hydropower project by considering engineering/economic/environmental criteria and social impact‖. Journal of Renewable and sustainable energy reviews, 13: 2336–2348 xii. Santasmita D, Paul, P.K. (2006) ―Selection of Site for Small Hydel Using GIS in the Himalayan Region of India‖. Journal of Spatial Hydrology, 6: 18-28. xiii. Shobitha, M.P. (2012) ―Estimation of hydropower potential in Nethravathi river basin using RS and GIS.‖ M.Tech thesis, NITK Surathkal. xiv. Vani, S. (2011) ―Mapping of suitable sites for small hydropower generation using RS and GIS‖, M.Tech thesis, NITK Surathkal.

Water and Sediment Yield Modeling for Micro Watershed Nagargoje Sonali R1 & D.G.Regulwar2 1. Research scholar, Dept. of Civil Engineering, Govt. college of engineering, Aurangabad, Maharashtra state India 2. Associate Professor, Dept. of Civil Engineering, Govt. college of engineering, Aurangabad, Maharashtra state India Email: 1. [email protected] 2. [email protected] ABSTRACT: Land is the most important natural resource, which embodies soil, water and associated flora and fauna involving the total ecosystem. Now a days degradation of land from water-induced soil erosion is becoming a serious global

MANIT Bhopal

Page 4

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

problem, which is not only eroding the top fertile soil but is also responsible for swelling of river beds and reservoirs thereby causing floods and reduction in the life span of costly reservoirs and dams. Reliable estimates of soil erosion, sediment and water yield are, therefore, required for design of efficient erosion control measures, reservoir sedimentation assessment, and evaluation of watershed management strategies. Watershed parameters such as channel network, location of drainage divides, water and sediment yield of the catchment etc are obtained from maps or field surveys traditionally. Since last two decades this information has been increasingly derived directly from digital representations of the topography. Measurement of sediment yield on a number of watersheds is operationally difficult, expensive, time consuming, and tedious. Therefore modeling is carried out for generating the sediment yield data base. Present study explores development of sediment and water yield model for micro watershed (627 ha) located in Khuldabad village of Aurangabad District, Maharashtra state India. For this topographical features such as LULC, soil map and DEM are prepared under GIS environment and meteorological data like temperature and rainfall has been made in gridded format. SWAT divided watershed into HRUs by merging digital elevation model land use and soil pattern. Annual average basin value for water and sediment yield for present study are 24.96 mm and 1.035 T/ha respectively. The study reveals the values and areas of sediment sources from the watershed which helps in adopting suitable soil conservation practices in basin. Keywords: Land use/ Land cover, Digital Elevation Model, Soil and Water assessment tool, Hydrological response units 1. INTRODUCTION Soil erosion/sedimentation is an immense problem that has threatened water resources development in all over the world. An insight into soil erosion/sedimentation mechanisms and mitigation methods plays an imperative role for the sustainable water resources development. This paper presents daily sediment yield and water yield simulations in micro watershed under different Best Management Practice (BMP) scenarios. The Soil and Water Assessment Tool (SWAT) was used to model soil erosion, identify soil erosion prone areas and assess the impact of BMPs on sediment reduction. For the existing conditions scenario, the model results showed a satisfactory agreement between daily observed and simulated sediment concentrations as indicated by Nash-Sutcliffe efficiency greater than 0.83. However, a precise interpretation of the quantitative results may not be appropriate because some physical processes are not well represented in the SWAT model. Literature review shows there are many catchment models that include the soil erosion/sedimentation processes and simulate the effect of mitigation measures. 2. MATERIALS AND METHODS 2.1 Description of study area Location of case study Watershed GV-41 is a significant drainage system contributing to river Godavari. The watershed lies between longitude 74° 58' 55” and 75° 07' 24” East and latitude 19° 53' 33” and 19° 45' 27” North. It is included in

HYDRO 2014 International

Survey of India topographic sheet No. 47 I/13 and 47 M/1 on 1: 50,000 scale.

Fig.1: Location Map of Study Area 2.2. MODEL INPUT The spatially distributed data (GIS input) needed for the Arc SWAT interface include the Digital Elevation Model (DEM)s 2.2.1 Digital Elevation Model Topography was defined by a DEM that describes the elevation of any point in a given area at a specific spatial resolution. A 90 m by 90 m resolution DEM (Fig. 2) was downloaded from SRTM (Shuttle Radar Topography Mission). The DEM was used to delineate the watershed and to analyze the drainage patterns of the land surface terrain. Sub basin parameters such as slope gradient, slope length of the terrain, and the stream network characteristics such as channel slope, length, and width were derived from the DEM.

Fig.2. DEM of Study Area (Source: SRTM) 2.2.2 Land Use/ Land cover Map and Soil Map Detailed classification of land use /land cover is shown in fig. 3.Database of LULC collected from Bhuvan (NRSC) . Distribution of various soil types among the study area is shown in fig.No.4.

MANIT Bhopal

Page 5

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Fig.3. Detailed LULC classification of study area

Fig.4. Classification of Soil in Study area Table No. 1 Spatial model input data for the Watershed.

soil temperature, crop growth, pesticides agricultural management and stream routing. The model predicts the hydrology at each HRU using the water balance equation, which includes daily precipitation runoff, evapotranspiration, and percolation and return flow components. The surface runoff is estimated in the model using two options (i) the Natural Resources Conservation Service Curve Number (CN) method (USDA-SCS, 1972) and (ii) the Green and Ampt method (Green and Ampt, 1911). The percolation through each soil layer is predicted using storage routing techniques combined with crackflow model (Arnold et al., 1995). The evapotranspiration is estimated in SWAT using three options (i) Priestley-Taylor (Priestley and Taylor, 1972), (ii) Penman-Monteith (Monteith, 1965) and (iii) Hargreaves (Hargreaves and Riley, 1985). The flow routing in the river channels is computed using the variable storage coefficient method (Williams, 1969), or Muskingum method (Chow, 1959). The SWAT model uses the Modified Universal Soil Loss Equations (MUSLE) to compute HRU-level soil erosion. It uses runoff energy to detach and transport sediment (Williams and Berndt, 1977). The sediment routing in the channel (Arnold et al., 1995) consists of channel degradation using stream power (Williams, 1980) and deposition in channel using fall velocity. Channel degradation is adjusted using USLE soil erodibility and channel cover factors. 2.4 SWAT model setup The SWAT model inputs are Digital Elevation Model (DEM), land use map, soil map, and weather data, which is shown in Table 1. The ArcGIS interface of the SWAT2005 version was used to discretize a watershed and extract the SWAT model input files. The DEM was used to delineate the catchment and provide topographic parameters such as overland slope and slope length for each sub basin. The land use map of the Global Land Cover Characterization (GLCC) was used to estimate vegetation and their parameters input to the model. The GLCC is part of the United States Geological Survey (USGS) database, with a spatial resolution of 1 km and 24 classes of land use representation. The parameterization of the land use classes is based on the available SWAT land use classes. The soil types of the study area were extracted from the soil map obtained from NBSS database. 3. RESULTS AND DISCUSSIONS Using above materials and models SWAT model is performed. DEM and LULC,soil map having 8 classification each taken as input. Output is obtained for each subbasin in delineated watershed of study area. Whole SWAT procedure is followed using SWAT Manual 2005. Results of the present study are as shown in Table No.2. Table No.2 Average monthly basin output

2.3 SWAT model description The Soil and Water Assessment Tool (SWAT) is a physical process based model to simulate continuous-time landscape processes at a catchment scale (Arnold et al., 1998; Neitsch et al., 2005). The catchment is divided into hydrological response units (HRUs) based on soil type, land use and slope classes that allows a high level of spatial detail simulation. The major model components include hydrology, weather, soil erosion, nutrients,

HYDRO 2014 International

MANIT Bhopal

Month 01 02 03 04 05

Water Yield (mm) 30.36 32.61 33.86 43.18 60.56

Sediment Yield (T/Ha) 0.19 0.29 0.21 0.43 0.18

Page 6

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

06 47.07 0.02 07 22.86 0.00 08 10.15 0.00 09 12.90 0.01 10 19.70 0.08 11 27.63 0.18 12 27.59 0.16 SWAT gives the average monthly basin values of water and sediment yield in mm and Tonnes/Hector respectively. From the output it seems easier to estimate sediment yield using hydrological model i.e. SWAT. Using this model identification of the soil erosion area becomes easier from which management of sediment yield can be done. Thus SWAT gives each basin values present in watershed through which Soil and Water conservation practices can be done for sustainable development of water resources. 4. CONCLUSIONS: ARC-SWAT is powerful hydrological model to identify erosion prone areas and it is also useful for watershed prioritization. Using hydrological models identification and solution of such critical soil erosion areas is in water resources engineering can be achieved for sustainable development. 5. REFERENCES i. Chen,B. (2012) ―Development of an integrated adaptive resonance theory mapping classification system for supporting watershed hydrological modeling‖ Journal of Hydrologic Engineering, ASCE, vol. 17, pp 679-693 ii. Gabriel, G., 2008 ―Fitting of time series models to forecast stream flow and groundwater using simulated data from SWAT‖, Journal of Hydrologic Engineering, ASCE, pp: 554-562. iii. Gong Y., 2010 ―Effect of watershed subdivision on SWAT modelling with consideration of parameter uncertainty‖, Journal of Hydrologic Engineering, ASCE, December, pp: 1070- 1074. iv. Kim, N.W., 2012 ― Assessment of flow regulation effects by dams in the Han River, Korea, on the downstream flow regimes using SWAT‖, Journal of water resources planning and management, ASCE, pp: 2435. v. Kirby, J.T. and Durrans, S.R., 2007 ―Modelling the combine effect of forests and agriculture on water availability‖, Journal of Hydrologic Engineering, ASCE, pp: 319-326. vi. Mishra A. and Kar S., 2012 ―Modelling hydrologic processes and NPS pollution in a small watershed in sub humid subtropics using SWAT‖, Journal of Hydrologic Engineering, ASCE, pp: 445- 454. vii. Pikounis M. (2003) ―Application of the SWAT model in the Pinios river basin under different land-use scenarios‖ 8th International Conference on Environmental Science and Technology, Vol 5, pp 71-79 viii. Sang, X., and Chen Q, 2010 ―Development of SWAT tool model on human water use and application in the area of high human activities‖, Journal of Irrigation and Drainage Engineering ASCE, pp: 23-30. ix. Setegn, S. G. (2008) ―Hydrological modeling in the Lake Tana Basin, Ethiopia using SWAT model‖ Journal of Hydrology, ASCE, vol.2, pp. 49-62

Approaches to Hydrological Modeling of the Heterogeneous Catchment of the Dal Lake HYDRO 2014 International

S. Raazia1 R. Khosa2 Department of Civil Engineering, Indian Institute of Technology Delhi, New Delhi, 110016, India 2 Department of Civil Engineering, Indian Institute of Technology Delhi, New Delhi, 110016, India Email: [email protected] 1

ABSTRACT: Dal Lake situated in the state of J&K along with its associated wetland system, forms a highly complex and vulnerable hydrological system. The lake catchment comprises of gently to steeply sloping mountains on three sides and a low relief, highly urbanized landscape on one side. Owing to these differences in physical features of the landscape, the catchment exhibits a spatially varying hydrological behaviour. The study identifies the catchment components with dissimilar hydrological response and, in recognition of these distinct but dominant hydrological features, has proposed similarly distinct approaches to hydrological modeling for these appropriately designated sub areas of the overall catchment. Briefly, the entire catchment was divided into 3 subbasins namely (i) DaraDachhigam subbasin with a mild to steep mountainous relief and a prominent network of drainage channels, (ii) Zabarwan subbasin with gently sloping foothills along the lake shore having a backdrop of highly steep mountains further from the lake, and (iii) the urban subbasin consisting of a nearly plain urbanized area and wetlands spread over an undulating topography. In the Dara-Dachhigam subbasin, runoff generation has been modeled in accordance with the Hortonian mechanism using the hydrological model SWAT. The hydrology of the Zabarwan basin is characterized by saturated foothills and presence of springs in the lower reaches. Presence of preferential flow paths is likely on the forested peaks. A dual porosity hillslope runoff model that quantifies Hortonian overland flow, saturation overland flow and lateral subsurface flow as well as extent of foothill saturation was used to simulate the hydrology of this region. The urban subbasin, having historically been a wetland, has a shallow water table with high surface water-groundwater interactions and, accordingly, the region was modeled using hydrological model MIKE SHE. Keywords: Heterogeneous catchment hydrology, hydrological modeling, Dal Lake catchment 1. INTRODUCTION The Dal Lake is s shallow, fresh water lake situated in the summer capital Srinagar, of the state of Jammu and Kashmir. The lake catchment extends over an area of 336 square kilometres including the area of the wetland system which is about 24 square kilometers. The catchment is located between 34002‟ and 34013‟ N latitudes and 74048 and 75009‟ E longitudes. The lake is situated at an altitude of 1583 m with the highest point in the catchment at 4390 m height above the mean sea level. The lake forms the central body of a complex wetland system and is connected to a number of smaller water bodies through numerous water channels. This urban lake along with its associated wetland system forms a highly complex and vulnerable hydrological system. The lake is surrounded by gentle to steep sloping mountains on three sides and a nearly

MANIT Bhopal

Page 7

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

plain urbanized area of mild topography meshed with wetlands spread over an undulating topography on the west. Floating gardens along the west shore of the lake are among the unique features of this lake. The spatial diversity in the landscape of the catchment surrounding this lake adds to the complexity of this system. Landscape heterogeneity results in spatial variability of hydrological states and incomplete process understanding (Troch et al., 2008). Catchment morphology often acts as a dominant control on water flow paths and may be used as a clue to understand the catchment hydrological response (Beven et al., 1988). The present study identifies the catchment components with dissimilar hydrological response based on the physical features of the landscape. In recognition of these distinct but dominant hydrological features, the study has proposed similarly distinct approaches to hydrological modeling for these appropriately designated sub areas of the overall Dal catchment. 2. MATERIAL AND METHODS 2.1 Data For Hydrological Modeling Data most relevant to hydrological modeling includes meteorological data such as precipitation, wind speed and temperature, and catchment characteristics such as topography, soil types and land use. For the present study, meteorological data including daily accumulated precipitation, daily minimum and maximum temperatures and daily wind speed was obtained from the weather observatory of Sher i Kashmir University of Agricultural Sciences and Technology, Kashmir situated within the catchment. Information about the terrain was obtained from the ASTER Global Digital Elevation Model (ASTER GDEM) of resolution 30 m (Figure 1). Information regarding land use land cover (Figure 2a) and soil types (Figure 2b) were taken from the available literature.

Figure 1. DEM of the Dal Lake catchment

(a)

HYDRO 2014 International

(b)

Figure 2. (a)Land use land cover (2005) map of the Dal catchment (b) Map showing soil types in the Dal catchment (Badar et al., 2013) 2.2 Delineation of the Catchment The catchment of the Dal Lake was delineated using ASTER DEM of 30 m resolution using the Automatic Watershed Delineator of the hydrological model ArcSWAT. Based on the visually observed differences in the physical features of the landscape and thereby in the hydrological response, the catchment was broadly divided into three subbasins (Figure 3). The Dara-Dachhigam subbasin comprises of the mountains on the north of the lake and those extending far in the east behind the Zabarwan hills. The subbasin constitutes nearly 74 per cent of the total catchment area. The Zabarwan subbasin comprises of the steep slopes of the Zabarwan hills lying along the entire east coast of the lake. The urban subbasin on the west comprises of wetlands, floating gardens and urban settlements.

Figure 3. Subbasins of the Dal catchment 2.3 Hydrological characterization and selection of modeling approach The Dara-Dacchigam subbasin is characterised by mountains with slopes in the range of 6 per cent to 50 per cent drained by a very prominent network of drainage channels, the main channel being initiated by a glacial lake known as the Marsar Lake. The drainage pattern is dendritic in the north region of this subbasin whereas it is of trellis type in the east region as shown in Figure 3 (Badar et al., 2013). In this subbasin, runoff generation has been modeled in accordance with the Hortonian mechanism. This runoff concentrates towards the drainage channels wherefrom it is carried to the lake through a number of streams dominated by the Telbal creek (nallah). The outflow hydrograph for this feature constitutes mainly of the surface runoff and with an added component, though small, of return flow from lateral subsurface flow. Hydrological model SWAT (Soil Water Assessment Tool) was used to model the hydrology of this subbasin. The model incorporates an algorithm capable of generating stream network from the topographic information. The SCS curve number method was used to model runoff generation. SWAT uses kinematic storage model (Sloan et al., 1983) to compute return flow. The Zabarwan subbasin consists of gently sloping foothills near the lake shore that make up nearly 25 per cent of the subbasin followed by steeply sloping mountains having upto 68 percent slope with forested peaks as we go further from the lake shore.

MANIT Bhopal

Page 8

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

This subbasin is devoid of any prominent drainage network and therefore, runoff flows mostly in diffused form towards the lake. Another notable hydrological characteristic of this subbasin is that the foothills are wet and remain inundated at many places for a considerable part of the year. This can be attributed to the saturation of the soil upto ground level at the lower end of the hillslope caused as a result of vertical percolation and lateral subsurface flows from the higher reaches (Dunne, 1978). Saturation of soil profile upto ground surface is also evident from the presence of springs in this region. Further addition of subsurface flow to the saturated profile causes water to seep through the surface and flow as overland flow, and is known as the return flow (Pilgrim et al., 1978; Corbett, 1979; Mosley, 1979). Moreover, the saturated soil profile does not allow any further infiltration and therefore, these regions act as source areas for generating runoff by the mechanism known as saturation excess overland flow (Dunne and Black, 1970; Hewlett and Hibbert, 1963). High levels of saturation at the foothills also points to high amounts of lateral subsurface flow. It can be postulated that secondary porosity of the forest covered peaks plays a major role in conducting water as subsurface flow (Mosley, 1979; Beven and Germann, 1982). The hydrologic behaviour of such regions can be modeled by appropriately superimposing a macroporosity on the natural hydraulic conductivity of the soil (Shakya and Chander, 1995; Jain et al; 2013). A physically based lumped parameter hillslope runoff model that calculates unsaturated zone flow in dual porosity domain was used to model the hydrology of this subbasin. The model incorporates a modified form of the Horton's infiltration model which is the original Horton's infiltration equation corrected for lesser actual antecedent infiltration than infiltration at capacity rate and recovery of infiltration capacity. The model considers the macropore domain to be comprised of only two size pores. Flow in the smallest size pores is assumed to be laminar and calculated using Stokes law. For the largest size macropores, Mannings equation is used to quantify flow of water assuming turbulent flow (Equation 1). (1) where Qm is the total flow through the macrpores, r min and rmax are the radii of minimum and maximum size macropores, respectively, g is the acceleration due to gravity, A m is the total area of the macropores and ν isthe kinematic viscosity of water. The model also takes into account the transaction through the walls of the macropore into the soil matrix which is quantified using Philip's absorption equation. Preciptation in excess of the combined capacity of the soil matrix and the macropores flows as surface runoff. The return flow is quantified using the kinematic storage model of Sloan et al. (1983). To account for catchment storage effects (lagged and attenuated response), the model routes the surface runoff through a non linear reservoir of the form given in equation 2. (2) where S is the storage, Q is the outflow and k and n are nonlinear reservoir parameters.

HYDRO 2014 International

The subbasin was divided into two regions, steeply sloping mountains with upto 68 percent slope constituting nearly 75 percent of the total subbasin area and foothills with slopes upto 9 percent to be modeled separately. The model was setup to calculate water table fluctuation and thereby, the length of foothill saturation besides total outflow from the subbasin. The urban subbasin along the west coast of the lake comprises of floating gardens, small wetlands with undulating topography and urban setups. The subbasin has been historically a large wetland. The subbasin has a shallow water table with the depth to water table varying in the range of 1.1 to 1.5 m below the ground surface (Jeelani et al., 2013). High groundwater-surface water interactions exist in this region. Owing to the undulating topography, there are pockets of specific flow directions. Accordingly, the region was modeled using the hydrological model MIKE SHE. MIKE SHE is a 3 dimensional hydrological model having capabilities of modeling unsaturated and saturated zone flows together with the surface flows. 3. RESULTS AND DISCUSSIONS The outflow hydrograph for the Dara-Dachhigam subbasin is shown in Figure 4. The hydrograph indicates that the subbasin shows a direct response to precipitation. Peak flows occur mostly during the rainy months of March and April. Occurrence of zero flows during the months of December and January indicates that the flows are intermittent.

Figure 4. Outflow hydrograph of the Dara-Dachhigam subbasin Hydrological modeling of the Zabarwan subbasin revealed that the entire precipitation falling on the unsaturated length of the hillslope is either absorbed by the soil matrix or bypassed through the macropores, leaving zero amount of precipitation to flow as surface runoff.

Figure 5. Saturated slope length in the steep region of Zabarwan subbasin for different initial conditions of water table (Ls1: Initial length of saturation, H1: Initial height of water table above the impervious bed in the soil profile, equal to depth to the impervious bed if Ls1> 0)

MANIT Bhopal

Page 9

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

For a number of initial conditions of the water table, it was observed that nearly 100 m slope length (out of an average slope length of 2300 m) at the lower end of the steep zone always remains saturated (Figure 5), whereas the entire length of the foothills remains wet during all seasons (Figure 6). The same is also evident from the presence of a number of springs in the foothill region of this subbasin.

Figure 6. Saturated slope length in the foothills of the Zabarwan subbasin The entire overland flow component (appearing as peaks) in the outflow hydrograph of the steep region (Figure 7) is due to saturation excess overland flow occurring at the saturated lower end of the slope. The outflow hydrograph of the foothills which also represents the outflow of the entire subbasin (Figure 8) has a constant return flow component and peaks due to overland flow during precipitation events.

Figure 9. (a) Overland flow depth and (b) infiltration in the urban subbasin

Figure 7. Outflow hydrograph at the lower end of the steep region of Zabarwan subbasin

Figure 8. Total outflow from the Zabarwan subbasin The urban subbasin exhibits a highly complex hydrology. The hydrological response is a result of a number of factors like land cover, soil type and depth of water table below the ground surface. Results Figure 10. Overland flows in (a) x and (b) y directions in the urban subbasin

HYDRO 2014 International

MANIT Bhopal

Page 10

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

show that the overland flow depths are dominantly affected by the infiltration rate of the soil (Figures 9a and 9b). Existence of positive as well as negative values of overland flows in x and y directions (Figure 10a and 10b) shows that there are pockets of specific flow direction in this region. 4. CONCLUSIONS Different regions of the catchment of the Dal Lake exhibit hydrological behaviours which are markedly different from each other. This varied response is mainly on account of the diverse landscape across the catchment. Therefore, a single modeling approach is not appropriate to model the hydrology of the entire system. In the present study, an attempt was made to understand the hydrological response in various regions of the Dal Lake catchment and the physics underlying that response. Based on this understanding, appropriate modeling approaches were selected and used to model the hydrology of the system. Suitably chosen approaches could closely represent the observed hydrological phenomena in the three subbasins of the catchment. More such attempts are necessary to precisely understand and model the hydrology of heterogeneous catchments. REFERENCES i. Badar B, Romshoo SA, Khan MA (2013) Modelling catchment hydrological responses in a Himalayan Lake as a function of changing land use and land cover. Journal of Earth System Science 122(2): 433-449 ii. Beven K, Germann P (1982) Macropores and water flow in soils. Water Resources Research 18(5): 1311-1325 iii. Beven K, Wood EF, Sivapalan M (1988) On hydrological heterogeneity, catchment morphology and catchment response. Journal of Hydrology 100: 353-375. iv. Corbett ES (1979) Hydrologic evaluation of the storm flow generation processes on a forested watershed. Report: PB80-129133 National Technology Information Service, Springfield v. Dunne T, Black RD (1970) Partial area contributions to storm runoff in a small New England watershed. Water Resources Research 6(5): 1296-1311 vi. Dunne T (1978) Field study of hillslope flow processes. In: Hillslope Hydrology, John Wiley and Sons, Chichester, U. K. vii. Hewlett JD, Hibbert AR (1963) Moisture and energy conditions within a sloping soil mass during drainage. Journal of Geophysical Research 68: 1081-87 viii. Jain L, Haldar R, Khosa R (2014) Hillslope runoff processes and modelling. International Journal of Earth Sciences and Engineering 7(1): 193201 ix. Jeelani G, Shah RA, Hussain A (2013) Hydrogeochemical assessment of groundwater in Kashmir Valley, India. Published manuscript: http://www.ias.ac.in/jess/forthcoming/JESS-D-13-00128.pdf x. Mosley MP (1979) Streamflow generation in a forested watershed, New Zealand. Water Resources Research 15(4): 795-806 xi. Pilgrim DH, Huff DD, Steele TD (1978) A field evaluation of subsurface and surface runoff, II, Runoff processes. Journal of Hydrology 28: 319-341 xii. Shakya NM, Chander S (1998) Modeling of hillslope runoff processes. Environmental Geology 35: 115-123 xiii. Sloan PG, Moore ID, Coltharpa GB, Eigel JD (1983) Modeling surface and subsurface stormflow on steeply sloping forested watersheds. Report 142: 167 Water Resources Institute, University of Kenya, Lexington Kenya xiv. Troch PA, Carrillo GA., Heidbüchel I, Rajagopal S, Switanek M, Volkmann TH, Yaeger M (2008) Dealing with landscape heterogeneity in watershed hydrology: A review of recent progress toward new hydrological theory. Geography Compass 2: 10.1111/j.1749-8198.2008.00186.x

HYDRO 2014 International

Probability Analysis for Estimation of Annual One Day Maximum Ainfall of Devgarhbaria Station of Panam Catchment Area Kapil Shah 1 T.M.V. Suryanarayana 2 PG Student, Water Resources Engineering and Management Institute, Faculty of Technology and Engineering, The M.S. University of Baroda, Samiala-391410, Vadodara, Gujarat, India 2 Associate Professor, Water Resources Engineering and Management Institute, Faculty of Technology and Engineering, The M.S. University of Baroda, Samiala-391410, Vadodara, Gujarat, India Email: [email protected] 1

ABSTRACT: Daily rainfall data of 30 years (1961-1990) were analyzed to determine the annual one day maximum rainfall of devgarhbaria situated near panam dam, Gujarat, India. The study area receives mean annual rainfall 903.13 mm which is distributed in 45 rainy days. The observed values were estimated by Weibull's plotting position and expected values were estimated by four well known probability distribution functions viz., normal, log-normal, log-Pearson type-III and Gumbel. The expected values were compared with the observed values and goodness of fit was determined by chi-square test. The results showed that the log-Pearson type-III distribution was the best fit probability distribution to forecast annual one day maximum rainfall for different return periods. Based on the best fit probability distribution, the minimum rainfall of 42.69 mm in a day can be expected to occur with 99 per cent probability and one year return period and maximum Of 481.32 mm rainfall can be received with one per cent probability and 100 year return period. Keywords: recurrence interval, frequency, AODMR, probability distribution 1. INTRODUCTION A good understanding of the pattern and distribution of rainfall is important for water resource management of a country. Rainfall is one of the most important natural input resources to crop production and its occurrence and distribution is erratic, temporal and spatial variations in nature. Most of the hydrological events occurring as natural phenomena are observed only once. One of the important problem in hydrology deals with the interpreting past records of hydrological event in terms of future probabilities of occurrence. The design and construction of certain projects, such as dams and urban drainage systems, the management of water resources, and the prevention of flood damage require an adequate knowledge of extreme events of high return periods. In most cases, the return periods of interest exceed usually the periods of available records and could not be extracted directly from the recorded data. Therefore, in current engineering practice, the estimation of extreme rainfalls is accomplished based on statistical frequency analysis of maximum precipitation records where available sample data could be used to calculate the parameters of a selected frequency distribution. The fitted

MANIT Bhopal

Page 11

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

distribution is then used to estimate event magnitudes corresponding to return periods greater than or less than those of the recorded events, hence accurate estimation of extreme rainfall could help to alleviate the damage caused by storms and can help to achieve more efficient design of hydraulic structures. The specific objective shall include the following: 1) To analyse maximum one day rainfall in every year. And 2) To compute severity of rainfall by various return period. 2. MATERIAL AND METHODS Daily rainfall data of devgrhbaria raingauge station has been used for the present investigation. Time series rainfall records for the period of 30 years (1961 to 1990) have been collected from State Water Data Centre, Government of Gujarat, and Gandhinagar. Devgarhbaria is situated in the catchment area of panam dam in the panchmahal district of Gujarat state at 22 0 41' N latitude and 730 55‟ E longitude with survey of india(SOI) toposheeet (1.4 miles), no. 46/F,46/j and 46/E. The mean annual rainfall was 903.1287 mm. Area receives 85 per cent of annual of the rainfall during south-west monsoon i.e. from June to September. The study area is mostly hilly and covered with forests except near the Panam dam site where it is relatively flatter. It has the expansion of Soils of the derived from rocks like quartzites, schists and phyllites. Deep soils cover about 79% of the culturable and is watered dominantly by Mahi River. The area experiences three marked seasons – summer (Mar-May), Monsoon (June-Sep) & winter (Oct-Feb). Project area experiences tropical climate with minimum temperature of 12°C in January and maximum temperature of 39°C in May. Table 1: One day maximum daily rainfall for the period of 1961 to 1990

observed rainfall. The distribution of one day maximum rainfall received during different months in a year is presented in Fig. 1.

Fig. 1: AODMR in different months Annual one day maximum rainfall was sorted out from the data collected and using statistical techniques for data analysis. The statistical behavior of any hydrological series can be described on the basis of certain parameters. The statistical tests were carried out in accordance with the procedure. The computation of statistical parameters includes mean, standard deviation; coefficient of variation and coefficient of skewness were taken as measures of variability of hydrological series. All the parameters have been used to describe the variability of rainfall in the present study. 2.1 Return period Return period or recurrence interval is the average interval of time within which any extreme event of given magnitude will be equaled or exceeded at least once. Return period was calculated by Weibull's plotting position formula (Chow, 1964) by arranging one day maximum daily rainfall in descending order giving their respective rank as: T=

The daily rainfall data are sorted out and filtered to compute annual one day maximum. The maximum (189.1 mm) and minimum (54.8 mm) annual one day maximum rainfall(AODMR) was recorded during the year 20 th Sep 1962 and 26th August 1981, respectively. The mean value of AODMR was found to be 134.77 mm with coefficient of variation as 0.5281. The coefficient of skewness was observed to be 1.3464. August month received the highest amount of one day maximum rainfall (53%) followed by September (17%) and July (13%). it can be observed that the estimated annual AODMR for different probability distributions are following the same trend of

HYDRO 2014 International

(1)

Where, N - the total number of years of record and R- the rank of observed rainfall values arranged in descending order. Weibull's plotting position formula was used for computation of observed AODMR amounts at the return periods of 1.01, 1.05, 1.11, 1.25, 2, 4, 5, 10, 20 and 40 years. 2.2 Frequency analysis using frequency factors Values of Annual one day maximum rainfall can be estimated statistically through the use of the Chow (1951) general frequency formula. The formula expresses the frequency of occurrence of an event in terms of a frequency factor, KT, which depends upon the distribution of particular event investigated. Chow (1951) has shown that many frequencies analyses can be reduced to the form XT= (1+CVKT) (2) Where, is the mean, CV is the coefficient of variation, is the frequency magnitude of a factor and XT is the event having a return period T. KT is the frequency factor which depends upon the return period T and the assumed frequency distribution. The expected value of annual maximum daily rainfall for the same return periods were computed for determining the best

MANIT Bhopal

Page 12

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

probability distributions. Calculations of frequency factor of the four distributions namely normal, log-normal, log-Pearson typeIII and Gumbel are discussed as 2.2.1 Normal distribution The normal distribution, a two parameter distribution, has been identified as the most important distribution of continuous variables applied to symmetrically distributed data. The probability density function is given by: (3) Where, σ is the standard deviation and µ is the mean of the sample. 2.2.2 Log normal distribution A random variable x is said to follow a lognormal distribution if the logarithm (usually natural logarithm) of is normally distributed. The probability density functions of such a variable y=ln x:

x

0

  (4) Where, σy is the standard deviation and µy is the mean of y = ln x 2.2.3 Log-Pearson type-III In log-Pearson type-III distribution, the value of variate 'X' (rainfall) is transformed to logarithm (base 10). The expected value of rainfall 'XT' can be obtained by the following formulae XT = Antilog X Log X = M + KTS (5) where, 'M' is the mean of logarithmic values of observed rainfall and 'S' is the standard deviation of these values. Frequency factor KT is taken from Benson (1968) corresponding to coefficient of skewness (Cs) of transformed variate as (6) 2.2.4Gumbel distribution In Gumbel distribution, the expected rainfall 'XT ' is computed by the formula given by Chow in equation (2) KT - frequency factor which is calculated by the formula given by Gumbel (1958) as KT= -

(7)

2.3 Testing the goodness of fit The expected values of maximum rainfall were calculated by four well known probability distributions, viz., normal, lognormal, log-Pearson type-III and Gumbel distribution at different selected probabilities i.e. 99, 95, 90, 80, 50, 25, 20, 10, 5, 2.5, 2, 1 and 0.5 per cent levels. Among these four distributions, the best fit distributions decided by chi-square test for goodness of fit to observed values. The chi-square test statistic is given by the equation χ2 =

square value (Agrawal et al. 1995). If > 2 ñ for (N - k 1) degrees of freedom. Then the difference between observed and expected values is considered to be significant. 2.4 Regression model Regression models were developed for estimating the AODMR to return periods in the present study and found the coefficient of determination (R2). 3. RESULTS AND ANALYSIS The average, standard deviation, coefficient of variation and skewness of Annual One Day Maximum Rainfall for 30 years and their respective formulas are given in Table 2. These statistical parameters can be used to find the estimated one day maximum rainfall from different probability distribution functions. The variation of standard deviation over the mean is shown in Fig. 2. It was also observed that 10 years (33.3%) received one day maximum daily rainfall above the average. Table 2:Computation of statistical parameters of annual one day maximum rainfall Statistical Parameter Mean Median

Mode

x = Σx / n x=L0+

Computed Value 134.77 106.95

152.4

Standard deviation

71.179

Coefficient of variance

0.5281

Coefficient of skewness

1.3464

(8)

Where, Oi is the observed rainfall and Ei is the expected rainfall and will have chi-square distribution with (N - k -1) degree of freedom (d.f.). The best probability distribution function was determined by comparing Chi square values obtained from each distribution and selecting the function that gives smallest chi-

HYDRO 2014 International

Formula

MANIT Bhopal

Fig 2:- Standard Deviation Variation over the mean

Page 13

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The AODMR for the period of 30 years was plotted against return period in years which was calculated from Weibull's method and presented in Fig. 3. Observed rainfall were found for various return periods of 1.01, 1.05, 1.11, 1.25, 2, 4, 5, 10, 20 and 40 year and for different probability distributions such as normal, log-normal, log-Pearson type-III and Gumbel were calculated and presented in Table 3. It is generally recommended that 2 to 100 years is sufficient return period for soil and water conservation measures, construction of dams, irrigation and drainage works (Bhakar et al., 2006). It was observed that all the three probability distribution functions fitted significantly i.e. null hypothesis accepted except normal distribution. Log-Pearson type-III distribution was found to be the best fitted to AODMR data by Chi-square test for goodness of fit. A maximum of 116.84 mm rainfall is expected to occur at every 2 years and 50 per cent probability which is nearer to the mean AODMR. For a return period of 5,10,20,50 and 100 years the AODMR, annual one day maximum rainfall is 178.98 mm, 226.77 mm, 277.73 mm, 351.68 mm and 413.58 mm which including other return periods are shown in Table 3.

value determined the best probability distribution function. The chi-square values (Table 4) for normal, log-normal, log-Pearson type-III and Gumbel distributions were 2.38,-0.04, 0.20 and 2.25, respectively. Log-Pearson type-III distribution gave the lowest calculated chi-square value that is selected among the four probability distributions. Hence, log- Pearson type-III has been found the best probability distribution for predicting AODMR for Devagarhbaria station of panam catchment area.

Table 4:

Chi-square values at different probability levels for different distributions

Table 3: Observed and expected one day maximum rainfall at different probability levels

The expected AODMR for different probabilities are graphically represented in Fig. 4.Regression models were developed from the observed AODMR against different return period by using Weibull's method. The trend analysis (Fig. 4.) for prediction of one day maximum rainfall for different return period was carried out and it is found that the exponential trend line gives better coefficient of determination (R2) = 0.9342 and the equation is: Y = 79.95X0.428 where Y is AODMR in mm and X is Return period in Years.

Fig. 3:Predicted AODMR with different probability distribution vs return period From the figure, it can be observed that the estimated annual AODMR for different probability distributions are following the same trend of observed rainfall. All four probability distribution functions were compared by chi-square test of goodness of fit and then selecting the function that gave the smallest chi-square

HYDRO 2014 International

Fig.4:- Annual One Day Maximum Rainfall with various return period 4. CONCLUSIONS

MANIT Bhopal

Page 14

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The mean value of AODMR was found to be 134.77 mm with standard deviation and coefficient of variation of 71.179 and 0.5281, respectively. The coefficient of skewness was observed to be 1.3464. The frequency analysis of AODMR for identifying the best fit probability distribution was studied for four probability distributions such as normal, log-normal, logPearson type-III and Gumbel by using Chi-square goodness of fit test. It was observed that all the three probability distribution functions fitted significantly i.e. null hypothesis accepted except normal distribution. Log-Pearson type-III distribution was found to be the best fitted to AODMR data by Chi-square test for goodness of fit. Based on the best fit probability distribution, the minimum rainfall of 42.69 mm in a day can be expected to occur with 99 per cent probability & one year return period and maximum of 413.58 mm rainfall can be received with one per cent probability & 100 year return period. This study gives an idea about the prediction of Annual One Day Maximum Rainfall to design the small and medium hydraulic and soil and water conservation structures, irrigation, drainage works, vegetative waterways and field diversions. This study also helps in developing cropping plan and estimating design flow rate for maximizing crop production. 5. REFERENCES: i. Adegboye, O.S and Ipinyomi, R.A (1995) ―Statistical tables for class work and Examination.‖ Tertiary publications Nigeria Limited, Ilorin, Nigeria, pp. 5 – 11 1765 – 1776. ii. Agarwal, M. C., Katiyar, K.S. and Ram Babu (1988). ―Probability analysis of annual maximum daily rainfall of U. P., Himalaya.‖ Indian Journal of Soil Conservation, 16(1): 35-42. iii. Barkotulla, M. A. B., Rahman, M. S. and Rahman, M. M. (2009). ―Characterization and frequency analysis of consecutive days maximum rainfall.‖ at Boalia, Rajshahi and Bangladesh. Journal of Development and Agricultural Economics, 1: 121-126. iv. Benson, M. A. (1968). ―Uniform flood frequency estimating methods for federal agencies.‖ Water Resources Research, 4(5): 891-908. v. Bhakar, S. R., Bansal, A. N., Chhajed, N. and Purohit, R. C. (2006). ―Frequency analysis of consecutive days maximum rainfall at Banswara, Rajasthan, India.‖ ARPN Journal of Engineering and Applied Sciences, 1(3) : 64-67. vi. Bhim Singh, Deepak Rajpurohit, Amol Vasishth and Jitendra Singh (2012). “Probability analysis for estimation of annual one day maximum rainfall of jhalarapatan area of rajasthan,india.‖ Plant Archives Vol. 12 No. 2, 2012 pp. 10931100 vii. Chowdhury, J.U. and Stedinger, J.R. (1991) ―Goodness of fit tests for regional generalized extreme value flood distributions.‖ Water Resource. Res., 27(7) : viii. Chow, V. T. (1951). ―A general formula for hydrologic frequency analysis.‖ Transactions American Geophysical Union, 32: 231237. ix. Chow, V. T. (1964). ―Hand book of applied hydrology.‖ McGrawHill Book Company, New York. x. ―Introduction To Probability and Statistics In Hydrology‖ a Book By Dr. Miguel A. Medina xi. Murray, R.S and Larry, J.S (2000) ―Theory and problems of statistics‖ Tata Mc Graw – Hill Publishing Company Limited, New Delhi, pp. 314 – 316, Third edition. xii. Olofintoye, O.O, Sule, B.F and Salami, A.W (2009). ―Best–fit Probability distribution model for peak daily rainfall of selected Cities in Nigeria.‖ New York Science Journal, 2009, 2(3), ISSN 1554-0200 xiii. Salami, A.W (2004). Prediction of the annual flow regime along Asa River using probability distribution models. AMSE periodical, Lyon, France. Modeling C2004, 65 (2), 41-56. (http://www.amsemodeling.org/content_amse2004.htm) New York Science Journal, 2009, 2(3), ISSN 1554-0200 http://www.sciencepub.net/newyork, [email protected] xiv. Singh, R. K. (2001). ―Probability analysis for prediction of annual maximum rainfall of Eastern Himalaya (Sikkim mid hills).‖ Indian Journal of Soil Conservation, 29: 263-265.

HYDRO 2014 International

Experimental and three Dimensional Numerical Studies for A Sluice Spillway Kulhare, A.1 Bhajantri, M.R.2 Research Officer, Central Water & Power Research Station, Pune - 411024, INDIA 2 Dr., Chief Research Officer, Central Water & Power Research Station, Pune - 411024, INDIA Email: [email protected] 1

ABSTRACT:Hydraulic modelling of spillways can be done through physical modelling or computer based numerical modelling. Experimental investigation through physical model studies is widely adopted common practice to optimize the design of spillway components. The advent of high-speed and large-memory computers has enabled to obtain numerical solutions to many complicated hydraulic problems of spillways. Numerical simulation has become a viable complementary tool for physical modelling of spillways. In the present work, hydraulic model tests were carried out on a 1:45 scale 2-D sectional model. In numerical studies, a Computational Fluid Dynamics (CFD) software 'FLUENT' was used which runs on a Finite Volume method for simulation. The results of the numerical model in respect of discharging capacity, pressures at different locations over spillway profile and sluice roof profile were compared with the physical model results. The numerical results obtained by simulating the system as two phase problem showed close agreement with the results obtained from physical model studies. Keywords: Computational Fluid Dynamics; FLUENT; VOF; Sluice spillway; Ski-jump bucket; Discharge capacity 1. INTRODUCTION Innovative designs of spillways have been evolved based on the concept of flushing. The design of spillway is required to perform the dual function of flushing of the reservoir as well as passing of the flood discharge. Low level Breastwall/Sluice spillways (also called Orifice spillway) combine the advantage of greater depth of flow over the crest and moderately sized gates. Orifice spillways have been widely recognized as the most appropriate, especially for run-of-the-river projects for handling both flood releases and flushing of sediment. Orifice spillway is an effective hydraulic structure for keeping the reservoir clean from the sediments along with the advantage of reduced gate height and reduced overflow crest length. Though the provision of breast wall or sluice has many advantages, there is no specific design method for its configuration. Spillway designs have been investigated through physical as well as numerical modelling. The drawback of physical model studies of spillways are the cost of construction, delay in time for fabrication and construction of model parts and conducting experiments and difficulty in changing structural details of various components of spillway while doing parametric studies. Numerical simulation has become a feasible complementary tool to physical modelling of spillways. The data obtained from the physical model studies can be used for model calibration and validation of the numerical models. To simulate the actual flow by providing an alternative cost-effective means of fluid

MANIT Bhopal

Page 15

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

dynamics, CFD complements experimental and theoretical analysis. However, the utility of a numerical model depends on the validity of the governing equations and numerical methods. CFD design tool as a more reliable and in order to become acceptable, numerical studies should be carefully validated with experimental results. Hydraulic design of spillways with CFD is a new application, it requires especially careful verification. Many researchers have conducted numerical modelling experiments on different types of spillways. But most of the investigations have been done for the Ogee crested overflow spillway. Savage and Johnson, Bijan Dargahi, Unami et al. and Ho et al. have done some recent work in field of overflow spillway and they found reasonable agreement with experimental data. Hu Cheng Yi et al. have studied the configuration of the spillway with breastwall and based on physical and numerical modelling they suggested some design configuration, which has a greater discharging capacity, less negative pressures on profiles and having a simpler configuration of profiles. The main concern of the present work is to investigate the flow phenomena over the sluice spillway and to compare the results with 3D numerical flow simulation. A commercial CFD code known as FLUENT was used for the present study. With the help of a numerical model, an attempt is made to investigate hydraulic characteristics by simulating the discharge, pressure distribution and water surface profile over the spillway. 2. MODEL INVESTIGATION Experiments were conducted on 1:45 scale 2-D sectional model to optimize roof profile as well as spillway bottom profile of the sluice spillway. In the model one full span and two half spans on either side were fabricated in transparent Perspex sheet of 12 mm thick. The fabricated spillway was installed in a one metre wide flume. The discharge was measured by means of a calibrated Rehbock weir. The accepted equations for similitude, based on Froudian criteria were used to express the mathematical relationship between the dimensions and hydraulic parameters of the model and the prototype. Discharging capacity, pressures distribution over the roof profile of sluice and spillway profile were measured for the head of 26 m over the crest of the spillway. The measurements were taken along the centre line section of the spillway span as well as along the side of the pier. The pressures were measured at 21 different locations over the spillway profile and at 12 locations over the roof profile of sluice. Pressures on the spillway were measured using a piezometer board with plastic tubes vented to the atmosphere. The piezometer board was leveled with respect to the spillway elevations. The piezometer board readings provided the average pressure readings at each pressure tab location. Measurements on the piezometer board were readable to within 0.045 m. Detailed measurements of water surface profiles normal to the flow were made in the centre line of the spillway span. A pointer gauge was used to measure the free water surface profile over the spillway structure. Figure 1 and 2 show the section the spillway and model view of the spillway respectively.

HYDRO 2014 International

Figure 1. Section of spillway

Figure 2. Model view 3. NUMERICAL MODEL SET-UP Flow over the sluice spillway was simulated with CFD software FLUENT. FLUENT is a commercial computer program for modelling fluid flow and heat transfer in complex geometries. FLUENT provides complete mesh flexibility, including the ability to solve flow problems using unstructured meshes that can be generated about complex geometries with relative ease. It solves the full three dimensional equations of fluid motion in general orthogonal curvilinear coordinates for both laminar and turbulent flows. 3.1 Computational DomainThe geometry of the spillway is prepared with prototype dimensions by using ”GAMBIT” software. For building the domain for upstream of the spillway dam axis, reservoir length of 100 m chainage was taken for inlet of flow and in the downstream side, domain was extended upto 240 m chainage with pressure outlet. The domain height is chosen around 32.5 m above the crest at spillway surface so that the water level can be attained in tank as well as interface with air can be captured properly. The domain sufficiently extended in the downstream region around distance of 180 m from the end of spillway structure. The objective of the extension of the domain in downstream is to capture the water behaviour after leaving the spillway and where the water hits the bottom surface of the domain. Figure 3 shows the Section of the domain.

MANIT Bhopal

Page 16

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 3. Section of the domain 3.2 Grid Generation Three dimensional grid was developed in Gambit software itself. The 3-D mesh generation consists of the geometry generation and 3-D grid development over the spillway geometry, water tank upstream for reservoir and downstream region. The objective behind this grid generation is to provide the mesh to simulate flow through two spans, mixing of flows coming out of two spans and the flow over spillways. The full domain is decomposed into the smaller volumes, so that they can be meshed by structured mesh. The cells have been clustered near the sluice roof profile and spillway surface to capture wall bounded effects and predict the wall pressures in the flow simulation. The grid is made finer in those regions where the water and air have interface. Minimum height of the grid cell is 0.1 m and maximum height of the grid cell is 0.9 m in the domain. The hexahedral cells are used for grid generation with the cell count 885261. The surface grid is shown in Figure 4.

Figure 4. Meshing of the domain 3.3 Boundary Conditions

Air was defined as a primary phase and water as a secondary phase. For the calculation of air water interface i.e. free water surface, volume of fluid (VOF) model was used. For simulation of spillway flow, two inlets were needed to define the water inflow to domain and air inflow over the top of domain. Water inflow was defined as a pressure inlet with the initial water level and initial velocity at the inlet face. Also the air inflow over the domain upstream and downstream of the spillway was defined as pressure inlet boundary condition. The water outflow at the end of domain was defined as a pressure outflow boundary condition. All the solid boundaries including side walls, Sluice walls, piers and spillway bottom were defined as wall boundaries with no slip condition. Figure 5 shows boundary conditions of the numerical domain.

Figure 5. Boundary conditions of the domain 4. SOLUTION PROCEDURE The numerical model of sluice spillway was run with unsteady free surface calculations with pressure based solver, which enables the pressure-based Navier-Stokes solution algorithm. The VOF method was used to capture the interface between water and air and governing equations are solved by the Finite volume method. For the VOF - method the Body force weighted scheme is used for pressure interpolation as the gravity force is high and the modified HRIC scheme is used for the volume fraction equations in order to improve the sharpness of the interface between the two phases i.e. water and air. Second order upwind scheme is used for momentum and pressure equations. The k- ε turbulence model was used to simulate the threedimensional turbulent flow. Figure 6 shows simulated flow after run of 27.36 seconds.

They are the critical components of simulations and it is important that they are specified appropriately. When solving the Navier-Stokes equation and continuity equation, appropriate initial conditions and boundary conditions need to be applied. Setting the appropriate boundary conditions can have a major impact on whether the numerical model results are reflecting the actual simulation one is trying to simulate. Poorly defined boundary conditions can have a significant impact on the solution. A set of boundary conditions such as pressure inlet, mass flow inlet, velocity inlet, pressure outlet, outflow, wall boundaries etc. are available in FLUENT. It is significant that the boundary conditions accurately represent what is actual physics occurring, to simulate a given flow close to real.

HYDRO 2014 International

MANIT Bhopal

Figure 6. Simulated flow over the spillway

Page 17

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

5. SIMULATION RESULTS AND ANALYSIS There is an unlimited level of details of the results in the numerical model analysis. Observations and analysis can be made very minutely for each and every component of model in respect of fluid properties such as velocity, pressure, and water surface profiles etc., also the forces on the various locations. In this numerical study the main concern to obtain discharging capacity of sluice spillway, pressure distribution over the sluice profile and spillway profile and free water surface profile for for 26 m head over the crest. It was found from the physical model studies that the design maximum discharge of 2983 m3/s could be passed through two sluices fully open with the reservoir water level (RWL) El. 26 m above the crest. Also in the numerical model, the discharging capacity of the spillway was found adequate to the passing discharge of 3030 m3/s at RWL El. 26 m, which is 1.6% higher than what we found from physical model studies. Also the coefficient of discharge is coming around 0.63 which is closer to 0.62 that was obtained by experimental studies. The result shows the good agreement between physical and numerical values in respect of discharge values. Water surface profile have been measured over the spillway surface on the physical model and compared with the numerical solution. Figure 7 shows the plot of both the water surface profile elevations. It has been observed in the numerical model study that after the lip of the ski-jump bucket, the water surface elevations obtained lower than what obtained in the physical model. Numerical model was solved with prototype dimensions and in the prototype, the jet after ski-jump bucket has been thrown out fully into the air downstream of the spillway structure. There may be more interaction of air and water because there will be free surface from either side of jet. This may be the reason of the deviation in elevation values after the spillway structure.

Figure 7. Water surface profile Pressure distribution were computed over the sluice bottom profile for numerical studies and compared with the physical model results. In both the studies water surface follow the sluice profile and corresponding pressures having same trend over the profile. Figure 8 shows the plots for pressure values of experimental and numerical modelling results over the sluice surface for comparison. It shows the good agreement between both the values. Pressure distribution over the sluice profile were

HYDRO 2014 International

found satisfactory and having a good agreement between both the studies in most of the locations. Figure 9 shows pressure contour near the sluice profile obtained from numerical model. It shows the low pressure zone in downstream portion of the sluice surface as observed in physical model.

the the the the the

Figure 8. Pressures over the sluice profile

Figure 9. Pressures contour near the sluice profile Figure 10 shows the plots for pressure values of experimental and numerical modelling results over the spillway surface. The plot shows good agreement between both the values except some locations near the entrance. Due to absence of upstream curve the separation of flow is observed at the crest of the spillway near entrance of the sluice. Flow at the entrance of spillway in the numerical model and the physical model are shown in Figure 11 in the form of velocity vectors. It shows the separation of velocity vectors near the entrance, so that the pressures on the spillway surface in this zone are reduced compared to other locations. Whereas in the physical model the velocity components cannot be minutely observed and also due to the wide river valley in the model the vertical component of velocity vectors was dominated by the horizontal components of velocity vectors. Because of this reason the pressure distribution is not following the same trend in this region in both the cases of centre line as well as side of pier.

MANIT Bhopal

Page 18

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 ii. Cheng, Xiangju, Yongcan, Chen and Lin, Luo. (2006), Numerical Simulation of Air-Water Two-Phase Flow over Stepped Spillways. Science in China Series E: Technological Sciences, Volume 49, Number 6, 674-684. iii. Dargahi B. (2006), Experimental Study and 3-D Numerical Simulations for a Free-Overflow Spillway. Jour-nal of Hydraulic Engineering, ASCE 132-9,899-907. iv. Fluent Manual ver. 6.3 v. Gadge,P.P., Kulhare,A. and BhosekarV.V., Application of Computational Fluid Dynamics in Hydraulic Structures, National Conference on Hydraulics and Water Resources, HYDRO -2011, Dec 29-30, at SVNIT, Surat, Gujarat. vi. Hu, C. Y., Wei, Y., and Zheng, Z. P. (1990), Study on Configuration of Overflow Dams with Breast Wall. 7th Congress APD-IAHR. vii. Savage, B. M., and Johnson, M. C. (2001), Flow over Ogee Spillway: Physical and Numerical Model Case Study. International Journal of Hydraulic Engineering, ASCE 127-8, 640-649. viii. Unami, K., Kawachi, T, Munir, Baber M., Itagaki, H, (1999), Two Dimensional Numerical Model of Spillway Flow, Journal of Hydrol. Engg. ASCE 125, 369-375. ix. Versteeg, H. K., and Malalasekera, W. (1995), An Introduction to Computational Fluid Dynamics-The Finite Volume Method. Longmaon Scientic &Technical, England.

Figure 10. Pressures over the spillway profile

Figure 11. Velocity vectors at the entrance of spillway 4. CONCLUSION In this paper, a finite volume-based CFD software FLUENT was used to investigate the hydraulic characteristics of flow through sluice spillway. The water surface profile, pressure distribution and discharge characteristics of the chosen spillway were computed and compared with existing physical model data. The computed and experimental values of the coefficient of discharge were 0·63 and 0·62, the computed value being 1.6 % higher than the experimental value observed on the physical model. As seen from the figures depicting pressure values and water surface elevations, it shows the good matching trend and values in case of breastwall bottom profile for both numerical as well as experimental studies. The upstream slope was not guiding the flow over the crest properly, as a result of which a mild separation zone was seen forming over the horizontal crest in the numerical model as depicted in the figure of velocity vectors, which was not predicted by physical model. Except the entrance the pressure distribution was found good agreement between the physical and numerical results. Reasonable agreement is observed with the numerical and physical model results, showing the applicability of the CFD software for the numerical simulation of real case study of spillway. Further refinement in mesh generation and cell count may improve the results of the simulation of flow through a sluice spillway. REFERENCES i. Bhajantri, M.R., Eldho T.I, Deolalikar P.B. (2006), Hydrodynamic Modelling of Flow over a Spillway using a Two-Dimensional Finite VolumeBased Numerical Model. Sadhana, Vol.31, part 6, 743-754.

HYDRO 2014 International

Physical Model Study for Energy Dissipation Arrangements to the Pick up Weir Across Pachaiyar River in Tamilnadu C. Prabakar1 P. K. Suresh2 T. Ravindrababu3 A. Parthiban4 A. Muralitharan5 1 Assistant Engineer, Institute of Hydraulics & Hydrology Poondi 602 023, India 2 Research Head, Centre of Excellence for Change, P W D Campus, Chepauk, Chennai 600 005, India 3 Assistant Director, Institute of Hydraulics & Hydrology Poondi 602 023, India 4 Assistant Director, Institute of Hydraulics & Hydrology Poondi 602 023, India 5 Assistant Engineer, Institute of Hydraulics & Hydrology Poondi 602 023, India Email: [email protected] ABSTRACT: The Agricultural development in Tamil Nadu mainly depends upon the surface irrigation as well as lift irrigation. But the state has almost utilized its surface water potential and ground water potentials. Hence, the further expansion of irrigation and agriculture in Tamilnadu depends on inter-linking of rivers by utilizing the surplus flood water which flows into the sea as unused. This scheme is proposed for interlinking of rivers Tamirabarani, Karumeniyar, and Nambiyar by connecting surplus water from Tamirabarani through kanadian channel and a new flood carrier canal for a length of 73km. The diversion of surplus water of Tamirabarani basin to its sub basin of Pachaiyar and adjoining basin of Nambiyar and Karumeniyar will be a milestone for linking the south flowing rivers. Under the Formation of Flood carrier canal with a carrying capacity of 3200 cusecs crosses the river Pachaiyar. At the place of canal crossing the river Pachaiyar, to utilise the river water of pachaiyar to divert in the flood carrier canal the Pickup weir is

MANIT Bhopal

Page 19

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

proposed to be constructed across Pachaiyar for a length of 250m to pass the Maximum flood discharge of 31664 cusecs safely. The physical model study for the energy dissipation arrangements for the stilling basin of the proposed weir and the scour vents is studied in this Institute and a 6 numbers of trials were conducted to evolve efficient Energy dissipation arrangements for the proposed pick up weir and scour vents. Various energy dissipation structures were introduced in the above 6 trials and the optimal performance is ascertained in the model studies and suggested for the stilling basin is given in this report in detail. Keywords: Physical model, Energy dissipation, Friction blocks 1. INTRODUCTION: The Agricultural development in Tamil Nadu mainly depends upon the surface irrigation as well as lift irrigation. But the state has almost utilized its surface water potential. Hence, the further expansion of irrigation and agriculture in Tamilnadu depends on inter-linking of rivers and their tributaries by utilizing the surplus flood water which flows into the sea as unused. This scheme is proposed for interlinking of rivers Tamirabarani, Karumeniyar, and Nambiyar by diverting water from Tamirabarani through the existing Kannadian Channel by increasing the carrying capacity and excavating a new flood carrier from LS 6.50km of existing Kannadian Channel through drought prone areas of Sathankulam, Thisayanvilai in Tirunelveli and Thoothukudi Districts respectively for a length of 73km after fulfilling the needs existing Kannadian Channel. The diversion of surplus flood water from Tamirabarani basin will be effectively utilized in the farther most gross command area but also in the adjoining basins of Pachaiyar, Nambiar and Karumeniyar rivers. The flood carrier canal will be operated only in the time of flash flood when the surplus flow of Tamirabarani water through the last anicut namely Srivaikundam anicut goes into the sea after meeting out the full demand of Tamirabarani basin. The diversion of surplus water of Tamirabarani basin to its sub basin of Pachaiyar and adjoining basin of Nambiyar and Karumeniyar will be a milestone for inter-linking the south flowing rivers. The Director, Institute of Water studies, Chennai formulated a proposal from tail end of Kanadian channel (nearby Melaseval) by using remote sensing and GIS taking into consideration of the existing Kanadian channel alignment. It is proposed to excavate a flood carrier canal from Kannadian Channel at LS 6.50 km to ML theru for a length of 73 km. The carrying capacity of the flood carrier at LS 0m is 3200 cusecs (90.61 Cumecs). On its length of run the new flood carrier crosses the river Pachaiyar and Karumeniyar. At the place of crossing the river Pachaiyar, a Pickup weir was proposed to be constructed at LS 20599 to LS 20690m of flood carrier canal. The design and drawing was prepared by the Superintending Engineer, Designs Circle. Chennai-05, it was suggested in the design that the energy dissipation arrangements proposed for the weir and scour vents are only tentative and should be finalized based on the rock level available at the downstream side during execution and conducting model studies at Institute of Hydraulics and Hydrology, Poondi. 1.1 LOCATION

HYDRO 2014 International

The River Thamirabarani originates from eastern slope of Western Ghats and traverse to a length of 120 kms and it is More Or Less Perennial River. There are 12 numbers of tributaries confluences with this river on its length of traverse. The following reservoirs were constructed across Thamirabarani River and its tributaries. 1. Papanasam Reservoir 2. Manimuthar Reservoir 3. Servalaru Reservoir 4. Gadana Reservir 5. Ramanadhi Reservoir 6. Gundar Reservoir 7. Karuppanathi Reservoir 8. Adavinainarkoil Reservoir 9. Vadakku Pachaiyar Reservoir The following anicuts were constructed across Thamirabarani River and its tributaries. 1. Kodaimelazhagian Anicut 2. Nathiyunni Anicut 3. Kanndain Anicut 4. Ariyanayagipuram Anicut 5. Suthamalli Anicut 6. Pazavoor Anicut 7. Maruthur anicut 8. Srivaikundam Anicut. The pickup weir and scour vents were proposed to construct across the River Pachaiyar located at LS 20599 to LS 20690m of the proposed flood carrier canal. The Chief Engineer, PWD, WRO, Design Research & Construction Support has given the approved drawing for the proposed pick up weir and scour vents. 2.0 OBJECTIVE OF STUDY 1. To evolve efficient Energy dissipation arrangements for the proposed pick up weir and scour vents to pass the Maximum flood discharge of 31664 cusecs safely. 2. To evolve good flow performance on stilling basin and surplus course. 3.0 HYDRAULIC PARTICULARS 1 Maximum Flood Discharge 31664 Cusecs 2 Front maximum flood level + 60.175m 3 Rear maximum flood level +53.45m 4 Crest level 58.745m 5 Length of the structure 240m Pick up weir 1 Discharge through Weir 29773 cusecs 2 Length of Weir 226.80m 3 Crest level 58.745m 4 Downstream bed level +52.00m 5 Stilling basin length 18.80m 6 Stilling basin level 50.60m 7 Rock level/Foundation level +48.80m

Scour vents 1 Discharge through Scour vents

MANIT Bhopal

1901 cusecs

Page 20

International Journal of Engineering Research Issue Special3 2 3 4 5 6 7 8 9

Sill level Number of vents Size of vents Basin level at Left side scour vents Basin level at Right side scour vents Top of operating platform Foundation level @ Right side Foundation level @ Left side

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

+53.00m 4 nos 2.1m x 0.9m +52.00 49.50m +61.175m +49.00m +50.50m

4.0 MODEL SET UP A comprehensive rigid bed, geometrically similar physical model with a scale of 1:50 is selected and the model discharge is calculated. Model discharge of the river was allowed through 'V' notch. Necessary gauge well have been constructed for measuring the water levels for the required maximum flood discharge. 4.1 Rigid Bed Model The model was constructed with the hydraulic components as per the design drawings and the downstream bed levels given by the Field officials. Right scour vents Spillway Stilling basin Left scour vents

Figure- 1. Dry Model of Pachaiyar Spillway 5. MODEL RUN Maximum flood discharge of 31664 Cusecs for Pachaiyar River is taken and the model discharge was computed and allowed through "V" notch to arrive the energy dissipation arrangements in the stilling basin. Trial I After incorporating the SE/Designs proposal and downstream bed levels furnished by the field officials with embankments on both sides of the river width, the model was run with the maximum flood discharge. Observation The hydraulic jump was found satisfactory, the flow concentrates on the central portion of the river since the banks of the river has high bed level ranging from +54.00 to +55.00m. The velocity ranges to 4.0 to 4.5m/sec. Cross flow was observed in the Left and Right scour vents since the bed level in the downstream of scour vents is in higher level, when comparing to the stilling basin. The flow of the left side scour vent tends to move towards the stilling basin of the weir. This trial needs alterations.

HYDRO 2014 International

Trial II In this trial the downstream of the river bed below the stilling basin is regarded to a bed slope of 1 in 400 up to a length of 500m along the river, keeping the SE/Designs proposal of hydraulic components. Observation The hydraulic jump was found satisfactory; the flow spreads uniform to the entire width of the river. The velocity in the range was of 3.5 to4.0 m/sec on the downstream. The left side scour vent baffle wall top has a level of +54.00m and after regarding the downstream bed level of the river course from the stilling basin of the river course is +52.00m to a slope of 1 in 400. The water from the left side scour vents experience a fall of 2m and water plunges in the downstream with an impact. With this condition the river bed will experience a heavy scour, which can damage the hydraulic structures. This trial needs additional alterations and requires suggestions from the Design wing. Trial III The site officials informed that the downstream portion consists of hard rock so that water can be allowed in the downstream bed of the left side scour vents. To assess the site condition the site was inspected and it is found that left side scour vents portion has hard rock up to a level of +53.00m; hence it was suggested that the Stilling basin of Left side scour vent can be kept as +52.80m and baffle wall top as +54.00m. The fall of 2m in left scour vent portion to the river portion can be provided with necessary water cushion arrangements by extending the wing wall and divide wall of the scour vent which can be fixed after conducting the model trial. The design shall be got revised from the designs wing. The model trial was done with the suggestions made by the officials as above with a ramp to negotiate the fall and the Observed velocity on the left scour vents is 4.5 m/sec. Hence this trial needs further alterations. Trial IV The Designs wing has revised the drawing and based on the details, the trial was done and the velocity is in the range of 1.5 to 2.0 m/sec in the weir portion and 7.3m/sec in the left side scour vent portion and 3.7m/sec in the right side scour vent portion. In order to reduce the velocity further, trial was conducted by introducing friction blocks of size 4mx1mx1m in the entire river width on the downstream side of the baffle wall with two rows in zig zag arrangement, the velocity is in the range of 1.0 to 2.0m/sec at the weir portion and 6.0 m/sec at the left side scour sluice. This trial needs alteration. Trial V In this trial the following alteration were done as follows 1. Downstream of the river is provided with a reverse slope arrangements keeping the level as +52.00m at LS 30M and +53.00m at LS 60m and continuing the level of +53.00m up to a distance of LS 170m on the river course. 2. Reducing the stilling basin level of Left side scour vent portion to +51.50m and baffle wall top as +.52.00m. The trial was conducted and the velocity ranges from 1.98 to 2.90m/sec, hydraulic jump in the stilling basin is satisfactory but during initial period the cross flow of water is observed from the

MANIT Bhopal

Page 21

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

left and right side scour vents and concentrate on the weir portion. Trial VI The Designs wing visited the site of the Pachaiyar and had discussion with the field engineer about the re grading of the river. They have finalized that the river can be re graded to the entire width of the river as only weathered rock and soft disintegrated rock are available in the river course portion. The Designs wing has request to conduct the model study for the regraded section of the river for the entire width as suggested in the approved design and to maintain the computed Rear Water Level(RWL) +53.45M @50m downstream of the weir alignment. The running model trial was inspected by Designs wing and the model was run with the maximum flood discharge. Observation The hydraulic jump formed in the stilling basin is found satisfactory and velocity on the downstream of the weir portion is in the range of 1.0 to 2.0 m/sec is within the permissible range. But the downstream of left side scour vent portion measured a velocity of 4.0 m/sec. To reduce the velocity further in the left side scour vent, friction blocks of size 1.0x1.0x1.0m is introduced 3 rows with 3 numbers in the first and second rows and 2 nos in the second row. Thus making zig zag arrangements in the left side scour vent portion. By introducing the baffle blocks the velocity got reduced to 3.28 m/sec. Thus this trial is giving satisfactory performance in the stilling basin and also velocity is got reduced to the permissible range. The Velocity observed in the model trial is furnished below. This can be taken as the final trial and the following recommendation has been given to incorporate in the construction of the weir at the site. Table No.1 Statement showing the observed velocity in Trial No. VI

SL NO

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

LS Chainage from the axis of spillway in "m" 40 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500

Observed Velocity in “m/sec”

LEFT

CENTRE

RIGHT

3.28 2.21 1.38 1.71 1.98 0.99 1.40 0.99 0.99 0.99 0.99 1.40 0.99 0.99 0.99 1.40 1.40 0.99 1.40 1.40

2.21 1.98 1.71 1.40 1.40 1.40 1.40 1.40 1.40 1.40 1.40 1.40 1.40 0.90 1.40 0.99 0.99 0.99 0.99 0.99

2.80 1.98 1.71 1.71 1.40 0.50 0.50 0.50 1.40 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 1.40 0.99

HYDRO 2014 International

Figure 2. & 3. Running Model of Pachaiyar River

Figure 4. View of Left side scour vents Hydraulic Jump at stilling basin

Figure 5.

CONCLUSION The spillway and scour vents design approved by the Designs wing is functioning satisfactorily for the given maximum flood discharge and the following alteration is to be made in the surplus course and the stilling basin portion. 1. Introducing friction blocks on the downstream left side scour vent portion with three rows of friction block of size 1.0x1.0x1.0m with 3 numbers in the first and third rows and 2 nos in the second row as shown in the sketch. 2. Raising the Right downstream divide wall between stilling basin weir portion and the Right side scour vent portion to a level of +54.00m. ACKNOWLEDGEMENT The authors acknowledge the services of Designs Wing, PWD, WRO, Chennai and the field engineers for collection of field data and suggestions during the course of model studies. REFERENCES i. Allen. J Scale Models in Hydraulic Engineering ii. Chow, V.T. (1959), Open Channel Hydraulics, McGraw-Hill, New York, NY iii. Elevators Key, Hydraulic Energy Dissipation.

MANIT Bhopal

Page 22

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Experimental Investigations For Estimation of the Height of Training Wall of Convergent Stepped Spillway P. J. Wadhai 1 N. V. Deshpande 2 A. D. Ghare 31 Associate Professor, Department of Civil Engineering, G. H. Raisoni College of Engineering, C.R.P.F. Gate No. 3, Hingna Road, Digdoh Hills, Nagpur - 440 016, Maharashtra, India., Email: [email protected] Principal, Guru Nanak Institute of Engineering & Technology, Kalmeshwar Road, Dahegaon, Nagpur - 441 501, Maharashtra, India, Email: [email protected] 3 Associate Professor, Department of Civil Engineering, Visvesvaraya National Institute of Technology, Nagpur - 440 010, Maharashtra, India, (Corresponding Author), Email: [email protected] 2

ABSTRACT: Amongst hydraulic engineers worldwide, there is enough interest generated for the construction of stepped chutes. Ease of construction and enhanced energy dissipation of flow over the control structure itself, are the primary reasons for its growing popularity. There are a good number of literature references available for the design of stepped spillways with straight side walls, but a very limited literature is available on the design of stepped spillways with convergent training walls. This paper presents the experimental findings carried out on a 45o stepped spillway set up having 1:1 convergent training walls. The step height variation is accounted for, in the proposed expressions which can be used for assessment of the flow bulking and the requirements of the height of training walls, in convergent stepped spillways.

Keywords : Stepped spillways, convergence angle, step height ratio 1. INTRODUCTION : A stepped spillway has conventional ogee spillway profile. However, it is provided with steps from just below the crest up to the toe of the spillway. The provision of steps on the downstream face of the spillway chute increases the rate of energy dissipation and in turn, reduces the size of energy dissipater downstream. A typical cross section of stepped spillway is indicated in Figure 1. Thus a stepped chute not only significantly increases the dissipation energy rate but also decreases the construction costs of the downstream stilling basin. h1 = Flow depth measured vertically above the extreme corner of each step along the training wall Y = Minimum depth of flow immediate after the toe

Hd = Head over crest of the spillway

D = Point of tangency E = Toe of spillway C = Crest

h = Normal size step height

D

Down stream side

Up stream side

H’ = Drop height

h

H

h1 Crest axis

θ

E

Y

Cavitation risk resulting from excessive sub-pressures decreases due to lower flow velocities and occurrence of high amount of air entrainment. But, this aeration produces flow bulking and therefore the spillway requires higher side walls. The effect of convergence enhances this effect due to shock waves and taller training walls are required. In the present study, it is proposed to experimentally determine the effect of converging training walls on flow characteristics of stepped spillway. Literature survey reveals that a limited literature is available on stepped spillways with convergent training walls as compared to stepped spillways having straight training walls. In due course of time, many of the stepped spillways are expected to be made with convergent training walls because of the geological or topographical constraints or due to limited scope for right-of-way caused by urbanization. Sorensen (1985), Peyras et al. (1992), Christodoulou (1993), Chanson (1994), Chamani and Rajaratnam (1999), Barani et al. (2005), Chanson (2006), Chinnarasri and Wongwises (2006), and others focused on study of stepped spillways. Hunt et al. (2008) conducted a study utilizing a three-dimensional, 1:22 scale physical model to evaluate the flow characteristics over a sloping stepped chute ( 3H: 1V) with varying training wall convergence angles. It was found that the required training wall height varied from critical depth for 15o convergence angle to thrice the critical depth at 52o convergence angle. As a follow up work, a major reference on training wall height requirements of convergent stepped spillway was presented by Hunt et al. (2012), wherein a simplified expression was developed to predict the vertical height of training wall as a function of centerline depth of flow. This expression was developed on the basis of simplified control volume momentum analysis and hence can be supposed to be a generalized one. However, more testing of this expression was warranted, due to requirement of an empirical adjustment associated with the force term during the derivation of the proposed expression. In background of this, it was felt necessary to conduct the experiments to develop an expression for estimation of height of training wall for 45o convergent stepped spillways. 2. EXPERIMENTAL FACILITY : Experimental setup consists of a convergent stepped spillway of ogee type 2.66 m long crested weir with stepped chute of (θ = 45° i.e. 1:1) and a toe channel of 0.5 m wide and 10 m long. The side wall of the stepped spillway converges from point of tangency to down the chute with a convergence angle Ø = 45 o . A storage reservoir having 9 square meter plan area and1.75 m depth constructed on the upstream side of the convergent stepped spillway crest. Stepped spillway experimental set up followed by arrangement of water recirculation system consisting of a pump of capacity 10 HP connected with G.I. suction and delivery pipe of 150 mm diameter. The pump fetch the water from underground sump which in turn discharged in to an upstream reservoir through delivery pipe provided with an arrangement for venturimeter with U-tube manometer for flow rate measurement.

Figure 1. Indicative cross section of stepped spillway

HYDRO 2014 International

MANIT Bhopal

Page 23

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Experimental testing was done for model step height (h) of 0.12 m, 0.06 m, 0.03 m and 0.015 m which in turn corresponds to respective step height ratios { H* = (H‟/ h) }of 10, 20, 40 and 80. Water from upstream reservoir flow over the convergent stepped spillway which further allowed to flows freely through a toe channel. For volumetric measurement the flow from toe channel empties in to a collecting tank of plan area 5.31 m2. For measurement of different values of rate of flow in the range of 0.02 m3/sec to 0.064 m3/sec, head over crest of spillway was measured at a distance of 0.15 m upstream of the crest. For measurement of water surface levels along and across the steps and also for measurement of flow depths at other locations vernier type point gauges were used with a sensitivity of 0.1 mm.

maximum depth of flow observed along the converging training walls in dimensionless form and the regime of flow (nappe or skimming) for the different experimental runs. Table 1. Experimental observations and computations for θ = 45º, Ø = 45º and H* = 10

Figure 2 shows the photographs of convergent stepped spillway experimental setup constructed at G. H. Raisoni College of Engineering, Nagpur in collaboration with VNIT, Nagpur, Maharashtra State, India.

Table 2. Experimental observations and computations for θ = 45º, Ø = 45º and H* = 20 Dimensionl ess discharge,

Discha rge, Q, cumec

Discharge per unit width at crest, q, cumec/m

Critical depth of flow, Yc, m

Step height, h, m

0.025

0.00940

0.020735

0.06

71.31

2.1583

Partly Skimming

0.032

0.01203

0.024444

0.06

60.08

2.3733

0.040

0.01504

0.028365

0.06

51.61

2.5867

Partly Skimming Partly Skimming

0.051

0.01917

0.033352

0.06

44.07

2.8017

Skimming

0.060

0.02255

0.037168

0.06

39.75

3.0167

Skimming

hmax /h

Flow regime

Table 3. Experimental observations and computations for θ = 45º, Ø = 45º and H* = 40

Figure 2. Photographs of convergent stepped spillway experimental setup during experimental runs Training walls of convergent stepped spillway and side walls of toe channel were fabricated with acrylic sheets for visibility of flow. Prior to begin with the experimentation, calibration of venturimeter and a triangular was done by the volumetric measurements using a collecting tank. 3.

EXPERIMENTAL OBSERVATIONS AND COMPUTATIONS : All the data sets of observations and computations for the experimental runs for the different step height ratios (H*) and also for smooth ogee spillway are presented in Table 1, Table 2, Table 3, Table 4 and Table 5. These tables also show the

HYDRO 2014 International

Disc harg e, Q, cum ec

Disch arge per unit width at crest, q, cumec /m

0.02 4 0.03 5 0.04 2 0.05 2 0.06 3

0.009 02 0.013 16 0.015 79 0.019 55 0.023 68

Critical depth of flow, Yc, m

Step height, h, m

0.020178

0.03

0.025949

0.03

0.029302

0.03

0.033786

0.03

0.038397

0.03

Dim ensi onle ss disc harg e,

72.7 0 56.6 2 50.2 3 43.4 5 38.1 3

hmax /h

Flow regime

3.7467

Skimming

3.8533

Skimming

4.0300

Skimming

4.2133

Skimming

4.4933

Skimming

Table 4. Experimental observations and computations for θ = 45º, Ø = 45º and H* = 80

Discharge, Q, Cumec

MANIT Bhopal

Discharge per unit width at crest, q, cumec/m

Critical depth of flow, Yc, m

Dimensionle ss discharge, Step height, h, m

hmax /h

Flow regime

Page 24

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

0.020735

0.015

70.77

3.9867

Skimming

0.033

0.012406

0.024951

0.015

58.39

4.4467

Skimming

0.041

0.015414

0.028835

0.015

51.33

4.7067

Skimming

0.051

0.019173

0.033352

0.015

44.07

5.1867

Skimming

0.064

0.024060

0.038802

0.015

37.90

5.8000

Skimming

Table 5. Experimental observations and computations for θ = 45º, Ø = 45º and smooth ogee chute Dimensionles s discharge,

Discharge, Q, cumec

Discharge per unit width at crest, q, cumec/m

Critical depth of flow, Yc, m

Step height, h, m

0.026

0.00977

0.021284

0

0.035

0.01316

0.025949

0.042

0.01579

0.052 0.063

hmax , m

Flow regime

69.72

0.0526

0

56.28

0.0596

0.029302

0

50.23

0.0624

0.01955

0.033786

0

43.65

0.0664

0.02368

0.03840

0

38.13

0.0726

Skimmi ng Skimmi ng Skimmi ng Skimmi ng Skimmi ng

4. ANALYSIS OF EXPERIMENTAL DATA : Due to convergence of the chute walls, the required training wall height is governed by the flow run- up. Visual observations indicated that there were no transverse waves for any of the step height ratios and the air entrainment occurred for nearly all the observations. Experimental data has been collected for plotting the water surface profiles along the centerline of the spillway and also along the convergent walls. As anticipated, the flow depths near the wall were more than those at the centerline of the spillway. The flow depths along wall shall form the basis for deciding the minimum training wall height requirement so that the flow does not overtop the convergent training walls endangering the safety of the structure. Figure (3) illustrates the observed water surface profiles along the wall for different discharge for a step height ratio H* = 40. As the maximum depth of flow along the wall (hmax) would determine the training wall height, a dimensionless plot showing its variation with dimensionless discharge is presented in Figure (4). The regression equations have been obtained and are as follows.

These expressions are proposed to be used for computation of training wall height of convergent stepped spillway with convergence angle of 45o and chute slope of 1:1. The maximum flow depths along wall depths were compared with the corresponding critical depths. For H*=80, the maximum flow depth was found to be between 2.25Yc to 2.9Yc, for H*= 40, the maximum flow depth was found to between 3.5Y c to 5.6Yc, for H*= 20, the maximum flow depth was found to lie between 4.85Yc to 6.25Yc whereas for H*= 10, the maximum depth of flow was observed to be in the range of 5.35Y c to 7.6Yc. 1.4000 1.2000 Stepped Spillway Profile

1.0000 0.8000

H* = 40, Q1 = 0.024 Cumec

0.6000

Elevation, m

0.009398

0.4000 0.2000

0.0000 -0.50

0.00

0.50

1.00

1.50

Station, m

Figure 3 (a). Water surface profiles along the side wall of spillway for step height ratio, H* = 40, Q1 = 0.024 cumec. 1.4000 1.2000 Stepped Spillway Profile

1.0000 0.8000

H* = 40, Q2 = 0.035 Cumec

0.6000

Elevation, m

0.025

0.4000 0.2000 0.0000 -0.50

0.00

0.50

1.00

1.50

Station, m

Figure 3 (b). Water surface profiles along the side wall of spillway for step height ratio, H* = 40, Q2 = 0.035 cumec. (1) 1.4000 1.2000 Stepped Spillway

(2) Profile

1.0000 0.8000

H* = 40, Q3 = 0.042 Cumec

Elevation, m

0.6000

(3)

0.4000

0.2000 0.0000 -0.50

0.00

0.50

1.00

1.50

Station, m

(4)

HYDRO 2014 International

Figure 3 (c). Water surface profiles along the side wall of spillway for step height ratio, H* = 40, Q3 = 0.042 cumec.

MANIT Bhopal

Page 25

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

1.4000

1.2000 Stepped Spillway Profile

1.0000 0.8000

H* = 40, Q4 = 0.052 Cumec

Elevation, m

0.6000 0.4000 0.2000

0.0000 -0.50

0.00

0.50

1.00

1.50

Station, m

Figure 3 (d). Water surface profiles along the side wall of spillway for step height ratio, H* = 40, Q4 = 0.052 cumec. 1.4000 1.2000

Stepped Spillway Profile

1.0000 0.8000

H* = 40, Q5 = 0.063 Cumec

Elevation, m

0.6000 0.4000 0.2000

0.0000 -0.50

0.00

0.50

1.00

1.50

Station, m

Figure 3 (e). Water surface profiles along the side wall of spillway for step height ratio, H* = 40, Q5 = 0.063 cumec. 1.4000

1.2000 H* = 40, Q1 = 0.024 Cumec 1.0000 H* = 40, Q2 = 0.035 Cumec 0.8000 H* = 40, Q3 = 0.042 Cumec

5. CONCLUSIONS : Stepped spillways with convergent training walls will have to be employed when there is limited space available for spillway rehabilitation work. Only a few guidelines are available in the literature for design of convergent stepped spillways, a three dimensional experimental study has been carried out on 45 o convergent stepped spillway having 1:1 chute slope and different step heights. The flow over the convergent stepped spillway was observed to be air entrained and more bulked as compared to ogee spillway. With increase in dimensionless discharge, the maximum flow depth at the convergent training wall normalized by the step height, was found to decrease. Based on the experimental observations and its analysis, the regression equations for maximum depth of flow near the converging walls { Eq. (1) to (4)} have been proposed. A high value of coefficient of determination for all the regression equations indicated that the correlation was good. In general, the maximum flow depth near the convergent training wall was found to lie between 2.25 to 7.6 times of the critical depth of flow, depending up on the step height ratio. The regression equations presented in this paper, may be useful for the hydraulic designers engaged in estimation for deciding the appropriate training wall height for convergent stepped spillways. However, more experimental studies with different convergence angles shall be required, to formulate more generalized expressions for estimation of requirement of adequate training wall heights for convergent stepped spillways. 6. ACKNOWLEDGEMENTS : The research presented in this paper is based on a research project funded by Raisoni Group of Institutions, India, which is gratefully acknowledged. 7. NOTATION :

0.6000 H* = 40, Q4 = 0.052 Cumec 0.4000

Elevation, m

H* = 40, Q5 = 0.063 Cumec 0.2000

0.0000 -0.40

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

Station, m

Dimensionless Maximum Depth of Flow along the Side Wall , hmax / h

Figure 3. Water surface profiles along the side wall of spillway for step height ratio H*= 40 with varying discharge i.e. Q1 = 0.024 cumec, Q2 = 0.035 cumec, Q3 = 0.042 cumec, Q4 = 0.052 cumec, Q5 = 0.063 cumec. 7.0000

H* = 10 H* = 20

6.0000

H* = 40

5.0000

H* = 80 4.0000

y = 9.455x-0.48 R² = 0.978

3.0000

y = 20.56x-0.52 R² = 0.988

2.0000

y = 12.21x-0.28 R² = 0.929

1.0000

y = 48.87x-0.59 R² = 0.992

0.0000 0.00

20.00

40.00

60.00

80.00

100.00

Dimensionless Discharge , Q / (Hd5/2.g1/2)

Figure 4. Dimensionless maximum depth of flow along the side wall versus dimensionless discharge

HYDRO 2014 International

The following symbols are used in this paper : A = L * Hd = Area of flow at crest of spillway; A1 = B * Y = Area of flow at toe of spillway; B = Width of flow channel; C = Discharge coefficient; Er = ∆E / Eo = Relative energy dissipation; Eo = H + 1.5 Yc = Energy at crest of spillway; Et = Y + (V12/ 2.g ) = Energy at toe of spillway; g = Acceleration due to gravity; h = Normal size step height; hmax = Maximum depth of flow observed along the converging training wall; h1 = Depth of flow observed along the converging training wall; H = Datum head measured from toe up to crest of Spillway; Hd = Head over crest of spillway; H' = Drop height; H* = H' / h = Step height ratio; L = Length of crest; Lr = Lp /Lm = Scale ratio; n = Number of regular size steps; q = Q / L = Intensity of Discharge; Q = C * L * (Hd)1.5 = Rate of flow i.e. Discharge over crest of spillway;

MANIT Bhopal

Page 26

International Journal of Engineering Research Issue Special3 R V V1 Y Yc ∆E Ø θ 8.

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

= Hydraulic radius; = Q / A = Velocity of flow at crest of spillway; = Q / A1= Velocity of flow; = Depth of flow; = Critical depth of flow; = Eo - Et = Energy loss due to stepped spillway; = Convergence angle; = Chute angle; REFERENCES :

i. Sorensen, R. M. (1985). “Stepped spillway hydraulic model investigation.” J. of Hydraul Eng., 111(12), 1461 - 1472. ii. Peyras, L., Royet, P. and Degoutte, G. (1992). “Flow and energy dissipation over stepped gabion weirs.” J. of Hydraul Eng., 118(5), 707717. iii. Christodoulou, G. C. (1993). “Energy dissipation on stepped spillways.” J. of Hydraul Eng., 119(5), 644 - 649. iv. Chanson, H. (1994). “Hydraulics of skimming flow over stepped channels and spillways.” J. of Hydraul Res., 32(3), 445 - 460. v. Chanson, H. (1994 a ). “Comparison of energy dissipation between nappe and skimming flow regime on stepped chutes.” J. of Hydraul Eng., Res., IAIHR, 32(2), 213 - 218. vi. Chamani, M. R. and Rajaratnam, N. (1999). “Characteristics of skimming flow over stepped spillways.” J. of Hydraul Eng., 125(4), 361 - 368. vii. Barani, G. A., Rahnama M. B. and Sohrabipoor, N. (2005). “Investigation of flow energy dissipation over different stepped spillways.” American Journal of Applied Sciences., ISSN 1546-9239, 2 (6): 1101- 1105. viii. Chanson, H. (2006). “Hydraulics of skimming flow on stepped chutes : The effects of inflow conditions.” J. of Hydraul., Res., 44(1), 51 - 60. ix. Chinnarasri, C. and Wongwises, S. (2006). “Flow pattern and energy dissipation over various stepped chutes.” J. of Irrig. Drain. Eng., 132 (1), 70 - 76. x. Sherry L. Hunt, Kem C. Kadavy, Steven R., and Darrel M. Temple (2008). “Impact of converging chute walls for roller compacted concrete stepped spillways.” J. of Hydraul Eng., ASCE, 134 (7), 1000 - 1003. xi. Sherry L. Hunt, Darrel M. Temple, Steven R., Kem C. Kadavy, and Greg Hanson. (2012). “Converging stepped spillways: simplified momentum analysis approach.” J. of Hydraul Eng., ASCE, 138 (9), 796 - 802.

Studies For Location of Bridges in the Vicinity of Existing Hydraulic Structures B. Raghuram Singh1 , Dr. R. G. Patil2 , M. N. Singh3 1 Research Officer, CWPRS, Pune, India; Email:[email protected] 2 Chief Research Officer, CWPRS, Pune, India; Email:[email protected] 3 Joint Director, CWPRS, Pune, India; Email:[email protected]. ABSTRACT: The rapid urbanization and increased traffic volume has forced the planners to construct additional bridges to cross the river passing through cities. These bridges are being constructed at increased interval, adjacent to the existing bridges, and barrages. The case being discussed here is the River Yamuna at Delhi. The river in this reach is constricted with the construction of series of bridges. Due to this

HYDRO 2014 International

construction, the passage of flood and silt gets modified near these bridges and creates problems to these hydraulic structures over a long period, reflecting either afflux or drawdown. The objective of the present study is to decide the suitable location of the proposed bridge in the close vicinity of existing bridge and a barrage. The studies were conducted on a composite model with a horizontal scale of 1:300 and vertical scale of 1:60 constructed at CWPRS, Pune. Series of studies were conducted to assess the movement of sediment through the reach by changing the location of the proposed bridge. The results and findings of the same are presented in this paper. Keywords: bridge pier, velocity, discharge intensity, water level 1. INTRODUCTION Present day New Delhi, national capital of India was original situated on the western bank of River Yamuna. After the independence and receiving tremendous impetus, New Delhi has developed into a populous city extending on either banks of River Yamuna. Being national capital region (NCT), Government of India, the state of Delhi and adjoining states have accorded high priority for the infrastructure development to connect the satellite cities around the city of Delhi. This envisage construction of bridge across the River Yamuna in addition to the existing barrages and bridges. The River Yamuna which drains the southern Himalaya region, originates in Yamunotri and flows through the gangetic plain beyond Yamuna nagar, enters the state of Delhi at Palla and leaves it after traversing a distance of about 50 km near the village of Jaitpur. The sediment being very fine the river is alluvial in nature. The Yamuna joins the Ganga at Allahabad. The huge pressure of development has forced the authorities to construct roads and bridges to connect the areas on either banks of the river Yamuna at Delhi. These bridges are being proposed to be constructed adjacent to existing barrages and bridges. The water way and alignment automatically gets fixed up due to the existing structures. However, the afflux gets accumulated and possibly may lead to additional resistance to the flow. Afflux may affect adversely the sensitive flooding conditions existing on the upstream areas of Delhi. In addition the reduction in velocities over the length of the river due to increase in depth (Afflux) of flow may accentuate the sediment deposition, which over a long period may lead to aggradation of river bed and increase in the flood levels. These issues need to be assessed before construction of bridge and avoid any such difficult situation. Model studies were conducted at CWPRS on a comprehensive model of River Yamuna at Delhi built to a scale of 1:300 horizontal and 1:60 vertical for a proposed bridge to be constructed between Okhla barrage and DMRC bridge. The existing two structures were at a distance of around 85 m and it was proposed to insert one more road bridge between these two structure or adjacent to them based on the model studies. 2. PHYSICAL MODEL The existing mobile bed model of river Yamuna constructed to a horizontal scale (L) of 1:300 and a vertical scale (D) of 1:60 covering a river reach from Palla to Jaitpur was utilized for present model studies.

MANIT Bhopal

Page 27

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 3. Model prototype conformity ( Q= 7022 m3/s ) Figure 1. Plan of River Yamuna at Delhi In order to reproduce proper bed movement and roughness, the model bed was made mobile by laying sand having a mean diameter (D50) of 0.34 mm. Figure 2 shows the grain size distribution of sand used. In order to establish flood slope and to observe water levels at various locations, gauges were installed on the right side upstream and downstream of Wazirabad barrage, upstream of ISBT road bridge and upstream of Indraprastha barrage and on the left side at Kailashnagar downstream of old rail-cum-road bridge, near Okhla weir and at the proposed road bridge site.

Figure 2. Grain size distribution curve for the sand used in the model 3. PROVING STUDIES: The maximum flood discharge of 7,022 m3/s occurred in Yamuna at Delhi in the year 1988. Discharge equivalent to 7,022 m3/s was let into the model and by controlling the gauge upstream of the Indraprastha barrage as per the gauge discharge curve, water levels were observed at various gauge locations. Figure 3, shows the comparison of the water levels observed on the model. These are in close agreement with the prototype values. In view of this, the model was considered as "proved"

HYDRO 2014 International

4.0 MODEL STUDIES Studies were carried out to examine the following aspects of design (i) Suitable location of the proposed bridge. (ii) Effect of water levels and velocities on the proposed bridge. (iii) Flow conditions in the vicinity of the bridge. The model studies were carried out for the following discharges. (a) 7,022 m3/s (2.48 lakh cusec) (maximum discharge observed in 1988 at Wazirabad Barrage) (b) 9,910 m3/s (3.5 lakh cusec) (design discharge considered for ISBT bridge and bridge proposed subsequently on Yamuna) (c) 12,750 m3/s (4.5 lakh cusec) (check flood for substructures, foundation and protection works suggested by Central Water Commission) Bridge location: The project authorities were interested in locating the proposed road bridge between the Okhla barrage and the under construction DMRC bridge spaced about 85 m apart. This helped in connecting the road bridge with the approach road on either banks of the river. However, insertion of bridge between the two existing structures would entail introduction of additional resistance to the flow which could pose difficulties in general movement of sediment from the Okhla barrage. This difficulty in a long run can pose aggradation of bed on upstream which in turn can increase the flood level. To avoid this, it was decided to study the effect of the bridge insertion at various possible locations which were decided after discussion with the project engineers.Model studies were carried out for the bridge alignment at the following locations. Alignment - 1: Studies for the proposed road bridge approximately 57.5 m downstream of Okhla barrage (i.e. 27.5 m upstream of proposed DMRC bridge). Alignment - 2: Studies for the proposed road bridge approximately 185 m downstream of Okhla barrage (i.e. 100 m downstream of proposed DMRC bridge). Alignment - 3: Studies for the proposed road bridge approximately 50 m downstream of Okhla barrage (i.e. 35 m upstream of Proposed DMRC bridge). Alignment - 4: Studies for the proposed road bridge

MANIT Bhopal

Page 28

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

approximately 120 m downstream of Okhla barrage (i.e. 35 m downstream of Proposed DMRC bridge). 4.1 Model studies with existing condition The preliminary model studies include, assessing and understanding the flow conditions existing in and around the structures to be introduced. This study is conducted by passing the predecided Alignm ent1&3 Alignm ent-2

Figure 6. Flow pattern in the vicinity of pattern in the vicinity of proposed bridge with Q = 12,750 m3/s (Alignment –1) Figure7. Flow proposed bridge with Q=12,750m3/s (Alignment –2)

Alignm ent-4

Figure 4. Model set-up with existing conditions

Figure 5. Flow pattern in the vicinity of proposed bridge with Q = 12750 m3/s (Alignment-1)

discharge through the model, but the structure of the proposed bridge is not inserted. However, to help recognize the structure, position and alignments are marked in such a way that it does not affect the flow conditions. In this case all four alignments are marked on the model as shown in Fig 4. And the experiments were conducted for above referred three discharges. The flow conditions were observed. The water surface elevation, and velocities at critical points were measured. These data would be used to compute the discharge intensities and afflux later. The measured values are presented in Table. 1. Fig. 5 depicts flow pattern along the proposed bridge under existing condition with river discharge of 12,750 m3/s (Alignment – 1). 4.2 Model studies with proposed road bridge The road bridge was proposed to cross the river Yamuna downstream of Okhla barrage, however its exact location was not decided. It was thought to be located between the Okhla barrage and the under construction DMRC railway bridge. The space available between these two structures was only 85 m. In view of this, four alignments (Alignment-1, 2 , 3 and 4 as discussed above) were studied on the model separately to decide the location of bridge and its effect on the overall functioning of barrage and movement of sediment downstream through various structures. The road bridge along the alignments 1 to 4 were separately inserted on the model and the studies were conducted. The measurements such as water levels and velocities at critical points were taken. General flow conditions and its effect on the river behavior was also assessed. The data in respect of velocity and water surface elevation is presented in Table 1. Fig 6, 7, 8 and 9 show the variations of flow pattern for four alignments from alignment-1 to 4 respectively.

HYDRO 2014 International

Figure 8. Flow pattern in the vicinity of pattern in the vicinity of proposed bridge with q = 12,750 m3/s ( Alignment –3) Figure 9. Flow proposed bridge with q = 12,750 m3/s (Alignment – 4)

Table 1. Maximum water levels and velocities observed during model studies Na me of the Str uct ure

Ok hla Bar rag e Pro pos ed Roa d Bri dge Ok hla Bar rag e Pro pos ed DM RC Bri dge Pro pos ed Roa d Bri

Case 1. 27.5 m upstream of proposed DMRC bridge (Alignment -1) Q= 7022 m3 /s Q= 9910 m3 /s Q= 12750 m3 /s Without With Without With Without With Bridge Bridge Bridge Bridge Bridge Bridg e WL V WL V W V W V WL V W V (m) ( (m) (m) L (m) L (m) (m) ( L m (m) (m) m ( ( ) ) m m ) ) 203 2. 203 2.0 20 2.9 20 2.9 204 3. 2 3 .19 0 .3 2 4.3 0 4.4 5 .58 4 0 . 0 8 6 4. 5 9 4 2 203 3. 203 3.0 20 3.8 20 3.8 204 4. 2 4 .14 0 .6 3 4.1 0 4.3 5 .41 5 0 . 5 4 4. 5 7 6 6

Case 2. 100 m downstream of proposed DMRC bridge (Alignment -2) 3. 203 3.1 20 3.8 20 3.8 204 4. 2 0 .23 0 4.2 2 4.4 5 .57 5 0 5 8 4 5 4. 8 0 203 3. 203 3.0 20 3.8 20 3.8 204 4. 2 .12 0 .18 1 4.1 4.3 2 .41 5 0 5 3 0 4. 6 6 203 .18

202 .98

MANIT Bhopal

2. 9 1

203 .06

2.9 4

20 4.0 5

3.6

20 4.2 5

3.6 5

204 .25

4. 0

2 0 4. 5 2

4 . 6

4 . 5 3

4 . 1 0

Page 29

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

dge Ok hla Bar rag e Pro pos ed Roa d Bri dge

203 .19

Ok hla Bar rag e Pro pos ed DM RC Bri dge Pro pos ed Roa d Bri dge

203 .18

203 .14

Case 3. 35 m upstream of proposed DMRC bridge (Alignment -3) 2. 203 2.0 20 2.9 20 2.9 204 3. 2 0 .28 3 4.2 1 4.4 7 .59 4 0 2 9 5 5 4. 9 0 3. 203 3.1 20 3.8 20 3.8 204 4. 2 0 .25 0 4.1 2 4.3 5 .41 5 0 5 5 3 1 4. 7 4

Case 4. 35 m downstream of proposed DMRC bridge (Alignment -4) 3. 203 3.1 20 3.8 20 3.8 204 4. 2 0 .24 2 4.2 1 4.4 5 .57 5 0 2 8 3 2 4. 8 1 203 3. 203 3.1 20 3.8 20 3.8 204 4. 2 .12 0 .19 5 4.1 2 4.3 7 .43 5 0 5 4 1 1 4. 6 9

203 .03

3. 0 2

203 .11

3.1 0

20 4.0 8

3.6 2

20 4.2 6

3.7 1

204 .27

4. 0 2

2 0 4. 5 5

6. DISCUSSION OF RESULTS 3 . 5 5 4 . 5 5

4 . 5 8 4 . 5 5

4 . 1 5

5. QUALITATIVE STUDIES Model studies were conducted with four alternate alignments with and without the proposed road bridge. During model studies of alignment -1 and alignment -3, it was observed that the sediment was depositing at the upstream of Okhla barrage and in between the Okhla barrage and under construction DMRC bridge. In view of this, to assess the effect of sediment movement through the Okhla barrage and downstream bridge qualitative studies were conducted by feeding particular quantity of sediment in to the flow about a kilometer upstream of the barrage. The movement of sediment with proposed road bridge at about 35 m upstream (Alignment -3) and 35 m downstream (Alignment – 4) of proposed DMRC bridge was studied by the silt injection. In case of studies with alignment-3, it was observed that relatively large quantity of sediment was depositing on the upstream and through the spillway bays as shown in Fig.10 compared with the aggradation seen in respect of alignment -4 as shown in Fig. 11.

Okhla barrage, under construction metro rail bridge and proposed road bridge are closely located in river reach of about 85 m. These structures with a waterway of 552 m of barrage and 574 m for bridges hold the river to a fixed course at their locations and therefore there is no possibility of any meandering. The river is about a kilometer wide in this reach and has already been constricted to about 552 m due to the construction of barrage and its guide bunds. For the alignment – 1, the maximum water levels observed at the proposed road bridge and Okhla barrage under existing conditions was 204.41 m and 204.58 m respectively with a discharge of 12750 m3/s. With the proposed road bridge in position, the maximum water levels observed at the bridge axis and Okhla barrage was 204.76 m and 204.92 m with a discharge of 12750 m3/s. This indicates to an afflux of about 35 cm near the proposed bridge axis and 34 cm at Okhla barrage. The water levels observed at the proposed road bridge (Alignment-2) and Okhla barrage without and with the bridge were 204.25 m and 204.52 m and 204.57 m and 204.80 m respectively with the discharge of 12750 m3/s. This indicates to an afflux of 27 cm near the proposed road bridge and 23 cm near the Okhla barrage. For alignment -3, the maximum water levels observed at the proposed road bridge and Okhla barrage under existing conditions was 204.41 m and 204.74 m respectively with a discharge of 12750 m3/s. With the proposed road bridge in position, the maximum water levels observed at the bridge axis and Okhla barrage was 204.59 m and 204.90 m with a discharge of 12750 m3/s. This indicates to an afflux of about 33 cm near the proposed bridge axis and 31 cm at Okhla barrage. The water levels observed at the proposed road bridge (Alignment - 4) and at the Okhla barrage without and with the bridge of waterway 574 m were 204.27 m and 204.55 m and 204.57 m and 204.81 respectively with the discharge of 12750 m3/s. This indicates to an afflux of 28 cm near the proposed road bridge and 24 cm at Okhla barrage. The studies conducted with alignment 3 & 4, by feeding equivalent sediment on the upstream of Okhla barrage, indicated that comparatively higher deposition of sediment on upstream of Okhla barrage and through the road bridge – DMRC bridge in case of alignment-3 when compared with alignment-4. The afflux measured at the Okhla barrage due to the bridge alignment-3 was 31 cm and due to bridge alignment 4, it was 24 cm. In view of this the alignment-4 is performing better than the alignment-3. 7. CONCLUSIONS

Figure 10. A view of proposed road bridge 35 u/s of DMRC bridge (alignment - 3)

Figure 11. A view of proposed road bridge 35 d/s of DMRC bridge (alignment -4)

HYDRO 2014 International

From the studies carried out with river discharges of 7022 m3/s, 9910 m3/s and 12750 m3/s following important conclusions were made : The site for the proposed road bridge 35 m downstream of proposed DMRC bridge (Alignment – 4) was satisfactory as revealed by model experiments from the hydraulic point of view and silt flow conditions.

MANIT Bhopal

Page 30

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The waterway of 574 m (14 spans of 41 m centre to centre each) for the proposed bridge 35 m downstream (Alignment -4) of DMRC bridge didn‟t cause any undesirable flow conditions at the proposed bridge axis and at the Okhla barrage. The movement of sediment with proposed road bridge at about 35 m upstream (Alignment -3) and 35 m downstream (Alignment -4) of proposed DMRC bridge was studied by the silt injection. It was seen that there was qualitatively large quantity of sediment deposition on upstream of barrage, through the spillway bays and at proposed road bridge in the model for Alignment -3 rather than for the Alignment-4. This will cause aggradations of river bed near proposed road bridge. In view of this, the proposed road bridge in between the barrage and under construction DMRC bridge was not recommended. ACKNOWLEDGEMENT We wish to express our deep sense of gratitude to Shri. S. Govindan, Director, CWPRS for constant encouragement and valuable suggestions during the course of this studies and kind permission given for publishing this paper. REFERENCES i. CWPRS Technical Report No.5092 of July 2013, ―Hydraulic model studies for the proposed road bridge downstream of Okhla barrage across river Yamuna at New Delhi‖. ii. Engelund.F.(1996).Hydraulic Resistance of Alluvial streams, Journal Hydraulic Division, ASCE, March . PP 315-327. iii. K.G.Ranga Raju, R.J.Garde and H.S.Yadav (1996) Modelling Bed level variations in Alluvial Streams, ISH, Vol. 2, PP 28-43. iv. S. B Kulkarni and V. M. Wakalkar (1998) Hydraulic Model Studies for Improvement of flow conditions at Samal Barrage, ISH, Vol.4, PP 24-33. v. SMITH D.W., (1977). Why do Bridges fail?, Civil Engineering, American Society of Civil Engineers.

Study of Sharp-Crested Triangular Weir M. Shaheer Ali1Talib Mansoor2 P. G. Student, Department of Civil Engineering, A.M.U. Aligarh 2 Associate Professor, Department of Civil Engineering, A.M.U. Aligarh E-mail: [email protected] 1

ABSTRACT : Triangular weir is a simple form of weir best suited for low discharge and is free from aeration difficulties. It is mostly used in various branches of engineering like hydraulics, environmental, chemical and irrigation for the purpose of discharge measurement. Earlier studies

HYDRO 2014 International

conducted on triangular weir indicate that the discharge coefficient related to head or head to weir height ratio covering a limited range of head and vertex angles. Further, no generalized equation proposed to compute either discharge coefficient or discharge for any head and vertex angle. In this study, a total of 65 experimental runs were taken for five weir vertex angles (from 30◦ to 90◦) at apex elevation of 20cm. Using the general formula for triangular weir dimensionless discharge and dimensionless head has been defined that helps in merging all the data points of five angles to one single curve. A generalized equation between dimensionless discharge and dimensionless head has been obtained. The maximum error obtained in the discharge computed from this equation is ±5%. This equation also validates the data of previous study (Wahaj, 1999). Keywords:Weir vertex angle, Discharge coefficient, Dimensionless discharge, Dimensionless head, generalized equation. 1. INTRODUCTION A weir is built across a river (or stream) in order to raise the level of water on the upstream side and to allow the excess water to flow over its entire length to the downstream side. Thus a weir is similar to a small dam constructed across a river, with the difference that a dam allows excess water to flow to the downstream side, only through a small portion called spillway, whereas a weir makes the excess water to flow over its entire length. Weirs have been mostly used for flow measurement in open channels. Since 1500 A.D. weirs have been a subject of interest for the mankind. In 1885, the investigations of Francis led to the application of weirs for accurate discharge measurements. Investigations of Thomson (1858) and Bazin (1888-1898) promoted the use of weirs. The triangular weir is used widely for measuring the flow of liquids in flumes and open channels. It is inexpensive, easy to use and maintain.Several assumptions are made to obtain a definite relationship between the actual discharge through the weir and the head obtained on the weir. These structures have been very often used in laboratories and in fields to know the nature of flow, nappe profiles and to determine the coefficient of discharge (Cd). The discharge coefficient takes into account the effects which are ignored in the derivation of the discharge equation for a triangular weir such as capillary action of water, viscosity, surface tension, approach velocity and influence of weir contraction on the nappe profiles. Thomson (1858) recommended Cd = 0.593 and 0.617 for the 90 o and 127o notch angles respectively. Barr (1910) concluded that the coefficient was increased by roughness and projections on the upstream face of the weir. Barralso concluded that the coefficient was independent of channel width if the width was at least eight times the head. For channel widths less than 8h, the coefficient increased asthe width decreased. Strickland (1910) quoted formula for 90°-notch weirs on the basis of Barr's experiments as . Cone (1916) gave the followi ng for mula: ,

MANIT Bhopal

Page 31

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Where, S = side slope of the notch, expressed decimally, and . Greve (1932) gave the following formula:

Lenz (1943) gave the formula:

WhereN, are functions of notch angle. Kindsvater-Shen (1980) developed a formula for the discharge over a triangular weir with angles notch angles between 20o and 100o, given by

,

WhereCe: coefficient of discharge, he: effective head (= h +kh), Ceis a function of three variables, i.e. Ce = f (h/p, p/B, Ɵ) where, kh is an experimentally determined quantity in metre which accounts for the combined effects of viscosity and surface tension .Capetillo et. al (2013) developed a discharge coefficient equation for sharp crested triangular weirs on the basis of free vortex theory as described by Bagheri and Heidarpour (2010); and measurement of the upper and lower nappe profiles using an adaptation of the low-speed photographic technique proposed by Salvador et al. (2009). The equation is given by:

Where Vb: lower nappe velocity at the maximum elevation section of the lower nappe (m/s); Rb: radius of the streamline curvature at the lower nappe of the profile (m); k: nonconcentricity coefficient; Y: the flow depth at the maximum elevation section of the lower nappe (m). From the literature surveyed above, it is clear that the coefficient of discharge is given for individual angles and most of the investigators related Cd with h and some of them related it with the wetted perimeter, Reynold‟s number, and Weber number. No generalized equation exists to compute discharge for all angles of the triangular notch. In the present study, an attempt has been made to compute discharge covering a wide range of notch angles and heads. The objective of the present study is to develop a generalized equation for the discharge through a triangular weir and to establish a relationship between the discharge coefficient, notch angle, h/p and p/B using an experimental data and regression analysis.

WhereQ is the discharge (m3/s); Ɵ is the notch angle; h is the head above the crest (m); Cdis the discharge coefficient (dimensionless). 2. EXPERIMENTAL SETUP: The experiments were conducted in a horizontal, rectangular (75 cm wide and 53 cm deep), prismatic glass walled channel having cement plastered bottom (Photo 1). Weirs were made of G.I. sheets. Weirs were installed at a distance of 8.5 m from the upstream end of the channel. Water was supplied through an inflow pipe from laboratory overhead tank provided with an overflow arrangement to maintain the constant head. A sharpcrested triangular weir was installed at a desired angle, Ɵ and apex height p. Discharge was controlled by means of a control valve. The flow was allowed in the set-up to fill the upstream channel up to the apex level of the triangular weir. The apex level was recorded with a point gauge of accuracy 0.1 mm. The discharge was allowed to flow in the channel and become steady and then the head difference in the two limbs of differential manometer attached to the bend meter mounted on the supply pipe was measured. The discharge flowing in the channel was computed using an accurate Calibration curve prepared for bend meter. Under the same steady state flow conditions point gauge reading at the free surface was recorded near the center of the channel at 1 m upstream of the weir to avoid the curvature effect of water surface (Photo 2). Five such readings were taken and averaged to obtain a precise value of gauge reading. Head over the apex was obtained by subtracting the apex level from averaged free surface reading. The discharge was changed by means of the control valve and a number of runs were taken to cover a wide range of h/p. The entire procedure was repeated for other weirs having different apex angles. Table - 1 gives an account of the different parameters of the triangular weir taken into consideration in this study. A total of 65 experimental runs were taken.

1.1 Governing equation: Fig. 1: Experimental Setup The discharge through a triangular weir is given by: (1)

HYDRO 2014 International

MANIT Bhopal

Page 32

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Fig. 2: Q v/s h Following best fit equations for Q – hhave been obtained: R2 = 0.9999

= 30◦,

Photo 1: Upstreamview of the channel

2

= 45◦,

R = 0.9999 2

= 60◦,

R = 0.9987 2

= 75◦,

R = 0.9818

(2) (3) (4) (5)

= 90◦, R2 = 0.9968 (6) These graphs show that there is an increasing trend for the discharge Q with increase in head above the crest.Further the discharge curve for 90o weir lies at the top while for 30o weir lies at the bottom. It is obvious from this figure that for a particular head discharge over 30o weir will be the least whereas discharge through 90o weir will be the highest. In other words, for a particular discharge, the head above the apex will be less in 90 o weir and more in 30o weir. Eqs (2) – (6) can be written as Photo 2: Point Gauge Table -1: Range of parameters for the triangular weir Notch angle

p (cm)

h (m)

Qo (m3/s)

Fr

( Ɵ, 0) 30

No. of runs

20 20

60

20

75

20

90

20

0.00510.013 0.00760.0184 0.0070.022 0.00750.033 0.00650.0356

0.00280.0053 0.00450.0093 0.01390.0304 0.01510.0452 0.02040.0531

8

45

0.1770.2605 0.17180.2466 0.1480.2355 0.13050.2469 0.1080.2233

Q= Ahn A generalized equation in the above form could not be obtained due a large scatter in the values of coefficients A and n and hence a large % error in the computed discharge. 3.2 Variation of Cd v/s h/p: Using Eq (1) Cd was computed and plotted against h/p as shown in Fig. 3

9 `11 21 16

3. ANALYSIS AND RESULTS: 3.1 Variation of discharge with head: The variation of discharge with head for five triangular weirs tested in the present study is shown in Fig. 2. Fig. 3: Cd v/s h/p

HYDRO 2014 International

MANIT Bhopal

Page 33

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The above graphs show that the value of Cd decreases as the angle is changed from 30o to 45o. If the value of notch angle is further increased from 60o to 90o, the values of Cd starts increasing. This variation is noticeable in the lower range of h/p (i.e., 0.5 to 1). In the higher range of h/p, the variation in Cd is insignificant. The variation of Cd with h/p shows a decreasing trend for angles 30◦, 45◦and 60◦ and an increasing trend for angles 75 o and 90o. However, the RMSE values for the best fit curves are small enough. So a generalized equation of the form could not be obtained as the trend of A and B shows a large scatter and the percentage error in the discharge computed with this equation is high. Fig. 4: Qn v/s Hn

3.3 Generalized equation Therefore an attempt has been made to make the discharge and head dimensionless in order to obtain a generalized equation for the discharge over a triangular weir. Rewriting Eq. (1) as:

The resulting discharge equation is agreement diagram in Fig. 8 shows that the computed discharge lies within an error band of ± 5 %

. The

Dividing both sides by p5/2,

Thus, both sides are changed to dimensionless quantities and can be expressed as: Fig. 5: Qov/s Qc Where,

Hn = h/p The data obtained from the experimental work is converted in the form of above mentioned dimensionless discharge Qn and dimensionless head Hn and graph plotted between Qn and Hn as shown in fig. 7:

SYMBOLS USED: A = flow area Cd = discharge coefficient Ce = effective discharge coefficient Fr = Froude number g = gravitational acceleration Q = discharge over weir p = weir height h = head above crest B = channel width

Ɵ = included angle at the apex of the triangle Qno = observed non-dimensional discharge Qnc= computed non-dimensional discharge REFERENCES i. Bengtson H.H. (), Sharp Crested Weirs for Open Channel Flow Measurement. ii. Bos (1989), Discharge measurement structures. iii. Capetillo Et.al (), Discharge coefficient analysis for triangular sharpcrested weirs using low-speed photographic technique. iv. Chow V.T. (), Open channel flow.

HYDRO 2014 International

MANIT Bhopal

Page 34

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

v. Hager W. H. (), Discharge measurement structures vi. Horton R.E. (1907), Weir experiments, coefficients, and formulas (Revision of paper no. 150, Department of the Interior United States Geological Survey). vii. Jain A.K.(), Fluid mechanics. viii. Jiwani R., Steffen P. E. (), Methods of Flow Measurement for Water and Wastewater. ix. King h.w.(1996), Handbook of hydraulics. x. Larsen D.C. (1992), Water measurement. xi. MasoudGhodsian (),Stage discharge relationships for triangular weir.Rao N.S. (), Theory of weirs. xii. Smith E.S., Providence, R. I. (),The v-notch weir for hot water.

Study of Elliptically Shaped Sharp Crested Weirs N.P. Singh1 R. Singh2 Ujjain Engineering College, Ujjain, Sanwer road Ujjain (MP), Pin 456010, India 2 Govt. Engineering College, Ahmadabad, Gujarat, Pin 380001, India Email: [email protected] 1

ABSTRACT : This is a study about behavior of elliptically shaped sharp crested weirs placed across open channels and used as flow measuring devises. Effects of surface tension and viscosity on coefficient of discharge are studied for values of Channel Reynolds Number less than 2000. Value of coefficient of discharge is established for Channel Reynolds Number greater than 2000. Suitability of elliptically shaped sharp crested weirs as flow measuring devices are analyzed. The study has generated experimental data for a new shape and a less studied flow regime. Keywords: Sharp crested weir, coefficient of discharge, head discharge characteristics 1. INTRODUCTION Sharp crested weirs are commonly used devices for flow measurement in open channels. Their advantage lies in the fact that they are cheaper as compared to electronic flow devices. In fact they become part of the same hydraulic structure in which they are installed. The accuracy of discharge measurement depends upon several factors such as the accuracy of fabrication of the device, accuracy of measurement of the head, the sensitivity of the device and also how well the control section is maintained. The triangular and the rectangular weirs are commonly used flow measuring devices. They are easy to fabricate. However, use of curvilinear weirs becomes incidental in many cases. Parabolic weir has the distinction that in this weir the discharge varies with the second power of head (Igathinanathane et al. 2007). This makes the calculation work easier and this also becomes the unique feature of the parabolic shaped weir. On the other hand the discharge in case of a triangular and rectangular weir varies as the 2.5th and 1.5th power of the head respectively. Baddour (2008) has described a method to determine the head discharge equation of sharp crested weirs with openings defined

HYDRO 2014 International

by polynomials of order n. He has described three different sharp crested weirs following the fourth order polynomial geometry, but each having their own head discharge equation so that for the same head each weir gives different discharges. Curve sectioned sharp crested weirs such as circular or elliptical shaped weirs have an advantage that they do not have a horizontal edge to be leveled. More over the circular shape can easily be cut and fabricated on electrically operated lathe machines. The ellipse is a curve drawn around two axes of unequal length. A circle is a special case of an ellipse where the two axes become equal in length. In fact if the eccentricity of the ellipse tends to become equal to zero the shape tends to resemble a circle whereas as the eccentricity of the ellipse tends to value one it assumes the shape of a straight line. The analytical solution of the discharge equation involves solution of similar kind of elliptical integrals for elliptical as well as circular weirs. For an ellipse behaving as a sharp crested weir for its major axis placed horizontal, Sommerfeld et al. (1996) have proposed the following equation for theoretical discharge: Qt 

32 g ab3 / 2 21  k 2  k 4 E k   2  k 2 1  k 2 K k   15

(1)

h and is called the modulus of the integral, 2b K k  and E k  are the elliptical integral of the first and second kind respectively. The intention of this work is to study the characteristics for elliptical shaped sharp crested weirs. where k 

2. THEORITICAL BACK GROUND AND FLOW COMPUTATIONS The theoretical discharges are computed as per the analysis given below: The equation 2 for an ellipse having semi major and minor axis a and b respectively is given by: x2 y 2 (2)  1 a 2 b2 Where a and b are the major and the minor axes respectively of the ellipse. The discharge equation for a shape sharp crested weir is obtained by summing up the discharge through a small strip at a distance x from the vertex and of thickness “dx”, the width of the strip is obtained by the use of the equation 2. The area of the strip is thus obtained by multiplying the chord length by the thickness “dx” of the strip. The velocity at the elemental area is obtained by the use of the Torricelli‟s formula in equation 3:

v

2 g h  x 

(3)

The discharge through the elemental strip is given by the product of the area and the velocity at the elemental strip. The discharge for a head h is obtained by summing up the discharges dq through all such elemental strips in the range 0 to h which is given by:

MANIT Bhopal

Page 35

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Qth   2 g h  x  dA , where dA is the area of the small elemental strip. The definition sketch is shown as per Figure1. The discharge for an ellipse with its major axis vertical yields the following equation for the theoretical discharge: h h 2b (4) Qth   dq  2 g  2ax  x 2 h  x dx a 0 0







there was no leakage through the weir section. White cement and m-seal were used for sealing the joints. It was thus ensured that the flow occurred only through the weir opening. To overcome accidental errors and each discharge was measured twice so as to make sure that there could be only one discharge corresponding to any particular value of head. To overcome systematic errors head values were measured once while discharges were increasing and once when the discharges were made to decrease. Water surface profile was determined by taking readings of the free surface of water in the open channel upstream of the weir section and it was observed that the flow in the channel was a uniform flow. The minimum distance of the vertex from the channel bottom was kept equal to 0.1 m or 10 cm. 4. THEORITICAL ANALYSIS

Figure 1. Elliptical section of wier In the present study the above definite integral is solved by using the Gaussian Quadrature technique so that estimates are made till five places of decimals. The solution of the above integral is also checked by finding the elliptical integrals of the first and the second kind as per equation 1. The experimental discharges are calculated by taking actual observations in the experimental channel. The coefficient of discharge can thus be calculated. 3. EXPERIMENTAL SETUP AND METHODOLGY Experiments were performed in a masonry channel 5 m long, 0.97m wide and 0.4 m deep. Flow was made to circulate into the channel by means of a 10 horse power pump. Flow from the pump entered into the channel through a stilling basin and a baffle wall so that the water that entered the approach channel became quiescent and without any wave formation in the vicinity of the head measurement. At the other end of the channel at the test section a metal frame was installed perpendicular to longitudinal axis of the channel in which the elliptical shaped sharp crested weirs of different sizes could be mounted by nut bolting. It was insured that the weirs were truly in plumb and perpendicular to the longitudinal axis of the flume. A vernier point gauge was mounted on the channel which could take readings up to one tenth of an mm. The point gauge was placed along the centre line of the channel and at a distance of five times the maximum head measured on the upstream of the weir section. Downstream of the weir section led to a rectangular measuring tank 0.97 m long 0.47 m wide and 0.7 meter deep. The discharge from the measuring tank drained into an underground sump which was also the source of water to be supplied into the channel. The discharge measurements were done by finding the rise in water level in the measuring tank in given time. A stop watch that could measure time up to 1/100th of a second was used for measuring time. Streamlined entry and exit were ensured into the channel. It was ensured that the head measurements were not affected by any kind of local turbulences in the vicinity of the control section. It was also made sure that

HYDRO 2014 International

The Guassian Quadrature technique is used for solving the discharge equation of the ellipse as represented by equation 3 iteration accuracy till 5 places of decimal was achieved. A program was prepared in the Fortran environment to implement the scheme of the Gaussian Quadrature technique. Heads of flows are used as input to get the theoretical discharges. The programme gives the coefficient of discharge as the output. The first set of the experiments are performed for low discharges so that the resulting flows are in the laminar transient zone for the open channel so that the Reynolds Number was less than 2000. The h / P ratio is varied from in between 0.25 to 0.54. To study the variation of coefficient of discharge with head, Cd is plotted against a dimensionless parameter h / P where P is the distance of the weir vertex from the channel bottom. 5. RESULTS The merit of a sharp crested weir is its simplicity of procedures for the discharge measurement. Once the head is measured the discharge can be read out from the calibration curves. While measuring the head over the vertex of the sharp crested weir care is to be taken that the head has stabilized and it is not rising or falling when the reading is being taken. According to Falve (2003), with the change of discharges in the channel it may take many minutes for the head to stabilize in the channel. To overcome this difficulty they suggested to take the head measurement with increasing as well decreasing discharges so as to eliminate the systematic errors. Therefore keeping this in mind the head measurements for the present study are done for once while the discharges are being increased and once while the discharges are being decreased. However, a stabilizing time of two minutes was also permitted while taking reading in either direction. The semi elliptical sections chosen are installed to work as sharp crested weirs with semi major axis in each case 0.25 meter and vertical minor axis of each section as 0.26 m, 0.30 meter and 0.34 meter with corresponding internal angles as 54.94o, 61.927o and 68.431o and corresponding eccentricities 0.854, 0.800 and 0.733 respectively. The lower value of the eccentricity is an indicator of flatness of slope of the elliptical curve. As such the eccentricity of the ellipse tends to zero for the flattest ellipse when the ellipse tends to assume the shape of the circle.

MANIT Bhopal

Page 36

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Head discharge variations are plotted as shown in Figures 2 for Channel Reynolds number from 635 to 1837 which is the laminar transition zone. The computed head discharge curve lies above the experimental curve as expected. However in the given range of flow and Reynolds number it is further observed that the two curves have not remained parallel. With increasing Reynolds number the gap between the two curves of computed and experimental discharges goes on widening which means that the ratio of experimental discharge to the theoretical discharge and therefore the coefficient of discharge goes on reducing for the given range of Reynolds Number .

Figure 4. Variation of Cd with surface tension

Figure 2. Variation of Cd with Reynolds Number It is observed on comparing the experimental discharges for the three different weir sections that for lower eccentricity of ellipse that is for the highest value of minor axis of 0.34 m the discharges for the same and similar heads are higher.

Figure 5. Variation of Cd with head

Figure 3. Variation of Cd with viscosity It is concluded that for the same head and similar hydraulic conditions a wider section is capable of handling higher discharge. This is due to the fact that for same head the flow velocity remains same, but the area of cross section is more for a wider section resulting into higher flows. Due to this reason the ellipse subtending an angle of 90 degree at the vertex gives the highest discharge as compared to elliptical sections having the same major axis and lesser value of the minor axis. After experimental results, the relationship between Cd and viscosity and surface tension is presented in Figure 3 and Figure 4 respectively. Variation of Cd with head is also presented in Figure 5.

HYDRO 2014 International

6. CONCLUSIONS The properties of the elliptical section as a weir are studied. Discharge sensitivity of elliptically shaped weir section is found to be more than the rectangular section. In fact its sensitivity lies between that of a rectangular and parabolic weir section. The carrying capacity of a semi elliptical section is inscribed in a rectangle is found to be 18.28% more than that of a parabolic weir inscribed in the same rectangle and under same hydraulic conditions. The carrying capacity of elliptical section is 57.7% more than that of an inscribed triangle the carrying capacity of the section is however less and is only 78.8% of the circumscribing rectangle. The dependence of Cd on surface tension parameter and viscosity parameters is studied. Their effect on Cd is more pronounced for low depths and discharges. From the present study it is concluded and reinforced that that coefficient of discharge becomes independent of surface tension parameter at a much lower depth while viscosity parameter still continues to control Cd for higher depths.

MANIT Bhopal

Page 37

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The values of coefficient of discharges are recorded for Re less h P ratio. The rate of decrease is faster in the laminar zone for and h lesser values of .The Cd values tended to approach a value P h 0.62 for the corresponding Reynolds number of 1837 and P ratio 0.54. The Cd values are also recorded for higher Reynolds numbers

than 2000. It is concluded that the Cd value decreases with

with Re varying between 2064 and 9369 in the turbulent regime of open channel flow. For the three sections with  = 30, 45 and 60o the average value of coefficient of discharge is found to be 0.45. The average value coefficient of discharge value for  = 90o which is actually a circular shape is found to be 0.53. It is concluded that the general value of Cd = 0.6 cannot be used under all circumstances for all shapes ,but will depend upon the weir geometry, the weir dimensions in comparison to channel dimension and the upstream flow conditions. Apart from being of academic importance the knowledge of elliptically shaped weir will become handy when its use becomes incidental. REFERENCES: i. Baddour RE (2008). Head discharge equation for sharp crested polynomial weir. Journal of Irrigation and Drainage Engineering, 134(2), 260-262 ii. Falvey TH (2003). Hydraulic Design of Labyrinth Weir, ASCE Publications. iii. Igathinanathane C, Srikant K, Prakash B, Ramesh AR (2007). Development of parabolic weirs for simplified discharge measurements. Journal of Biosystem Engineering, 96(2), 111-119 iv. Sommerfeld JT, Michael P (1996). Journal of Environment Science Health, 31(4), 905-912

Turbulence Characteristics of Flow Past Submerged Vanes Sharma, H., Research scholar, Department of Civil Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India-247667. E-mail: [email protected] Ahmad, Z., Professor, Department of Civil Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India247667. E-mail: [email protected].

generates the excess turbulence in form of helical flow structure in the flow due to pressure difference between approaching flow side and downstream side of vane. Experiments were performed in a recirculating concrete flume of width 1.0 m, 0.3 m depth and of 19 m length to observe flow pattern around submerged vane rows. It was observed that in the presence of submerged vanes all the turbulence quantities were observed to increase. It was also observed that optimum amount of flow was diverted with one vane row rather than utilizing multiple vane rows. INTRODUCTION Submerged vane is basically an aerofoil structure, which generates the excess turbulence in form of helical flow structure in the flow due to pressure difference between approaching flow side and downstream side of vane (Odgaard and Spoljaric, 1986; Odgaard and Mosconi, 1987; Odgaard and Wang, 1991; Wang and Odgaard, 1993). These vanes are in general placed at certain angle with respect to the flow directions which is usually in between, 10o – 40o (Fig. 1.). Submerged vane differs from the traditional methods like groins, dikes, etc., which are usually placed normally to the flow and produce flow distribution by drag force and are not so much efficient in controlling the sediment transport. Submerged vanes utilize vorticity to minimize the drag and produce flow redistribution in the flow such that longitudinal flow is compelled to get diverted towards the transverse direction (Wang and Odgaard, 1993). Many investigators like Odgaard and Wang (1991a), Wang and Odgaard (1993), Marelius and Sinha (1998), Tan et al. (2005), Ouyang et al. (2008) have studied analytically and experimentally the flow structure of the submerged vane. This paper presents the study of flow pattern around rows of submerged vanes. A BRIEF REVIEW OF LITERATURE OF FLOW AROUND SUBMERGED VANES Odgaard and Kennedy (1983) calculated by using KuttaJoukowski theorem and verified by physical modeling the utilization of submerged vane as bend protector. Odgaard and Wang (1991a) studied the flow pattern around the submerged vane by including various factors which can possibly affect the flow pattern and developed a formula to calculate lift and drag coefficient. Wang and Odgaard (1993) critically analyzed the theory of tip vortex and utilized method of images for two vanes and for multiple vane arrays they proposed a differential equation. Marelius and Sinha (1998) observed the flow pattern around the vane for α > 30o and also obtained the optimum angle of attack. Tan et al. (2005) studied the flow pattern around the vane and optimized the vane parameter so that vane can act as sediment manager. Ouyang et al. (2008) obtained an interaction model of vane by putting up the fact that vane interaction field associated with multiple vane array is different for different vane in the system in contradiction to the theory put forward by Wang and Odgaard (1991a). Han et al. (2011) experimentally studied the effect of submerged vanes on the flow characteristics of 90 o channel bend.

ABSTRACT : Submerged vane is basically an aerofoil structure placed at certain angle with respect to the flow directions which is usually in between, 10o – 40o, which

HYDRO 2014 International

MANIT Bhopal

Page 38

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 8H and 20H from the last vane row. Velocity was measured initially for four vane rows and after measuring the flow pattern around submerged vane, a vane row was removed and process was repeated and finally the final flow pattern was measured for plane shear flow condition.

Fig. 1. Submerged vane induced transverse irculations EXPERIMENTAL ANALYSIS OF FLOW PATTERN AROUND SUBMERGED VANE Experiments were performed in Hydraulic Engineering Laboratory of Civil Engineering Department, Indian Institute of Technology, Roorkee. Experiments were performed in a recirculating concrete flume of width 1.0 m, 0.3 m depth and of 19 m length (Fig.2.) The bed slope of flume was measured to be 6.32 ×10-4. The water was supplied to the flume through an overhead tank in which the level of water was kept constant to have constant discharge for a particular opening of the valve fitted in delivery pipe of the tank. After the experimentation the used water was taken to sump from where by the centrifugal pump water again sent back to the overhead tank. Flow strengtheners and wooden wave suppressors were provide to kill the surfacial disturbances and for straightening of the flow. A tail gate was provided at the end of the flume in order to maintain the uniform flow into the flume. An orificemeter was also provided in the delivery pipeline from overhead tank for the measurement of discharge. Four rows of submerged vanes were attached to the bed so as to perform experimentations of flow pattern around submerged vanes.

RESULTS AND DISCUSSIONS From the Fig. 4, it can be seen that in the presence of vanes, flow near to the vane is highly unstable and chaotic. The turbulence is clearly having heterogeneity as going up in vertical direction from bed towards the flow surface turbulence quantities decrease usually but in the presence of submerged vanes all the turbulence quantities varied having a peak. This peak signifies the area of separation and high shear stress. It was also seen that this peak occur at z/h ≈ 0.4.

Fig.2. Line sketch of experimental setup

Fig.3. Experimental set up with submerged vanes (H = 6 cm and L = 12 cm) Vanes used in experimentations were viz. 6cm x 12cm whose lateral spacing respectively was 12.5 cm (Fig. 3.). In order to measure the velocity mini ADV was used over sections x = 3H,

HYDRO 2014 International

MANIT Bhopal

Page 39

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Fig.4. Variation of various turbulence quantities and velocity profile for x = 3H for four vane rows

Fig.5. Variation of various turbulence quantities and velocity profile for x = 3H for no vane row.

HYDRO 2014 International

MANIT Bhopal

Page 40

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

When observed this point comes out to be z = 0.83 times height of vane. According to observations of Odgaard and Wang (1991 a) and Wang and Odgaard (1993) the point of origin of vortice was 0.8 times the vane height and present observation was very near to their observation. In case of Fig. 5, it can be clearly seen that variation of all turbulence characteristics was same in all direction and was nearly overlapping each other. It signifies that turbulence in case of without vanes was homogeneous in nature. Also, the turbulence quantities varied in accordance with the observations of Nezu and Nakagawa (1993). 0.6 b)

0.6 0.5

a) 0.5

0.4

z/H

z/H

0.4

0.3

0.3

0.2

0.2

0.1

no vanes 1 vane row 2 vane rows 3 vane rows 4 vane rows

0.1

no vanes 1 vane row 2 vane rows 3 vane rows 4 vane rows

0.0 -4

-2

0

2

4

v/u*

0.0 -6

-4

-2

0

2

4

6

v/u* 0.6 0.6

c)

b)

0.5 0.5

0.4

z/H

z/H

0.4

0.3

0.3

0.2 no vanes 1 vane row 2 vane rows 3 vane rows 4 vane rows

0.2 no vane 1 vane row 2 vane rows 3 vane rows 4 vane rows

0.1

0.1

0.0 0.0

-4 -6

-4

-2

0

2

4

6

-2

0

2

4

v/u*

v/u*

Fig. 7. Variation of transverse velocity with and without vane row for a) y = 0.45 m; b) y = 0.5 m and c) y = 0.55 m for x = 20h (h = vane height).

0.6 c) 0.5

z/H

0.4

0.3

0.2 no vanes 1 vane row 2 vane rows 3 vane rows 4 vane rows

0.1

0.0 -4

-2

0

2

4

v/u*

Fig. 6. Variation of transverse velocity with and without vane rows for a) y = 0.45 m; b) y = 0.5 m and c) y = 0.55 m for x = 8h (h = vane height). It was seen from Figs. 6 and 7 that with one vane row more flow was diverted in the transverse direction as transverse velocity then two, three and four vane rows. Hence, it signifies the fact by placing one vane row optimum diversion of flow can be done while other vane rows did not produced effective diversion as was expected.

HYDRO 2014 International

CONCLUSIONS It can thus be concluded from the experimental study that in the presence of submerged vanes all the turbulence quantities were observed to increased. It was also observed in the variation of turbulence quantities a peak was observed to occur at z/h ≈ 0.4. This represented the core of vortex having maximum turbulence. Height of core of vortex was observed to be z = 0.83 times height of vane which was close to value quoted in literature. It was also observed that with one vane row more flow was diverted in the transverse direction then two, three and four vane rows. Hence, it signified the fact that by placing one vane row optimum diversion of flow can be done while other vane rows did not produced effective diversion as was expected. REFERENCES i. Marelius, F., and, Sinha, S.K. 1998. Experimental analysis of flow past submerged vanes. Journal of Hydraulic Engineering, ASCE, 124 (5), 542545. ii. Nezu, I., and Nakagawa, N. 1993. Turbulence in open channel flows. IAHR, AA Balkema, Delft, Netherlands. iii. Odgaard, A.J., and, Kennedy, J.F. 1983. River bend bank protection by submerged vanes. Journal of Hydraulic Engineering, ASCE, 109 (8), 11611173.

MANIT Bhopal

Page 41

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

iv. Odgaard, A.J., and, Spoljaric, A. 1986. Sediment control by submerged vanes. Journal of Hydraulic Engineering, ASCE, 112 (12), 11641181. v. Odgaard, A.J., and, Mosconi, C.E. 1987. Streambank protection by submerged vanes. Journal of Hydraulic Engineering, ASCE, 113 (4), 520-536. vi. Odgaard, A.J., and, Wang, Y. 1991 a. Sediment management with submerged vanes. Theory: I. Journal of Hydraulic Engineering, ASCE, 117 (3), 267-283. vii. Ouyang, H.T., Lai, J.S., Yu,H., and, Lu, C.H. 2008. Interaction between submerged vanes for sediment management. Journal of Hydraulic research, IAHR, 46 (5), 620-627. viii. Tan, S.K., Guoliang, Y., Lim, S.Y., and, Ong, M.C. 2005. Flow structure and sediment motion around submerged vanes in open channel. Journal of Waterway, Port, Coastal and Ocean Engineering, ASCE, 131 (3), 132-136. ix. Wang, Y., and, Odgaard, A.J. 1993. Flow control with vorticity., Journal of Hydraulic Research, IAHR, 31 (4), 549-562. x. Han, S.S., Biron, P.M., and Ramamurthy, A.S. 2011. Three dimensional modeling of flow in sharp bends with vanes. Journal of Hydraulic Research, IAHR, 49 (1), 64-72.

Hydraulic Design Aspects of Stilling Basin with Sloping Apron V.S. Rama Rao1 K.T.More2 Dr. 3 M.R.Bhajantri Dr. V.V.Bhosekar4 [email protected] , [email protected] [email protected] , [email protected] Central Water & Power Research Station, Khadakwasla, Pune411 024 ABSTRACT: Stilling basins are very popular type of energy dissipators provided for high head / low head spillways, weirs, culverts and channels. Energy dissipation by stilling basins is governed by various factors like intensity of discharge, head causing flow, Froude number and tail water depth. When the tail water levels are sufficient to cope up with the sequent depth of hydraulic jump, stilling basins with horizontal apron are provided. If the tail water levels are higher than the required for sequent depth, sloping aprons are provided to contain the jump within the spillway glacis to avoid encroachment of jump further upstream. The design of sloping apron involves fixing of slope of apron, calculation of length of apron and provision of appurtenances like endsill. The slope of the apron has influence on the tail water depth and thereby the length of the jump and its location on the apron. The end sill is constructed at the downstream end of the stilling basin, whether solid or dentate and has function of reducing the length of the hydraulic jump and controlling scour. It is not possible to standardize design procedures for sloping aprons as for the horizontal aprons. The slope of the apron must be determined from economic considerations and the length must be judged by the type and soundness of the riverbed downstream. In this paper various aspects relating to sloping stilling basins are discussed with reference to hydraulic model studies conducted

HYDRO 2014 International

on Garudeshwar weir in CWPRS. Numerical modelling was also carried out for the weir and the results were found in good agreement with results from physical model studies. Key Words: Horizontal apron, Sloping apron, initial depth, sequent depth, end sill, downstream apron, maximum water level, crest elevation. 1.0 INTRODUCTION Energy dissipaters for spillways are required to dissipate the excessive energy generated by impounding water when gets released down. The huge amount of potential energy is converted into kinetic energy due to steep slope of glacis of spillway. This energy may cause serious erosion which depends largely on the rate of discharge, head causing flow and credibility of the river bed material and surrounding geological area on the proximity of the dam and cause problems to the downstream of spillways and sometimes create threat to the dam complex itself. The energy of released flows can cause problems in the following ways:  Erosion of banks and spillway undermining  Sedimentation problems  submergence of downstream areas To avoid the above mentioned problems the excess energy is to be dissipated to an allowable limit. The various structures which are required for this are called energy dissipators. The design of energy dissipator plays an important role in the dam safety issue. The common types of energy dissipators are stilling basin with horizontal and sloping aprons, ski jump type buckets and solid/ slotted roller buckets. 2.0 STILLING BASINS AND SLOPING APRONS Stilling basins are the most popular type of energy dissipators provided for spillways. When the Tail Water Rating curve matches with the Jump Height Curve, Stilling Basin is the suitable form of energy dissipation arrangement. For spillways on weak rock conditions and weirs and barrages on sand or loose gravel, hydraulic jump stilling basins are recommended. Design of stilling basins involves calculation of invert level of basin, length of basin and appurtenances provided for basin. When Tail water is too high as compared to the sequent depth, the jet left at the natural ground level would continue to go as a strong current near the bed forming a drowned jump which is harmful to river bed. In such a case, a hydraulic jump type stilling basin with sloping apron should be preferred as it would allow an efficient jump to be formed at suitable level on sloping apron. Figure 1 shows a typical sloping stilling basin with endsill. 3.0 HYDRAULIC DESIGN OF SLOPING APRON Stilling basin with sloping apron can be considered for high head spillways when tail water depth is more to achieve economy. The hydraulic jump may occur in different ways on sloping apron as shown in Figure 2. Type B jump forms at toe of slope and ends on horizontal apron, type C forms on slope and ends at junction of slope and horizontal apron, and type D forms entirely on slope. The length of apron required may range from 40 – 80 % of length of jump. When good rock is available downstream,

MANIT Bhopal

Page 42

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

that rock is supposed to act as apron. Figures 3 and 4 show length of jump in terms of conjugate depth D2 and ratio of conjugate depth D´2 to D1 (IS: 4997- 1968). Extensive studies were done on sloping apron stilling basins (Hager, 1974) by Kindsvater (1944), USBR (1948), Bradley and Peterka (1957), Ariyemma (1958), Bunyan (1958), Smith (1959), Van Beesten (1962), Rajaratnam (1963), Mahmood (1964) and Mura Hari (1973). Procedure adopted for designing sloping apron is given as under (Peterka, 1984): 1. Determine an apron arrangement which will give the greatest economy for the maximum discharge condition. This is a governing factor and the only justification for using a sloping apron. 2. These stilling basins are provided for spillways/ weirs whose heads are less than 15 m and intensity of flow less than 30 m3/s/m. 3. Position the apron so that the front of the jump will form at the upstream end of the slope for the maximum discharge and tail water condition. Several trials will usually be required before the slope and location of the apron are compatible with the hydraulic requirement. It may be necessary to raise or lower the apron, or change the original slope entirely. 4. With the apron design properly for the maximum discharge condition, it should then be determined that the tail water depth and length of basin available for energy dissipation are sufficient for, say,1/4,1/2 and 3/4 capacity.

Figure 1. Typical Sloping Stilling Basin with end sill

Figure 3. Length of jump in terms of conjugate depth D2 (IS: 4997- 1968)

Figure 4. Ratio of conjugate depth D´2 to D1(IS: 4997- 1968) 4.0 HYDRAULIC MODEL STUDIES ON GARUDESHWAR WEIR Garudeshwar weir is located about 12 km downstream of Sardar Sarovar Dam in Gujarat. The reservoir created by the weir would function as the lower reservoir for reversible operation of the turbines of river bed power house of Sardar Sarovar Dam. Total length of the weir is 1137 m which includes 339 m long rockfill dam and non overflow blocks of length 189 m. The ungated overflow portion is 609 m long. It has an ogee profile with crest at El. 31.75 m. The design discharge is 62,807 m3/s and the high flood level is El. 44.65 m. The FRL and MDDL are at El. 31.5 m and El. 25.91 m respectively. The original design of weir consisted of roller bucket as an energy dissipater with a 40 m long apron downstream of bucket and since the solid roller bucket was not functioning satisfactorily for the entire range of discharges, the design was changed to 95 m long Stilling basin with horizontal apron as energy dissipator. As the horizontal stilling basin was not performing satisfactorily, it was provided with the sloping apron with dentate end sill.

Figure 2. Hydraulic jump on sloping apron and the relationship between D´2 and D1 (Peterka, 1964)

Figure 5. Location plan of proposed Garudeshwar weir.

HYDRO 2014 International

MANIT Bhopal

Page 43

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Hydraulic model studies have been considered as best tool for assessment of suitability of spillways and energy dissipators. For Garudeshwar weir project, 1:55 scale 2-D sectional model was built in a glass sided flume. 55 m length of the weir and stilling basin with sloping apron as energy dissipator were constructed in brick masonry and the surface was plastered in smooth cement and painted with enamel paint. The upstream and downstream beds were reproduced rigid at El. 12 m. Piezometers were provided along the surface of the weir with sloping apron for hydrostatic pressure measurement. Necessary arrangements were made for measurement of discharge, water levels and pressures. The accepted relationship of hydraulic similitude, based on Froudian criteria were used to express the mathematical relation between the dimension and hydraulic quantities of the model and the prototype. The general relation expressed in terms of model scale is as given in Table 1. Table 1. Model Scale Relation for Various Dimensions Dimensions Length Area Velocity Discharge Time Pressure in m of water head Manning´s ´n´

Scale Relation 1 : 55 1 : 3025 1 : 7.42 1 : 22434 1:7.42 1 : 55

Figure 6. Tail Water Rating Curve and Jump Height Curves for different aprons of Stilling Basin.

Figure 7. Pressures on profile of Sloping Stilling Basin for the discharge of 15,700 m3/s

1: 1.95

5.0 STUDIES WITH SLOPING APRON (CWPRS T.R.No. 5027, 2012) 5.1 Studies with dentated end sill The performance of 60 m long stilling basin with sloping apron with dentate endsill was observed for the entire range of discharges up to the maximum discharge of 62,807 m3/s. The hydraulic jump on sloping basin is subjected to varied tail water levels for different discharges. Studies indicated that weak jump was forming for higher discharges above 31,400 m3/s but for discharges ranging from 31,400 m3/s (50%) up to 15,700 m3/s (25%), a clearly defined hydraulic jump was forming in the stilling basin but slightly encroaching upstream on the rear slope of the weir. Tail water rating curve versus jump height curve shows that tail water levels are 0 to 5 m higher than jump heights for entire range of discharges (Figure 6). The energy dissipation seems satisfactory for the given tail water levels. For discharges below 10,000 m3/s, the front of jump shifted downwards and showed tendency of further shift for 10 % retrograded tail water levels. The studies for pressures indicated that the pressures were positive on the surface of the weir and stilling basin for the entire range of discharges. Velocities observed downstream of sloping apron are of the order of 1.1 m/s. Figures 7 and 8 show pressure and water surface profiles on Sloping Stilling Basin for the discharge of 15,700 m3/s with dentated endsill.

HYDRO 2014 International

Figure 8. Water surface profiles on Sloping Stilling Basin for the discharge of 15,700 m3/s

Photo 1. Performance of stilling basin with horizontal apron with dentated endsill for discharge of 15,700 m3/s 5.2 Studies with Solid end sill The end sill, either dentated or solid, located at the downstream end of the stilling basin reduces the length of the stilling basin by creating additional tail water depth. It also deflects the flow along the stilling basin floor upward and away from the bed of the downstream channel and protects it from scour. The end sill also serves to hold the hydraulic jump in equilibrium within the

MANIT Bhopal

Page 44

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

basin resulting in improved efficiency. To allow a shift of toe of jump further upstream for lower discharges, the existing dentated endsill was converted into solid endsill and studies were carried out. From the model studies, it was observed that the front of the jump shifted slightly towards toe with the provision of solid end sill as compared to the jump with dentated end sill, though it did not form at the toe of the weir. But for discharge of 15,700 m3/s, jump was forming exactly at the toe without showing any tendency of shifting down as shown in photo 3. Velocities observed downstream of sloping apron are of the order of 1.5 m/s and were slightly more than the one with dentated endsill as shown in Table 2.

Photo 3. Performance of Stilling Basin with solid endsill for discharge of 15,700 m3/s Table 2. Velocities observed downstream of sloping apron Type of profile

Discharge, Q (m3/s)

60 m long sloping apron with dentate endsill

15700

Maximum observed velocity d/s of end sill @ Ch. 90 m (m/s) 1.17

60 m long sloping apron with solid endsill

15700

1.51

6.0 NUMERICAL MODELLING The commercial software Flow-3D, developed by Flow Science, was used for the numerical modeling of the flow. The Flow-3D uses finite-volume method to solve the Reynolds-averaged Navier –Stokes (RANS) equations over computational domain (Amorim et al, 2004). Tracking of free surface is performed using Volume-of-Fluid method. The numerical modelling of the flow inside the stilling basin is much complex due to the high intensity of the turbulence and the recirculation that is associated with the hydraulic jump. To represent these characteristics of the flow, Re-normalized Group (RNG) turbulence model was used. During simulation, upstream boundary was set as a Volume Flow rate and downstream boundary as a Pressure Outlet. The extent of the mesh in the upstream X-direction was adjusted until any further increases had negligible effect on the discharge, while the downstream boundary was placed past the energy dissipator to cover tail water level conditions. The simulation was run for 85 seconds which was found to be enough for the hydraulic jump stabilisation. During the simulation, flow starts from the rest and is settled by water level difference between

HYDRO 2014 International

upstream and downstream. There is an initial time gap, for which the hydraulic jump, still is not stabilised and characteristics flow parameters presents a great time fluctuation. When the jump becomes stable, these values have a small fluctuation around an average value. Simulation was carried out for 15,700 m3/s (25% of design discharge). Figure 9 shows numerical simulation in Flow- 3D for Garudeshwar weir for discharge of 15,700 m3/s with solid endsill.

Figure 9: Numerical Simulation in Flow-3D for Garudeshwar weir for discharge of 15,700 m3/s. 7.0 COMPARISION OF RESULTS OF PHYSICAL AND NUMERICAL MODELS. 8.0 The results obtained from numerical simulation were compared with the results obtained from experimental (physical) model studies. 8.1 Average Pressure Pressure at pre-defined points were measured from numerical simulation at 85 seconds, corresponds to occurrence of stable hydraulic jump. Figures 10 and 11 show results from numerical simulation and comparison of results for pressures obtained from numerical simulation and experimental studies for discharge of 15,700 m3/s, respectively. The results are in general agreement at most location. 8.2 Average Water Profile Water surface profile over surface of weir measured from numerical simulation at 85 s, corresponds to occurrence of stable hydraulic jump. Figures 12 and 13 show results from numerical simulation and comparison of results for water surface elevations obtained from numerical simulation and experimental studies with a discharge of 15,700 m3/s, respectively. The results are in general agreement at most location with minor differences.

MANIT Bhopal

Page 45

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Figure 13. Comparision of Water surface profile for discharge of 15,700 m3/s.

Figure 10. Average mean Pressure from Numerical Simulation for discharge of 15,700 m3/s.

8.0 CONCLUSIONS The sloping apron stilling basin is adopted when the tail water levels are higher than the sequent depths of horizontal apron. The design involves calculation of economical slope of stilling basin suited to frequent disposable floods. Though codal provisions mentioned the applicability of these for heads less than 15 m and intensities less than 30 m3/s/m, while designing these basins for other conditions, hydraulic model studies are necessary for verifying its performance. Garudeshwar weir of Sardar Sarovar Project, Gujarat was designed with sloping apron stilling basin after testing various alternatives through hydraulic model studies. The studies indicated that the length of apron is sufficient for containing the jump in the sloping basin. By carrying out numerical modelling, water surface and pressure profile were compared with results of physical model studies and were found in good agreement. Thus, it is inferred that the numerical modelling can be used as a complementary tool to physical modelling for studying various alternatives. However, final designs needs to be studied on physical model. ACKNOWLEDGEMENT The authors are thankful to Shri S Govindan, Director CWPRS for his encouragement in writing the paper. The authors are also grateful to staff of SED Division, CWPRS for their help in preparation of this paper.

Figure 11. Comparision of average mean Pressure for discharge of 15,700 m3/s.

Nomenclature D1 = Depth of flow at the beginning of the jump D2 = Depth conjugate to D1 for horizontal apron D´2 = Depth conjugate to D1 for sloping apron hs = Height of endsill L j = Length of hydraulic jump L b = Length of basin V1 = Velocity of flow at the beginning of the jump V2 = Velocity of flow at the end of the jump θ = Angle of sloping apron with horizontal F1 = Froude Number of flow at the beginning of the jump REFERNCES

Figure 12. Water surface profile from Numerical Simulation for discharge of 15,700 m3/s.

i. Amorim, J. C., Rodrigues, R.C., Marques, M. G., (2004) ―A Numerical and Experimental Study of Hydraulic Jump Stilling Basin‖ - Advances in Hydro-science and Engineering, Volume VI. ii. CWPRS Technical Report No. 5027 of Nov 2012 ―Hydraulic model studies for Garudeshwar Weir with sloping apron of Sardar Sarovar Narmada Project, Gujarat, 1:55 Scale 2-D Sectional Model‖. iii. Hager. W.H. (1992) ―Energy Dissipators and Hydraulic Jump‖. Kluwer Academic Publishers, The Netherlands. iv. IS: 4997- 1968 ― Indian Standard Criteria for Design of Hydraulic Jump Type Stilling Basins with Horizontal and Sloping Apron‖ v. Peterka A. J. (1984) ―Hydraulic Design of Stilling Basins and Energy Dissipators‖, Engineering Monograph No. 25, United States Department of the Interior Bureau Of Reclamation, Water Resources Technical Publication, Denver, Colorado.

Hydraulic Design of Barrage in Montane Terrains HYDRO 2014 International

MANIT Bhopal

Page 46

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Rajendra Chalisgaonkar 1, Mukesh Mohan1, Manish S. Sant2 and Pratibha S. Sant2 1 Superintending Engineer, Irrigation Department, Dehradun248001, Uttarakhand. 2 Assistant Engineer, Irrigation Department, Roorkee-247667, Uttarakhand. E-mail: [email protected] ABSTRACT:The bouldery reach of river is characterized by supercritical flow for the major portion of its length till it reaches the plains where the river runs at sub-critical stage. The river bed comprises of boulders, cobbles, gravels, etc. with a mean sediment size ranging from 10cm to 30 cm or more. The approach of planning and design of diversion structures for irrigation, drinking water or power generation in upper bouldery reaches of rivers having steep gradient and deep pervious foundation are entirely different from the design principles followed for structures in mild sloping lower reaches of rivers with flat and plain terrains flowing in fine alluvial soils and as such the existing guidelines by Bureau of Indian Standards for design of weirs and barrages do not apply to the planning and design issues of structures in bouldery reaches. In this paper, authors have described in detail the hydraulic design of barrage carried out by the prevalent BIS guidelines and the formulae developed by many researchers for hydraulic design of barrage in montane regions and presented a comparison.

characterized by supercritical flow for the major portion of its length till it reaches the plains where the river runs at sub-critical stage. The river bed comprises of boulders, cobbles, gravels, etc. with a mean sediment size ranging from 10cm to 30 cm or more. Fig. 1 gives an idea of rivers flowing in bouldery reaches with steep gradient and carrying large size boulders. In fact, current IS code on „Guidelines for Hydraulic Design of Barrages and Weirs: Part – I, Alluvial reaches‟ (IS: 6966 – Part I, 1989) and other related codes by Bureau of Indian Standards (BIS) are applicable for barrages on alluvial reaches of rivers with fine and medium size sediments. The Engineers and other design consultants are still using the Guidelines available for Design of Barrages in alluvial reaches due to non-availability of sufficient literature and guidelines of Bureau of Indian Standards. However, the Indian rivers of large magnitude, flowing over gravelly and bouldery beds in the Himalayan and sub-Himalayan regions, need more accurate studies and analysis as the planning and designing of these structures are entirely different from the design principles followed for structures in mild sloping lower reaches of rivers with flat and plain terrains flowing in fine alluvial soils. The paper describes in detail the hydraulic design of barrage carried out by the prevalent BIS guidelines and the formulae developed by many researchers for hydraulic design of barrage in montane regions.

Key words: Diversion structure, Bouldery River, Supercritical flow, Sediment size, Impervious apron, Cut-off Depths 1.0 INTRODUCTION A barrage, by a definition, is a weir fitted with a gated structure to regulate the water levels in the pool behind in order to divert water through canal. The importance of weirs or barrages to divert river water through a canal system for irrigation and other useful purposes in tropical and subtropical countries needs no emphasis. Outwardly, it would appear a comparatively straightforward task to divert water from perennial rivers. By following the general guidelines, the location and alignment of barrage axis and that of the canal head works may be decided but the other details like the width of barrage and head works, levels of weir crests, length of weir floors, river training works, pond level etc. have to be finalized based on the hydraulic conditions and geologic characteristics of the river bed and banks of the site. However, it poses a considerable challenge to hydraulic engineers to devise a safe and economical way of tapping the mighty rivers of the Indian subcontinent, with their highly variable flow over the year in montane terrains. A barrage is a costly structure involving an expenditure of several hundred million rupees. Any approach to reduce the cost of a barrage satisfying the design criteria would be appreciated as an innovative step. Generally 15m to 20m high barrage type diversion structure are constructed in bouldery reaches of a river with steep gradient and narrow cross section. The bouldery reach of river is

HYDRO 2014 International

Figure 1. Typical River in Bouldery Reach 2.0 DESIGN OF BARRAGE IN MONTANE REGION From the literature survey, it has been observed by the authors that mainly the researchers have developed rational formulae for estimating the water way and scour depth in montane region. Therefore in the succeeding paragraphs, only the formulae suggested by researchers for estimating water way and scour depth have been described. 3.0 WATERWAY 3.1 Alluvial Rivers To minimize shoal formations in meandering alluvial rivers, the following looseness factor, suggested by IS 6966(Part 1):1989, shall be applied to Lacey‟s waterway for determining the primary value of the waterway:

MANIT Bhopal

Silt Factor

Looseness Factor

Less than 1

1.2 to 1

Page 47

International Journal of Engineering Research Issue Special3 1 to 1.5

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

1 to 0.6

(7), R depth of scour below the highest flood level in m; Q is high flood discharge in the river in m3/s; q is intensity of flood discharge in m3/s per m width; and f is silt factor which may be

Lacey‟s waterway is given by

P  4.83 Q

(1)

Where, Q is design flood discharge in cumec. The IS 6966(Part 1):1989 also suggests that for deciding the final waterway, the following additional considerations may also be taken into account: (a) Cost of protection works and cutoffs, (b) Repairable damages for floods of higher magnitudes, and (c) Afflux constraints as determined by model studies. 3.2 Bouldery Rivers For deciding the preliminary waterway (P) of the barrage in Bouldery River, the following formulae developed by different researchers may be used as guidance. a) Using formula developed by P.Sen(1997)

f  1.76 d50

(2) Where, q is intensity of the discharge which is given by Eq. (3)

q  6.56 D1.17 d 50 0.354

(3)

Where, D is total depth of flow (regime depth), and d50 is average diameter of the stone in the bed. b) Using formula developed by R.Garde(2000)

P  3.872Qn 0.396 d50

(a) For design discharge upto 500cumec R. D. Hey(1986)

R  0.22Q 0.37 d 0.11

(b) For design discharge above 500cumec P.Sen(1997)

g  S

gb



 1)(d50 ) S   

(5)

Where, P is waterway required, d50 is median size of bed material, Q is design flood discharge in cumec, Sgb is Specific gravity of bed material and S is average bed slope of the river at the location of the proposed structure. It should be noted that Lacey‟s equation is applicable in the alluvium reach of the river. SCOUR DEPTH 4.1 Alluvial Rivers River scour is likely to occur in erodible soils, such as clay, silt, sand and shingle. In non-cohesive soils, the depth of scour may be calculated from the Lacey‟s formula which is as follows: 1/ 3

than 1)

(10)

where in Eqs. (9) and (10), R is regime depth below the HFL in m, Q is a total discharge in the river in cumec, d is median size of bed material in mm and q is the intensity of discharge in the river in cumec/m. The Scour depths around a barrage constructed on mobile gravel or bouldery bed will vary from point to point due to various factors affecting the flow condition at each point. 4.0 EXAMPLE OF BARRAGE DESIGN

stream power which is defined by Eq. (5) as

Q R  0.473    f 

(9)

(4)

Where, Qn is Non dimensional quantity, may be called as

Qn  Q /  d50 2  

(8)

4.2 Bouldery Rivers For calculating the regime depth of flow in gravelly or bouldery rivers, different formulae have been developed. For average diameter of bed material upto 0.4m(400 mm) the following formulae may be used:

R  0.2q 0.855 d 0.3

P Q/q

as

calculated from the relationship

In order to compare the changes in the design of barrage, due to the formulae developed for montane regions, an example has been presented in the paper to illustrate the effects on the various parameters of barrage design. 138m long barrage has been designed in the montane regions using the standard guidelines available for barrage design in alluvial regions and the formulae described in the preceding paragraphs. The basic data adopted for the detailed design are shown in Table 1. 5.0 COMPARISON OF METHODS OF BARRAGE DESIGN The perusal of detailed design of various elements of barrage, carried out by Lacey‟s and P. Sen method, given in Table 2 indicates that 5.1 Detailed Design Design parameters or elements of design obtained from the formulae suggested by Lacey, Sen and Garde have described in Table 2.

(applicable when looseness factor is more (6)

or 1/ 3

 q2  R  1.35    f  than 1)

(applicable when looseness factor is less (7) where, in the Eqs. (6) and

HYDRO 2014 International

MANIT Bhopal

Page 48

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

viii

ix

5.1.4 I

Ii

iii iv

Table 2 – Summary of Design of Barrage Elements 5.1.1 i ii

5.1.2

Fixation of crest levels 909.50m

vii 911.00m viii

Water way Calculation Lace y's form ula 478.8 m

Parameters

5.1.3

i

Water way (using Eqs. (1), (2) and (4))

ii

No. of bays

iii

Length of each bay

iv

Total overall water way Provided

v

Looseness Factor

ii iii iv

V

vi

vii

P. Sen form ula 139. 9m

8

8

15.0 m 140.5 0m

15.0 m 140. 50m

0.29

1.0

ix R. Garde formula 67.7m

x

8 15.0m

xi

140.50m 2.1 xii

Calculation for Depth of Cutoffs Lacey's formula

P. Sen formula

9829cumec

9829cumec

85.47cumec/m

85.47cumec/m

12.16m

25.19m

926.50m

926.50m

Parameters I

vi

Crest Level of Undersluice bay Crest Level of the other barrage bays

Design Flood Discharge Discharge Intensity Scour depth Upstream water level corresponding a discharge of 9829 cumec Upstream cutoff level corresponding a discharge of 9829 cumec Assuming upstream cutoff level to be Downstream water level corresponding a discharge of 9829 cumec

HYDRO 2014 International

xiii

xiv 908.26m

888.72m

907.25m

889.00m

924.10m

924.10m

xv

Hence, downstream cutoff level corresponding a discharge of 9829 cumec Assuming downstream cutoff level to be Calculation for Length of floor Maximum staic Head 'H' = 929.5 -908.5 GEC= (S-1)(1-n), where 'S' is specific gravity and 'n' is porosity Safe exit gradient „GE‟ According to Bligh's Creep Theory, Total Length of floor Taking depth of downstream cutoff „d‟ to be Length of sloping glacis Length of trough If the total downstream slope floor length is 95 m, level of the floor at the d/s with a river slope of 0.0131 Assuming level of the floor at the d/s with a river slope of 0.0131 to be Length of downstream slope from 905.00 to 908.25 Taking 2m horizontal length beyond downstream slope & 1.5m length of weir crest downstream of the gate, total essential downstream length Length of intake works on the upstream side abutments Provide total length of upstream side (since the total length of upstream side comes negative using P. Sen formula hence providing minimum length 1.5 scour depth for P. Sen) Total length of floor

899.78m

873.73m

899.00m

874.00m

21.00m

21.00m

-

0.99

1 in 5

1 in 4

105.00m

84.00m

10.00m

35.00m

18.00m

18.00m

65.00m

65.00m

908.26m

908.26m

908.25m

908.25m

6.50m

6.50m

93.00m

93.00m

32.00m

32.00m

117.00m

38.00m

210.00m

131.00m

6.0 CONCLUSION

MANIT Bhopal

Page 49

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The design of barrage in montane region carried out by the prevailing Laceys technique and formulae suggested by researchers P. Sen and others has been described in detail in the paper and the results have been illustrated in the Tables 1 to 2. The comparison of results, clearly indicates that there is an improvement in looseness factor, as compared to Lacey‟s water way, in fixing the water way of the barrage. Also for a given discharge as the average size of bed material increases, the scour depth and depth of cut offs increases substantially. However, it has been observed that the length of weir floors are deccreased, when formulae developed for montane terrains by researchers are adopted. It has also come to the notice of the authors that the Bureau of Indian Standards is planning to formulate Guidelines for the Design of Barrage in hilly terrains and therefore it is also recommended that some more studies be conducted in the montane regions before finalizing the draft of the proposed Guidelines for Hydraulic Design of Barrages and Weirs”, Part 2Bouldery Reaches by the Bureau of Indian Standards, New Delhi so that the results obtained from the formulae are authenticated. 6.1 Waterway Length of waterway, L is equal to the regime perimeter, P. In boulder reaches of the river, it would be economical to reduce the waterway to about (0.6 - 0.8) times Lacey's waterway. From the calculations, it is observed that the length of waterway, according to R. Garde formula is 0.14 times the Lacey‟s formula. Moreover the length of waterway, according to P. Sen formula is 0.29 times the Lacey‟s formula which is in the acceptable range for boulder reaches. 6.2 Looseness factor The ratio of waterway actually provided to waterway computed is known as looseness factor. Generally the overall width of barrage actually provided may be more or less as has been computed theoretically. The perusal of Table 2 indicates that the looseness factor computed by Lacey, P. Sen and R. Garde formulae are 0.29, 1.0 and 2.1 respectively. 6.3 Scour Depth It is obseved from Table 2 that the scour depth computed by Lacey and P. Sen formulae are 12.16m and 25.19m respectively for the same discharge and silt factor. It indicates that the Scour depth calculated by P. Sen formula is almost two times the scour depth, what has been estimated by Lacey‟s formula and therefore the formula suggested by P. Sen has to be validated with further studies before using it. 6.4 Total Length of Floor The perusal of Table 2 indicates that the the total floor length in montane terrains shall be less as compared to the alluvial regions, if formulae suggested for montane terrains are used. The total floor length obtained from Lacey and P. Sen formulae are 210m and 131m respectively. 7.0 REFERENCES i. Garde, R.J. and RangaRaju, K.G. (2000) ―Mechanics of Sediment Transport and Alluvial Stream Problems‖ 3rd Ed. New Age Int. Pub. Pvt. Ltd., New Delhi.

HYDRO 2014 International

ii. Hey, R. D., and Thorne, C. R. (1986) ―Stable channels with mobile gravel beds.‖ J. Hydraul. Div., 112(8), 671–689. iii. Khosla, M.N., Bose, K.K.and Taylor, M.T. (1954) ―Design of Weirs on Permeable Foundation‖, Publication No.12, Central Board of Irrigation and Power, Malcha Marg, New Delhi. iv. Lacey, G. (1929) ―Stable channels in alluviums‖. Journal Institution of Engineers, Paper No. 4736, 229. v. Mazumder, S.K. (2004) ―Scour in Bouldery Bed – Proposed Formula‖, Written discussion on Paper No. 508 by R. K. Dhiman, Journal of Indian Roads Congress, Vol 65(3). vi. Mazumder, S.K. and Yashpal Kumar (2005) ―Estimation of Scour in Bridge Piers on Alluvial Non- Cohesive Soil by different methods‖, IRC Highway Research Bulletin. Oct., 2006. vii. Sen, P. (1997) ―Depth of scour in gravelly and bouldery rivers‖, Journal of the Institution of Engineers (India), Civil Engineering Division, Vol. 77, pp. 209-214. viii. (1989) ―Guidelines for Hydraulic Design of Barrages and Weirs‖, Part 1-Alluvial Reaches (First revision), IS:6966, Bureau of Indian Standards, Manak Bhawan, New Delhi. ix. (1989) ―Guidelines for Operation and Maintenance of Barrages and Weirs‖, IS:7349 (First Revision), Bureau of Indian Standards, Manak Bhawan, NewDelhi. x. (1991) ―Criteria for Investigation, planning and Layout of Barrages and Weirs‖, IS:7720, Bureau of Indian Standards, Manak Bhawan, New Delhi xi. Guidelines for Hydraulic Design of Barrages and Weirs(DRAFT)‖, Part 2-Bouldery Reaches, IS 6966: Part-2, Under formulation, Bureau of Indian Standards, Manak Bhawan, New Delhi(Unpublished).

Optimal Design of Intake Upstream of A Weir – A Case Study Kuldeep Malik1, Dr. R. G. Patil2 and M.N.Singh3 1 Research Officer, Central Water and Power Research Station, Khadakwasla, Pune 411 024, India, Email: [email protected] 2 Chief Research Officer, Central Water and Power Research Station, Khadakwasla, Pune 411 024, India, Email: [email protected] 3 Joint Director, Central Water and Power Research Station, Khadakwasla, Pune 411 024, India, Email: [email protected] ABSTRACT: Intake is a very vital component in every power project, which facilitate drawal of sufficient uninterrupted raw water from the available water body in the vicinity. Locating intake is a unique exercise for every project because the kind and nature of water body differ in individual projects. An intake for Rourkela Power Plant for drawing 0.425 m 3/s water was to be located in the backwaters of Tarkera weir across

MANIT Bhopal

Page 50

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

INTRODUCTION Tarkera weir was constructed across Brahmani river near Rourkela about 50 years back to facilitate the assured supply of raw water for Rourkela Steel Plant (RSP), Orissa. Two intakes have been constructed near the left bank just upstream of Tarkera weir. M/s. NSPCL has now proposed to construct an additional intake adjacent to existing intakes to cater raw water requirement of 0.425 m3/s needed for the expansion of Rourkela Power Plant (Fig.1). The river Sankh and Koel join at Vedvyas to form river Brahmani and the confluence is about 5.6 km upstream of Tarkera weir. Mandira dam with a storage reservoir capacity of 326 MCM supply regular water for diversion to the intakes throughout the year. The first intake built upstream of Tarkera weir is working nicely, however, the functioning of second intake is not upto the mark. The second intake has siltation problem because of limitations in the opening levels. In view of this the project authorities were apprehensive of the design of third intake and wanted to properly design this intake to avoid future complications.

HYDRO 2014 International

Koel river

Shankh river

Brahmni river

Proposed Intake Location

Tarkera Weir

Figure 1 : Index plan of Intake site The intake design is mainly dependent on the river morphology adjacent to the intake. Since the intake is to be located upstream Figure 1 : Index Plan of a weir, the reservoir is subjected to sedimentation and the river tries to change its planform continuously. This change is due to the movement and deposition of sediment with respect to the flow passing downstream of the weir. To assist in proper location of the intake, morphological studies were conducted with the help of topo-sheet of 1970, Satellite Imageries for the years1989, 2000 and 2012 (Fig.2). In addition hydrographic survey data of Brahmani river, hydraulic data and observations made during site visit were used to locate the intake.

Koel River

Keywords: Bridge; minimum water level; power plant; river morphology; satellite imageries ; weir.

Mandira Reservoir

Toposheet of 1970

Imagery of 1989 Imagery of 2000 Imagery of 2012

Brahmani River

river Brahmani 100 to 150 m upstream of existing intakes of Rourkela Steel Plant near left bank. The desk studies were conducted, in CWPRS, to locate the intake and decide various hydraulic design parameters. The location of intake was decided on the basis of morphological analysis using Topo-sheet of 1970 and satellite imageries of the years 1989, 2000 and 2012. The same was confirmed by the analysis of river cross-section data in the upstream of Tarkera weir. The G-Q data at upstream gauging site and 1 in 100 year flood of 15,700 m 3/s was used to workout expected water levels at proposed intake site using 1-D mathematical model HEC-RAS. The maximum scour level for the intake well of 8.0 m diameter was worked out and foundation level was recommended considering the grip length. To draw required quantity of water and to minimize the entry of sediment, size of the openings of the intake structure were decided by limiting drawal velocity to 0.2 m/s so as to ensure minimal disturbance in the surrounding flow field. Orientation of the openings were decided in such a manner that drawal of sediment in the intake system is minimum and maximum portion of sediment travels in the down stream direction along with flow. The crest level of the opening was decided below LWL for 90% dependability. Openings in the intake well were suggested at two levels, one to draw surface water during floods and another from the bottom layer during lean flow to minimize entry of sediment into the intake system. Formation / Pump floor level was decided considering sufficient free board above the expected 1 in 100 year flood level. Various intricacies involved in locating an Intake well upstream of a weir and its design are discussed in the paper.

South Eastern Railway Bridge Panposh

Figure 2 : Brahmani river courses for past years

STUDY OF TOPO-SHEETS AND SATELLITE IMAGERIES The toposheet of the year 1970 (73 B ), showing Brahmani river from the confluence of Sankh and Koel rivers to the upstream of proposed Intake location, was compared with satellite imageries for years 1989 (IRS 1A), 2000 (IRS 1C) and 2012 (IRS P6) to study the changes in the deep channel courses of river Brahmni in the vicinity of proposed Intake site upstream of Tarkera weir.

MANIT Bhopal

Page 51

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 locate Intake about 70 to 80 m upstream of existing Intake.

Brahmani River

Figure 2 shows comparison of the river reach near proposed Intake location as well as in its upstream and downstream during years 1970, 1989, 2000 and 2012. It could be seen from the toposheet of year 1970 and satellite images of later period that there is very minimal change in the course of Brahmni river from its origin i.e. confluence of Sankh and Koel rivers to the Tarkera weir (near proposed Intake location). Although several changes have been observed in the past images in upstream reach of both rivers before the confluence, the river channel is quite stable at the proposed intake site. In the upstream of Tarkera weir deep channel portion is well spread from left bank to right bank, there are some rock exposures in the centre of channel also. In the reach under consideration, the deep channel is along left bank for more than last 40 years. The left bank upstream of Tarkera weir is on outer curve, therefore, deep channel has been following it. There were several rock exposures near right bank about 2 km upstream of Tarkera weir acting as a nodal point, it deflects the river course towards left bank. Afterwards river follows concave path and flows in wider area, one channel follows left bank and another along the right side upto Tarkera weir. Siltation between the channels is also noticed in an area of about 600 m long and 200 m wide, about 150 m upstream of existing intakes. A close view of satellite images ( Figure 3) shows presence of deep channel upstream of Tarkera weir well spread over the width of river.

Toposheet of 1970

Imagery of 1989 Imagery of 2000 Imagery of 2012

Fig. 3 : A close view of Satellite images for past 40 years STUDY OF RIVER CROSS-SECTION DATA The cross-section data was utilized to review and finalize the location of proposed Intake, considering location of deep channel, river bed levels and bank slope at different locations etc. From the cross-sections upstream of Tarkera weir, it was observed that deepest bed level near the left bank upstream of existing intake varied from RL 190 m to 190.5 m at a distance of 80 m to 100 m from left bank (Fig. 4) and the deep channel is about 60-70 m wide. Whereas, further upstream, river is showing tendency of shoal formation. Deep bed levels are more than RL 191.3 m and width of deep channel near the left bank is also very less. Therefore, it is considered more appropriate to

HYDRO 2014 International

Fig. 4 : River cross-sections 180 and 235 m upstream of Nalla confluence EXAMINATION OF GROUND REALITY To get familiarize with the site conditions or to know the ground truths, it is also necessary for the designers to carry out site inspection before finalizing the design. With this view site inspection was also carried out. The Brahmani river reach from confluence of Sankh and Koel rivers i.e. 5.8 km upstream of Tarkera weir (Photo 1) to 600 m downstream of proposed intake location was inspected along both of the banks of river. It was noticed that deep channel of river was along left bank in most of the portion. Within the reach under study, the river flow is between well defined & firm high banks. There existed solid rock exposures along river bed at number of places including vicinity of the proposed Intake location. It was observed that a very deep pool of water was present from Tarkera weir to about 500m upstream and deep channel was towards left side of the river (Photo 2). Two Intakes were already constructed by RSP just upstream of Tarkera weir to fulfill its requirement (Photo 1). Out of these two Intakes, the old one had multiple level openings and is working satisfactorily. Whereas, the new intake was provided with only one lower level opening. Therefore, it was facing severe siltation problem during monsoon.

MANIT Bhopal

Page 52

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 GQ PANPOSN 208

206

WATER LEVEL IN M

204

202

200

198

196 0

2000

4000

6000

8000

10000

12000

14000

DISCHRGE

Figure 5 : Gauge – Discharge relation at Panposh gauging site

FINALISING DIFFERENT WATER LEVELS Daily discharge and corresponding water-level data from June 1996 to May 2010 at Panposh gauging station about 4km upstream of Tarkera weir (Fig. 5) and daily discharge and corresponding water-level data from June 1972 to June 1996 at Bolani gauging station about 40 km downstream of Tarkera weir was utilized to decide minimum and maximum expected water levels at the Intake and thereby to decide various levels of opening and pump floor level. Statistical analysis of discharge data by gumble extreme value distribution for minimum yearly flow was used to ensure availability of required discharge in the river. It was informed by the project authority that the total water requirement for the project would be about 0.425 m3/s i.e. 15 cfs and sufficient water was available in the pool behind Tarkera weir due to regular releases from the Mandira dam, about 22 km upstream of Tarkera weir. It was informed by project authority that Rourkela Steel Plant has assured water drawal from the pondage created by Tarkera weir. The maximum flood discharge and corresponding water levels in Brahmni river were available at Panposh gauging site about 4 km upstream of Tarkera weir, which are used to workout scour and the foundation level of Intake.

HYDRO 2014 International

From the Gumble extreme value analysis of the gauge-discharge data of Panposh gauging site for 26 years, it was revealed that for 50 years frequency, maximum discharge would be 14,138 m3/s and minimum discharge would be 7.8 m3/s. The maximum discharge with hundred year frequency was found to be 15,700 m3/s, which is considered for design of foundation level of Intake structure. The project authority informed that they had never faced shortage of water supply at Tarkera Pump house for last 40 years due to regular releases from Mandira dam in the upstream. The Mandira dam having storage capacity of 326 million cubic meter was solely constructed for RSP and the releases from the dam are governed by the requirement at Tarkera weir. The requirement of water for NSPCL intake is only 0.425m3/s, for which availability is ensured on the basis of above data. For deciding the sill level of the opening of the Intake well, a realistic assessment of minimum water level is necessary. The sill level of the lowest opening should therefore be such that it is sufficiently below the lowest minimum water level satisfying the criteria of submergence. Generally the concentration of sediment near the bed is more. For minimizing the sediment entry into the Intake, the sill level of the opening should be provided with openings at different levels depending upon variation in water level with the arrangement to close bottom openings at the time of high flood. Also the area of Intake openings should be such that at minimum water level the velocities at opening / entry should preferably be below the standard drawl velocity of 0.2 m/s for drawl of required discharge with least disturbance to the surrounding area. After study of the hydraulic data and results of 1-D mathematical model studies provision of lowest level opening in the Intake has been considered at about 3 m above the river bed level in the vicinity of Intake i.e. at RL 193.0 m. As per standard drawl velocity of 0.2 m/s, one opening of size 2.2 m wide x 1.0 m high would be required at each of the levels at RL 193.0 m and RL 199.0 m as shown in Fig. 6. During high flood period water should be drawn from gates at higher level and lower level gates should be kept closed, otherwise high silt concentration bed load may enter the Intake system and clog the pump-sump. During the lean period flow with low sediment load, water can be drawn from low level openings. For the proper gate operation, openings at different level should be staggered. Intake

MANIT Bhopal

Page 53

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

openings should neither be provided facing upstream nor facing downstream, but these should be provided at the sides making an angle of 30 to 400 to the flow direction.

for a discharge of 14000 m3/s, water level and velocity were RL 206.32 m and 2.40 m/s respectively. Considering the water level for flood of 17,500 m3/s, the pump floor level is provided above RL 209.04 m taking into account free-board of 2.0 m. Table-1 Normal Depth (0.000224) Chainage Bed Level (m) (m) 6762.38 5774.49 4812.92 4678.78 4523.36 4418.34 4315.33 4212.22 4089.69 3986.56 3890.89 3793.38 3704.22 3609.84 3506.16 3368.44 3268.76 3163.12 3062.24 2984.19 2902.72 2805.54 2696.16 2507.88 2452.09 2255.78 2058.92 1907.94 1805.92 1639.82 1534.96 1432.23 1304.86 1204.36 1075.39 973.22 0.00

PREDICTION OF FLOW PARAMETERS AND HYDRAULIC DESIGN The one dimensional mathematical model HEC-RAS was used to predict water levels and velocities at the proposed Intake site. The Water level and corresponding discharge data available at Panposh gauging site of CWC about 4 km upstream of proposed Intake site was utilised for model calibration. For different discharges, the flow simulations were carried out by providing normal depth condition at the downstream boundary, for which the bed slope of river was taken as 1 in 4464 (as per available survey drawings) and about 1 in 2460 m in downstream of Tarkera weir. Discharges were used as the upstream boundary, and n value was taken as 0.04. The n value was decided considering lot of rock exposures in bed in this reach. It was seen from the extrapolated gauge – discharge data at Panposh that the water level was about RL 207.2 m for the discharge of 14,000 m3/s and matches well with the water level obtained by Mathematical model for this discharge (RL 207.49 m). Similarly water level at Panposh for discharge of 12,000 m3/s was RL 206.60 m from the G-Q curve and RL 206.50 m from the mathematical model, which shows a very good conformity. The high flood of 15,700 m3/s was also simulated by providing normal depth as the downstream boundary condition and discharge at the upstream boundary. The Table-1 shows the water levels and velocities worked out with HEC-RAS at different locations. From this table, it was seen that at proposed Intake site, water levels and velocities for flood of 15,700 m3/s were RL 207.04 m (Fig.7) and 2.55 m/s respectively, whereas

HYDRO 2014 International

194.21 193.47 190.53 190.13 189.95 188.91 189.12 189.17 190.24 187.60 186.77 185.10 188.46 187.70 189.44 190.02 189.89 189.98 189.99 189.61 189.48 190.24 190.22 190.23 190.48 190.07 191.26 190.41 190.95 190.91 190.34 190.63 190.80 190.65 190.44 190.05 190.18

Q=14000m 3 /s

Q=15700m 3 /s

WL( m)

Vel (m/s)

WL( m)

Vel (m/s)

207.49 207.36 206.98 206.97 206.94 206.88 206.82 206.76 206.66 206.61 206.59 206.57 206.57 206.63 206.59 206.59 206.57 206.55 206.50 206.48 206.44 206.43 206.40 206.32 206.25 206.23 206.16 206.18 206.15 206.09 206.05 206.04 206.00 205.99 205.96 205.92 205.75

2.55 1.97 2.63 2.53 2.45 2.56 2.68 2.77 2.95 3.00 2.91 2.85 2.70 2.21 2.29 2.08 2.11 2.11 2.19 2.23 2.28 2.22 2.28 2.40 2.58 2.41 2.49 2.14 2.15 2.26 2.33 2.23 2.25 2.18 2.20 2.26 1.95

208.28 208.16 207.75 207.73 207.70 207.64 207.57 207.51 207.40 207.34 207.32 207.30 207.31 207.38 207.33 207.34 207.32 207.30 207.25 207.21 207.18 207.17 207.13 207.04 206.97 206.95 206.88 206.91 206.88 206.80 206.76 206.75 206.72 206.71 206.67 206.62 206.46

2.68 2.07 2.79 2.68 2.60 2.71 2.84 2.94 3.12 3.18 3.09 3.03 2.85 2.34 2.42 2.19 2.21 2.21 2.31 2.36 2.41 2.35 2.41 2.55 2.74 2.57 2.63 2.25 2.25 2.40 2.48 2.36 2.39 2.29 2.31 2.40 2.04

Remark Panposh Gauging Station

cs 0 @ 236m U/S of Tarkera weir cs 27 @ 180m U/S of Tarkera weir cs 28 @ 16m D/S of Tarkera weir

Fig. 7 : Water Surface Profiles along Brahmani River The general scour was worked out considering maximum discharge of 15,700 m3/s and silt factor of 0.9799 for D50 = 0.31 mm of bed material (sand). Considering 450 m river width during high flood stage (as per the cross-section data), average discharge intensity would be 34.88 m3/s/m and increasing it by 40% for flow concentration, the maximum intensity of discharge has been considered as 48.84 m3/s/m. Taking into account local scour for 8.0 m diameter of Intake well, the maximum Scour levels were worked out on the basis of criterion laid down by the various investigators like Sir Claude Inglis, Dr. H.W. Shen etc. Considering the HFL of RL 207.0 m with the two different criterion, the maximum Scour levels were at RL 170.96 m and RL 183.88 m respectively. The foundation level is to be decided

MANIT Bhopal

Page 54

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

considering sufficient grip length below this level. Foundation level may however be restricted at higher level in case good quality rock is encountered above this level. Table-2 Maximum Scour level at Intake considering different approaches Sl.No.

1

Scientific Approac h Inglis

General Scour

q DL  1.34  f

2

  

Well Diamete r

Maximu m Scour level

8m

HFL- 2 DL = 170.96 m

8m

HFL- D – 1.4b = 183.88 m

1 3

= 18.02 m Water depth available at HFL=17m 1

2

Shen

 Q 3 D  0.473  = f 11.92 m Water depth available at HFL=17m

1

Research scholar, Department of Civil Engineering, Visvesvaraya National Institute of Technology, Nagpur, 440 010, India, Email: [email protected] 2 Assistant Professor, Department of Civil Engineering, Visvesvaraya National Institute of Technology, Nagpur, 440 010, India, Email: [email protected] ABSRACT: Dams are critical flood control devices and a major source of electric power, irrigation etc. An effort has been made in this paper to optimized the parameter of dam as even a small variation in the length or width of the dam can overall reduce the tremendous cost of the structure. An excel sheet has been prepared for this purpose with procedure followed by Indian Standard Code IS 6512 Criteria for the design of the gravity dam in which the parameter of the dam like length; width has been change to get the optimized parameter with the permissible stresses within the safe limit as per the standards of code for the hydraulic structure. Keywords: Gravity Dam, Design, Optimization parameter, Critical values, Safety limits criteria, stresses. 1. INTRODUCTION:

CONCLUSIONS Morphological and one dimensional mathematical model studies were carried out for deciding intake location in Brahmani river. The analysis of Topo sheet, satellite imageries and cross-sections of river Brahmni revealed that the course of river Brahmni is stable at proposed Intake location for more than 60 years. Hence, the proposed location of intake well about 70 m upstream of existing RSP intake and at 80 m from left bank in deep channel of Brahmni river was hydraulically satisfactory. The founding level of RL 170.96 m for the 8 m outer diameter Intake well was considered necessary from maximum scour depth analysis and adequate provision of grip length. Intake shall be provided with one opening of size 2.2 m X 1.0 m at each of levels at RL 193.0 m and 199.0 m and could be operated effectively during monsoon period to minimise the sediment entry into the intake well. The formation level / pump floor level could be kept at least 2.0 m above the HFL i.e. at RL 209.0 m at Intake location. The construction / sinking of intake well is to be undertaken in such a manner that the river flow conditions are least disturbed and cofferdams / sheet piles etc. provided during sinking should be removed as early as possible before the monsoon flood. ACKNOWLEDGEMENT Authors express deep sense of gratitude to Shri S. Govindan, Director, CWPRS for constant encouragement and valuable suggestions during preparation of papers and consent to publish this paper. The co-operation extended by all the CWPRS staff members in conducting studies is great- fully acknowledged. REFERENCES i.

CWPRS Technical report No. 5095, August 2013, ―Water Availability and Intake Studies for Expansion of Rourkela Power Plant of NSPCL, Odisha‖.

Study of Effect on the Stresses & Safety of Gravity Dam with Changes in Width Parameter B.S. Ruprai1

A.D Vasudeo2

HYDRO 2014 International

Hydraulic Structures are important components of Water Resources Engineering systems. Hydraulic structures such as dam‟s, weirs, spillways, stilling basins, energy dissipaters etc constitute major components of water resources projects. These are the main components of the system and the primary focus of analysis. Conventionally these structures are designed using standard methods and codes. The design methods adopted are also well established. But still it has been documented by many of the researchers that the structures do not perform well during the design life. It has also been observed in standard literature that these structures fail without prior warning which leads to catastrophic events. The hydraulic and structural analysis and methods adopted in designing of these structures are very complex. Even a small saving in the height or width of the dam without affecting the safety of the structure can give a lot of saving to the structure. The present study is aimed at proposing a research methodology for the design of big Water Resources Engineering systems. In India specific design codes are available which document step wise procedure for the design of Dams, Spillways, Conveyance channel etc. However these components of the water resources systems are treated in isolation. An algorithm is prepared to optimize the parameter of the design of the gravity dam in the present case. The design procedure is adopted by Indian Standard IS-6512:1984, “Criteria for the design of solid gravity dam”. To make the optimization procedure more understandable, a Microsoft Excel Sheet program is prepared to analyze the effects of varying dimensions and the factors on which the design is dependent. The sheet provides a good tool to check the permissible stresses and stability of the dam against sliding and overturning and safety within the permissible limits prescribed in the IS Code. 2. MATERIAL AND METHOD:

MANIT Bhopal

Page 55

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

By varying the width of dam on both the upstream side and the downstream side the stresses are studied for all the different cases of the dam which are discussed in brief in result and discussion. The Design constant taken is the height of the dam which will depend on water level and free board. Design variables are the sloping projection of height on the upstream side of the dam considered as X1, width of the dam on the upstream side of the dam is taken as X2 and width of the dam on the downstream side of the dam is considered as X3. A typical; Diagram Showing these parameters is in Figure No : 1.

Figure No. 1 : Typical diagram showing the parameters of dam

By observing the Figure No. 1, it is evident that the upstream and downstream slopes have a great impact on the values of X2 and X3 which in turn will govern the total base with of the dam. Whereas X1 remains unaffected as it is the Height which already is assumed to be constant. By varying the slope the decrease in any of the above parameter can directly change the dimensions of the dam and in either case reduce or increase the total size and will affect the stresses and stability. The above equation for design variables can be mathematically written as: X = f [x1, x2, x3]T (1) As for the geometric constraints if we consider the Y axis of the dam, the design variables X1 which limit from the geometry of the gravity dam can be studied from the figure no. 1 with the minimum level of the dam ie origin to the maximum level of the dam ie total height „H‟ of the dam and can be mathematically written in the form of equation as given below: (2) Similarly, the second design variable if we follow the X axis of the dam which will be the upstream side slope of the gravity dam is kept to the steeper limiting angle because in the angle is reduced the width will increase on the upstream side of the dam which is not advise due to less contribution to the safety and stability of the dam and also create the hindrance of the storage capacity of the dam. Hence the slope is steeper from the above point of the view and its limit can expressed in the mathematical from as given below: (3) Regarding the third design variable which is also along the Xaxis of the Figure No. 1 that will be the downstream side slope of the gravity dam and it will depend on the engineers decision whether to start the slope from the top width of the dam or lower

HYDRO 2014 International

than that. In our case the slope is not directly started from the top width from optimization point of the view and hence to reduce the amount of concrete the slope is kept less as that of the upstream height and the downstream height is kept H c as shown in the Figure No. 1 from the economy point of the view. Mathematically the equation can be written as given below: 0.6Hc2 – X3 ≤ 0 &X3 – 0.8Hc2 ≤ 0 Or

(4)

Thus by varying the above values the optimized width is obtained as in this paper a parameter on the downstream side of the width is only reduced as there is significant saving in the concrete and their by directly affecting the cost of the dam is studied which are discussed in the results. As there is not considerable saving in the dam if the parameter of X 2 is reduced because already taken very steeper as they play less importance in the stability of the dam as majority of the dam is have higher width on the downstream side only. For case 1, the dam is checked for the reservoir empty conditions in which the eccentricity is less than < b/6 means no tension will be developed and vertical stresses are checked at toe and heel and are within the permissible limits. Also the stability is checked by the formula as stated in Indian Standard Code IS 6512:1984 as under: ( w  u ) tan  CA  F Fo F P

(5)

F=Factor of safety, w=total mass of the dam, u=total uplift force, tan  = coefficient of internal friction of the material C=cohesion of the material at the plane considered A= area under consideration for cohesion F  =partial factor of safety in respect of friction, Fo=partial factor of safety in respect of cohesion, and P= total horizontal force Also the stability is checked for the overturning moment and given by the formula as under: Factor of safety against overturning = Resisting Moment/Overturning Moment and should be less than the IS code permissible limit. Similarly the stresses and stability are checked considering the reservoir full condition, considering uplift for case 2, reservoir full condition, considering no uplift for case 3 and reservoir full condition with drains chocked for case 4 which are discussed in details in result and conclusion by adopting the above algorithm for the programming. 3. RESULT AND DISCUSSION: Form the standard literature a generalized dam section is optimized by changing the width of the dam; stresses and stability are checked to satisfy the Indian Standard Code. The result and discussion are given below and by reviewing the graphs, it can be studied that there is very small change in the stresses as the width is reduced by 0.1m from 51m to 50m after

MANIT Bhopal

Page 56

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

which their the dam is not safe in sliding criteria. The detailed discussion of the result is as under.

Figure No. 4: Variation of Parameter of dam & Stresses with change in the width of toe for reservoir full condition with no uplift

Figure No. 2: Variation of Parameter of dam & Stresses with change in the width of toe for reservoir Empty Case Consider Case 1 for reservoir empty condition in which with the variation in parameter of dam & stresses are studied with change in the width of toe and as from the above Figure No. 2, the various parameter of dam like eccentricity, various stresses are checked for reservoir empty condition and the width of the dam at toe is reduced from 51m to 50m with height constant and the stresses are safe for this condition, but if we reduce further the width the stresses does not remain safe as per the Indian Standards. The factor of Safety against Sliding and Overturning satisfy the safety criteria for this case.

Consider case 3 of dam for reservoir full condition with no uplift in which with the variation in parameter of dam & stresses are studied with change in the width of toe and as from the above Figure 4, the various parameter of dam like eccentricity, various stresses are checked for reservoir full condition with uplift case and the width of the dam at toe is reduced from 51m to 50m with height constant and the stresses are safe for this condition, but if we reduce further the width the stresses does not remain safe as per the Indian Standards. The factor of Safety against Sliding and Overturning satisfy the safety criteria for this case.

Figure No. 5: Variation of Parameter of dam & Stresses with change in the width of toe for reservoir full condition drains chocked

Figure No. 3: Variation of Parameter of dam & Stresses with change in the width of toe for reservoir full condition with uplift Consider case 2 of dam for reservoir full condition with uplift in which with the variation in parameter of dam & stresses are studied with change in the width of toe and as from the above Figure 3, the various parameter of dam like eccentricity, various stresses are checked for reservoir full condition with uplift case and the width of the dam at toe is reduced from 51m to 50m with height constant and the stresses are safe for this condition, but if we reduce further the width the stresses does not remain safe as per the Indian Standards. The factor of Safety against Sliding and Overturning satisfy the safety criteria for this case.

Consider case 4 of dam for reservoir full condition with no uplift in which with the variation in parameter of dam & stresses are studied with change in the width of toe and as from the above Figure 4, the various parameter of dam like eccentricity, various stresses are checked for reservoir full condition with uplift case and the width of the dam at toe is reduced from 51m to 50m with height constant and the stresses are safe for this condition, but if we reduce further the width the stresses does not remain safe as per the Indian Standards. The factor of Safety against Sliding is the only component which is not safe while the overturning criteria satisfy safety for this case. Thus by providing Sliding key of small size the sliding criteria can also be satisfied. 4. CONCLUSION: The economy can be further achieved by reducing the width of the dam as the stresses are within the safety limit. Just by

HYDRO 2014 International

MANIT Bhopal

Page 57

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

decreasing the width of the dam by just one meter and satisfying the stress and stability a considerable saving in the cost is achieved. As the constraints of the width of the dam which cannot be reduced further due to sliding of the gravity dam is not within the permissible limit, but if we consider the structural aspect by provision of shear key the width of the dam can be further reduced to get more economy. As our limitation of the research is to have normal gravity dam without shear key hence the economy of the width can be optimized upto certain limit only. 5. APPENDIX I Consider the example in which the dam is design for the below mentioned case which is safe in all the design aspect as taken from standard literature, but if applying the algorithm stated above the width of the heel is reduced to 60m without affecting the safety their by achieving considerable saving in the concrete and hence the overall economy. Total Width of the dam = 61m Width of the heel = 51m Width of the toe = 10m Height of the dam = 65m 6. REFRENCES: i. Cohn, M. Z. and Dinovitzer, A. S., (1994). Application of structural optimization, Journal of Structural Engineering, ASCE, 120(2): 617–650. ii. E.J. Haug and J.S Arora, 1979. Applied optimal design, WileyInterscience, New York. iii. F. González-Vidosa, V. Yepes, J. Alcalá, M. Carrera, C. Perea and I. Payá-Zaforteza., (2000). Optimization of Reinforced Concrete Structures by Simulated Annealing, School of Civil Engineering,Universidad Politécnica Valencia, Spain. iv. IS 6512: Indian Standard Code of practice for design of gravity dam, 2010. v. Kirsch, U., (1997), How to optimize prestressed concrete beams, Guide to structural optimization. Edited by J.S. Arora. ASCE Manuals and Reports on Engineering Practice No. 90, American Society of Civil Engineers, New York. pp. 75–92. vi. Lazan, B. J., (1959). Energy dissipation mechanisms in structures with particular reference tomaterial damping, in Structural Dynamics, edited by J. E. Ruzcka, ASME Annual Meeting, Atlantic City, N. J. vii. S.S. Rao., 1977. Optimization theory and applications (Second Edition), 1977 Wilsey Eastern Limited, New Delhi. viii. U. Kirsch, 1981. Optimum Structural Design, McGraw Hill, New York. ix. Zienkiewicz, O. C. and Taylor, R. L., (1991). The Finite Element Method, McGraw-Hill, London, Fourth edition.

Assessment of environmentally stressed areas for soil conservation measures using usped model. Bikram Prasad1, R K Jaiswal2 and Dr H.L Tiwari3 1. Ph.D Scholar MANIT, Bhopal 2 Scientist, National Instiute of Hydrology Bhopal 3 Assistant Professor, MANIT Bhopal. Email: [email protected]

HYDRO 2014 International

ABSTRACT: A balanced ecosystem consisting of soil, water, and vegetation is essential for the Survival and welfare of human. However, over-exploitation of natural resources created disturbances in ecosystems and induces natural hazards. Erosion and Sedimentation are major issues in disrupted ecosystems. Soil erosion is a major environmental and agricultural problem worldwide. The loss of soil from farmland may be reflected in reduced crop production potential, lower surface water quality and damaged drainage networks. We have studied the environmentally stressed area in a catchment using USPED model. In this attempt my study area is The Kodar reservoir, constructed across river Kodar, a tributary of river Mahanadi. The dam is constructed on Raipur – Sambalpur national highway at a distance of 65 km from Raipur near village Kowajhar in Mahasamund district. We studied the soil stresses area of the Kodar reservoir using USPED model. This model is built on the backbone of the Universal Soil Loss Equation (USLE) and the Revised Universal Soil Loss Equation (RUSLE) models.. It depends on Rainfall erosivity factor, Soil erodibility factor, Topographic index, Cover and management factor and Support practice factor. It predicts the spatial distribution of erosion and deposition rates for a steady state overland flow associated with a given rainfall input. We have generated the thematic layers in GIS for development of USPED modelBy using the method we have given the priorities and divided subwatersheds as very high, high, moderate, low and very low priority. We have concluded that out of 67 sub-watershed 8 sub watershed comes under very high, 2 under high, 3 under moderate and rest under low and very low priority. Keywords: USPED, Watershed, soil erodibility. 1. INTRODUCTION Sediments deposited in the reservoir can be transported into the headrace tunnel and can lead to the wearing of mechanical parts of the Power station units such as buckets and the needle valves. The silting of reservoir can reduce the storage capacity of reservoir and high level of sediment deposited in the dam can also raise concern for the stability of the dam. Soil Erosion and sedimentation are the major environmental and agricultural problem worldwide. A balanced ecosystem consisting of soil, water and vegetation is necessary for the survival and fortunes of human being. Nearly 12×106 ha of available land are destroyed annually and to adequately feed people a diverse diet about 0.5 ha of arable land per capita is needed but only 0.27 ha per capita is available. The world population is increasing and there is continuously degradation of land by erosion resulting in food shortages and malnutrition. However, over-exploitation of natural resources created disturbances in ecosystems and induces natural hazards. Although the erosion has occurred throughout the history of agriculture it has intensified in the recent years. Hence in this study we will identify the erosion affected area and will conclude some preventive measures to minimize the soil loss. 1.1 Soil erosion by water Soil erosion is a naturally occurring process and is the wearing of a field's top soil by the natural physical forces of water and

MANIT Bhopal

Page 58

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

wind or through forces associated with farming activities such as tillage. Soil erosion is a slow process that continues relatively unnoticed, or it may occur at an alarming rate causing serious loss of topsoil. The loss of soil from agricultural land can lead to reduction in crop production potential, lower surface water quality and damaged drainage networks. It depends upon various factors such as rainfall erosivity factor, soil erodibility topographic factor vegetation and tillage practices 2. STUDY AREA The Kodar reservoir which is constructed on river Kodar, a tributary of river Mahanadi has been selected for the systematic and scientific study of reservoir sedimentation, sediment yield from catchment areas and prioritization of catchment for soil conservation measures. 3. METHODOLOGY 3.1. USPED This model is developed on the backbone of the Universal Soil Loss Equation (USLE) and the Revised Universal Soil Loss Equation (RUSLE) models. The USPED model considers divergence and convergence of slope by modelling, in a geographic information system environment, the entire upslope area that contributes to the overland flow of water across every point in the landscape. The model more fully accounts for topographic complexity by considering both in the downhill direction and the perpendicular to the downhill direction. It computes both soil erosion and sediment deposition as the change in sediment transport capacity in the direction of flow. This paper attempts to identify the spatial patterns of soil erosion within the catchment area on river Kodar, a tributary of river Mahanadi in Raipur. Maps of erosion and deposition were derived for catchment area of river Kodar, a tributary of river Mahanadi and its individual sub-basins by implementing the USPED model. The USPED model employs a stream powerbased sediment transport model with an expression of mass conservation to simulate soil erosion and deposition. The model departs from the RUSLE annual average soil loss equation expressed by E (tons/acre/year). (1) Where R represents the rainfall erosivity index, K the soil erodibility factor, LS the slope length and steepness, C the land cover management factor, and P represents the support practices factor. The USPED model assumes that sediment transport rates are determined by the erosional strength of flowing water, and never limited by the supply of transportable soil particles. Thus it is assumed that the sediment transport rate (capacity) is given by: (2) where b represents the local surface slope (degrees), m and n are constants depending on the type of flow and soil properties, where the constants m and n have the values 1.6 and 1.3 respectively for prevailing rill erosion and 1 for prevailing sheet erosion. The results of the USPED model represent relative magnitudes of the soil erosion and deposition rates rather specific soil loss values traditionally expressed in tons/acre/year. The net rate of soil erosion or deposition (ED) is given by the two-dimensional (horizontal plane) divergence of the sediment flux that expresses mass conservation:

HYDRO 2014 International

(3) Where, a represents the aspect of the terrain (the direction of maximum hill slope gradient in the horizontal plane in degrees). 3.1.1 Rainfall erosivity factor (R) The R factor is calculated by rainfall and the energy imparted to the land surface by the impact of rain drop. Rainfall erosion index implies a numerical evaluation of a rainstorm which describes its capacity to erode soil from an unprotected field. It is a function of intensity and duration of rainfall and mass, diameter, and velocity of the rain drop. Annual R factor, (4) Ra  79  0.363 * PA where, PA is the annual rainfall in mm and Ra are annual Rfactor in MJ mmha-1yr-1. The theissen map (Fig. 1) of Kodar catchment has been prepared using the ILWIS 3.0 software and it observed that Kodar catchment is affected by Kodar, Bagbahara and Bartunga R.G. stations. The weights and R-factor for different RG stations have been presented in The value of annual and seasonal R-factor for Kodar reservoir catchment has been obtained as 429.39 MJmmha-1hr-1 and 402.94 MJmmha-1hr1 respectively. The weights of Kodar, Bagbahara and Bartunga RG stations have been computed as 0.50, 0.48 and 0.02 respectively. The rainfall in the study area concentrated mainly in the month of July, August and September. By using the operation attribute map input as thessien polygon and table as R value output Rmap is generated is shown Fig 2.

Fig 1: Thessien polygon map of the study area

MANIT Bhopal

Page 59

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Fig. 3: Kmap for the Kodar catchment

Fig 2: Rmap for the Kodar catchment

3.1.2 Soil erodibility factor (K) The soil erodibility factor relates the rate at which different soils erode. K is expressed as soil loss per unit of area per unit of R from a standard plot (a plot of 22.3m long with a uniform slope of 9% under continuous fallow and tilled parallel to the slope. In case of USLE, the standard .14 100 K  2.1M 1Kodar (10 4 )(12  a )  3.25(b  2)  2.5( c  3) (5) Bagbahar a silt, very fine sand and clay [(% of where, M is the percent of Bar t unga very fine sand+% of silt)*(100-% of clay)], a is the organic matter, b is the structure of the soil (very fine granular=1, fine granular=2, coarse granular=3, lattic or massive=4) and c is the permeability of the soil (fast=1, fast to moderately fast=2, moderately fast =3, moderately fast to slow=4, slow=5, very slow=6). For determination of organic matter from organic carbon a factor 1.724 has been used (BUB, 2007; Wayne et al, 2003). The soil map of the study area has been taken from the soil map of National Bureau of Soil Survey & Land Use Planning (NBSS&LUP). By using the operation attribute map and feeding the table as Kvalue, Kmap has been generated in the ILWIS 3.0 software (Table 1 & Fig. 3).

3.1.3 Topographic index ( ) The topographic index was calculated using the Digital Elevation Model which has been generated using contour map and point elevation map obtained from the seamless data distribution database. The use of DEM has been documented by Mitasova et al (1996) to be the most reliable elevation data when higher resolution data is unavailable because it allows for lower levels of systematic errors and artifacts of analysis compared to the lower resolution DEMs that are available. The interpolation for contour map and rasterize operation for point elevation has been performed to get two separate raster maps. The „iff‟ statement of ILWIS has been used to combine both the raster maps to get the DEM. The points defining the flow line are computed as the points of intersection of a line constructed in the flow direction given by aspect angle a: and a grid cell edge. The Map Calculation option of raster operation in ILWIS has been used to determine topographic factor map (Fig. 4).

Table 1: Computation of K-factor for soils in the study area

573.63 515.99

Nomencl ature

% Fine sand

% Silt

% Clay

657 &670

11.03

11.32

689

8.60

23.87

1.80 12.2 2

710

6.30

5.41

0.00

733

4.47

14.12

2.14

746

3.20

26.87

3.22

747

10.03

19.83

0.00

HYDRO 2014 International

M

2668. 59 2850. 38 1171. 00 1819. 22 2910. 32 3086. 00

a

1.6 2 2.0 3 1.6 2 1.2 1 1.9 7 0.8 6

b

c

K Fact or

458.35 400.72 343.08 285.44

3

1

0.15

3

1

0.20

3

3

0.09

3

3

0.15

3

2

0.20

3

2

0.24

Fig 4: Digital elevation model for Kodar catchment

3.1.4 Cover and management factor (C) The main role of vegetation cover in the interception of the rain drops is that their kinetic energy is dissipated by them. The crop management factor is the expected ratio of soil loss from land cropped under specified conditions to soil loss from clean, tilled fallow or identical soil and slope and under the same rainfall. Available soil loss data from undisturbed land were not sufficient to derive C values by direct comparison of measured soil loss rates, as was done for the development of C values for cropland. The following equation suggested by Van der et al. 1999, 2000 has been used for estimation of C factor.

MANIT Bhopal

Page 60

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

 NDVI    

(6) C  exp   NDVI  The α-value of 2 and β-value of 1 gave good results ( Ioannis et al, 2009) have been used in the study. It has been observed that some values of C-factor may be reached to value greater than the limiting value of 1.0 and hence a scaling factor Z was used to keep the C-factor within the range of 0 to 1 (Mokua, 2009). The equation 3.6 can be written as:   

C  Z  exp

NDVI  

  NDVI 

(7) For computation of value of Z, a scalar graph can be plotted between NDVI and C-factor and value of Z has been determined by iterations to scale the values of C-factors from 0 to 1. NDVI has been calculated from the equation  RED  NIR  NDVI    RED  NIR 

conservation measures. The histogram of the resultant map has been used to estimate the rate of soil erosion from the catchment. The land use classification of the study area has been taken from IRS LISS IV data. Using spectral signatures of various land uses, sample sets for different land uses have been prepared. The maximum likelihood technique of classification has been used for generation of land use map of Kodar catchment. From the analysis, it has been observed that the Kodar catchment is an agriculture watershed covering nearly eighty percent of watershed with dense forest on the ridges only. Several small water bodies in the form of village tanks have been found in Kodar catchment which is used for bathing, cattle, recreation and other house hold work.

(8)

RED is Band III and NIR is Band IV of IRS satellites (IRS ID and P6). For determination of C-factor map of the study area, the NDVI image of LISS III data for the study area has been generated. The C-factor-map using equation 7 has been prepared and a graph between NDVI and C-factor values has been plotted. From the analysis of graph, it has been observed that the some of the C-factor values were going above the limiting value of Cfactor. Therefore, a correction factor of 0.6246 has been applied to keep all the values between 0 and 1 (Fig. 5).

Fig 6: P map for the Kodar catchment 4. ANALYSIS AND RESULT The map of all the factors responsible for developing the USPED model is generated. The sediment flux is than calculated by multiplication of all the maps separately for both rill and sheet. The directional derivative of the sediment transport capacity is than computed using the command Mapfilter. Finally using the Map slicing command the Erosional Depositional map for sheet and rill is generated.

Fig 3.5 C map for the Kodar catchment 3.1.5 Support practice factor (P) Conservation practice conditions consist mainly in the methods of land use and tillage, and the agro technology. The amount of soil loss from a given land is influenced by the land management practice adopted. The value of P ranges from 1.0 for up and down cultivation to 0.25 for contour strip cropping of gentle slope. In case of USPED model, the agricultural area of catchment has been divided in different slope ranges and according to slope, the values of P-factor have been assigned (Fig. 6). For other land uses, standard values considering no conservation measures have been given. All the thematic maps have been generated in ILWIS GIS for USPED model. After multiplication of thematic maps R, K, LS, C and P-factors, the annual and seasonal soil loss maps giving spatial distribution of soil losses have been generated. It has been observed from the field visits that presently no conservation measures are being implemented in study area, P-factor map has been generated using P-factor values for different land uses with no

HYDRO 2014 International

MANIT Bhopal

Page 61

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 SW-29 SW-31 SW-35 SW-37 SW-40 SW-42 SW-47 SW-52 SW-54 SW-56 SW-59 Total

Fig 7: Sheet and Rill Erosion and Deposition after map slicing in Kodar catchment. 4.1 Watershed prioritization using usped model : From the histogram all the erosional value for the sub-watershed has been taken and mean average value is calculated. Both the sheet and rill erosion value has been taken and the erosional value has been sorted0 between 0 and 1. The priorities of subwatersheds have been divided in the various ranges i.e. more than 0.50 as very high, 0.50 to 0.30 as high, 0.30 to 0.20 as moderate, 0.20 to 0.10 as low and less than 0.10 as very low priority. 4.2 Overall prioritization: The overall priority has been evaluated by taking the mean of the sheet and rill value. The final priorities of sub-watersheds have been divided in the various ranges i.e. more than 0.50 as very high, 0.50 to 0.30 as high, 0.30 to 0.20 as moderate, 0.20 to 0.10 as low and less than 0.10 as very low priority, so that environmentally stressed areas can be identified for soil conservation measures Table 4.1 Overall Prioritization of Sub Watershed S.N.

Priority Class

Range of final priority

No. of watershed

1.

V. high

Up to 0.50

08

2.

High

02

3.

Moderate

4.

Low

0.50 to 0.3 0.30 to 0.20 0.20 to 0.10

5.

V. low

Less than 0.10

47

03 07

HYDRO 2014 International

Sub-Watershed

SW-2, SW-38, SW-44. SW-45, SW -46, SW-48, SW-49 and SW63 SW-61 and SW64 SW-60, SW-62 and SW-65 SW-1, SW-5, SW-32, SW-50, SW-57, SW-66 and SW-67 SW-3, SW-4, SW-6, SW-7, SW-8 SW-9, SW-10, SW-11, SW-12 SW-14 SW-15 SW-16 SW-17 SW-18 SW-19 SW-20 SW-21 SW-22 SW-23 SW-24 SW-25 SW-26 SW-27 SW-28

Total area (sq. km)

24.29 9.93 23.65

41.08

208.76

SW-30 SW-34 SW-36 SW-39 SW-41 SW-43 SW-51 SW53 SW-55 SW-58 307.71

Fig 8: Overall Prioritization of Watershed 5. CONCLUSION Intensified pressures on the land and an improved understanding of human impacts on the environment are leading to profound changes in land management. This trend has a significant impact on the development of supporting GIS and modelling tools. In this paper, a soil erosion model at Kodar catchment with the integration of USPED (Unit Stream Power Erosion and Deposition) and GIS tools has been developed to estimate the annual soil loss. Different components of USPED were modelled using various mathematical formulae to explore the relationship between Rainfall emissivity, Soil erodibilty, Topographic factor, Crop factor and Practice factor maps. The USPED model was implemented in geographic information system (GIS) for predicting the spatial patterns of soil erosion risk required for soil conservation planning From the analysis of the Kodar catchment using USPED model it has been observed that 52.22 km2 area has been subjected to sheet erosion, while the eroded material may deposit in 42.48 km2 area of Kodar reservoir. The areas affected by sheet erosion may be treated with agronomic measures of soil conservation such as contour farming, contour bunding, bench terracing etc on cropped land and afforestation, agro- forestry on degraded forest and barren lands. Similarly, 55.25 km2 areas of Kodar reservoir may be affected by rill erosion where suitable mechanical soil conservation measures in the form check dams, gully plugs etc. may be constructed. According to this model, approximately in 67.8 % of the basin has very low erosion risk and 13.38 percent has low erosion risk 7.77 percent area has moderate risk. But erosion risk is high on 3.3% and Very High on 7.89% of the basin. In general, it is clear from the results of this study that the developed model is beneficial for the rapid assessment of soil erosion. REFERENCES

MANIT Bhopal

Page 62

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

i. Alejandra, Puerto, Rico., González, M, Rojas.,(2008) Soil erosion calculation using remote sensing and GIS in río grande de arecibo watershed, Annual Conference Portland, Oregon. ii. Bhattarai, Rabin., & Dutta, Dushmata., (2006) Estimation of Soil Erosion and Sediment Yield Using GIS at Catchment Scale Springer Science Business Media B.V.. iii. Jones, S, David., Kowalski, G, David., and Shaw, B, Robert., (1996) Calculating Revised Universal Soil Loss Equation (RUSLE) Estimates on Department of Defense Lands: A Review of RUSLE Factors and U.S. Army Land Condition-Trend Analysis (LCTA) Data Gaps. iv. Kumar, Suresh., and Kushwaha, SPS., Modeling Soil Erosion Risk based on RUSLE-3D using GIS in a Shivalik sub-watershed. v. Liu, Jinxun., Liu, Shuguang., Tieszen L. Larry and Chen, (2014) Mingshi Estimating Soil Erosion Using the USPED Model and Conservation Remotely Sensed Land Cover Observations vi. May, Linda., and Place, Chris., (2005) GIS-based model of soil erosion and transport Freshwater Forum vii. Mitasova, H. and Mitas, L., (1999) Erosion/deposition modeling with USPED using GIS. http://www2.gis.uiuc.edu:2280/modviz/erosion/usped.html. viii. Mitasova, Helena., Hofierka, Jaroslav., Zlocha, Maros., & Iverson, R, Louis., (1996) Modelling topographic potential for erosion and deposition using GIS, International Journal of Geographical Information Systems, , VOL 10, NO 5, 629-641. ix. Nearing, M.A., Jetten, V. , Baffaut, C., Cerdan, Couturierd, A., Hernandeza, M., Le Bissonnaise, Y., Nicholsa, H, M., Nunesf, P, J., Renschlerg, C.S., V. Souche`reh,. Oost, van, K., (2005) Modeling response of soil erosion and runoff to changes in precipitation and cover. x. Paige, Ginger., and Zygmunt, Jennifer.,(2012) The science behind wildfire effects on water quality and erosion. xi. Pistocchi, A., Cassani, G., and Zani, O., (2009) Use of the USPED model for mapping soil erosion and managing best land conservation practices, 47100 Forlì, Italy practices xii. Pricope, G, Narcisa., (2009) Assessment of Spatial Patterns of Sediment Transport and Delivery for Soil and Water Conservation Programs, Journal of Spatial Hydrology Vol.9, No.1 Spring. xiii. Wordofa, Gossa., (2011) Soil erosion modelling using GIS and RUSLE on the EURAJOKI watershed FINLAND.

A Novel Optimisation Model Applied to Godavari River Basin R.B.Katiyar2,Balaji Dhopte1, Tejeswi Ramprasad1, Shashank Tiwari2, Anil Kumar2, K.R.Gota2 1 Department of Chemical Engineering, Jawaharlal Nehru Engineering College, Aurangabad-431003 1 Department of Chemical Engineering, Maulana Azad National Institute of Technology, Bhopal-462051 Email: [email protected], [email protected] ABSTRACT: Integrated water resources management (IWRM) is a rapidly developing field encompassing many disciplines including ecology, engineering, economics, and policy. Generic integrated watershed management optimization model is developed to study efficiently a broad range of technical, economic, and policy management options within a watershed system framework and choose the optimum combination of management strategies and associated water allocations for designing a sustainable watershed management plan at minimum cost. The watershed management model integrates both natural and human elements of a watershed system and

HYDRO 2014 International

includes the management of ground and surface water sources, water treatment and distribution systems, human demands, wastewater treatment and collection systems, water reuse facilities, non potable water distribution infrastructure, aquifer storage and recharge facilities, storm water, and land use. The model was formulated as a linear program and applied to Godavari basin in India. Results according to the study carried out demonstrate the merits of integrated watershed management by showing the relative effectiveness and economic efficiency of undervalued management options , the value of management strategies that provide several functions such as the benefits of increased infiltration for meeting both storm water and water supply management objectives and that both human and environmental water needs can be met by simultaneously implementing multiple diverse management tools, which in this case study led to achieving 60-65% of the recommended in-stream flow with only 25% decrease in net benefits. Keywords: Optimization models; Integrated systems; Water supply; Watersheds; Water management; Storm water management; Land management; Wastewater management; Groundwater recharge 1. INTRODUCTION Water is an important resource which is used in each and every industrial sector. But the increasing demand on water from the sectors emphasizes the need of integrated watershed. It therefore becomes necessary to understand what is a watershed, the various kinds of interactions in a watershed, the side effects of degradation of a watershed and basic approach on how to implement a watershed management plan for a water source. (USEPA publication.,2013) A watershed is the area of land that delivers runoff water, sediment and dissolved substances to a river. It a unit which collects, stores and releases water through the networks to the main river. It is an integration of flora, fauna, land, water and their interacting elements. It is quite clear that in order to study the integrated watershed management we need to have a basic knowledge of the hydrological principles which govern the occurrence, distribution, movement and properties of the water. The hydrological cycle describes the various paths water may take during its continuous circulation from ocean to atmosphere to earth and back to ocean. Water is temporarily stored in streams, in lakes, in the soil and as groundwater. The basic watershed equation is given as: P=I + F + E + T+ Q ± S Where, P is precipitation, I is interception, F is filtration, E is evaporation, T is plant transportation, Q is runoff and S is storage. Atmospheric moisture is one of the smallest storage volumes of the earth‟s water, yet it is the most vital source of freshwater for humankind. The distribution and amount of precipitation (P)

MANIT Bhopal

Page 63

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

depends on air mass circulation patterns, distance and direction from large water bodies and local topography. Precipitation may be intercepted or captured by leaves, twigs, stems and soil surface organic matter and returned to the atmosphere as water vapour. This process known as interception (I) does not help to recharge soil moisture or generate stream flow in fact it lessens the impact of the raindrop on the soil surface and the danger of soil erosion. When water reaches the ground surface, a portion of it is absorbed by the soil. Infiltration (F) is the process of water seeping into the soil and is controlled by surface soil conditions, such as soil texture, vegetation type and land use. For the purpose of integrated watershed management, necessity is to develop models which focus on developing comprehensive watershed management models as opposed to the existing redundant hydrologic models. Such models are referred to as integrated watershed management models. Two of these models which are the most common models in practice are Water Evaluation and Planning (WEAP) (Yates et.al, 2005) and Water Ware (Jamieson and Fedra., 1996) 2. MODEL FORMULATION The model introduced here is a generic lumped parameter model that combines the principles of the hydrologic cycle, human water system and a wide range of management options. The natural components of the watershed system are depicted with white backgrounds. These include land use, runoff, percolation, surface water, groundwater and external surface water, and ground water. Run off and percolation is specified as unit values of flow per land area for each land use type for a hydrologic design condition. The land use component specifies the existing area of each land use type. Surface water, representing rivers and other landscape sources of water is assumed to have negligible channel storage and hence empties completely within each time step. The underground water is the only natural watershed component with a large storage capacity. The human components of the watershed system are depicted with gray and black backgrounds. Gray is used for components that exist and are managed by water and waste water utilities. The human system includes a reservoir, potable water treatment plant, potable distribution system, wastewater treatment plant, wastewater collection system, water reuse facility, non potable distribution system, septic systems, and aquifer storage and recharge facility. The reservoir may be a single reservoir or the sum of many reservoirs assumed to be operated together as a single reservoir system. The potable water treatment plant treats water from surface water, reservoir, or groundwater sources to drinking water standards. The wastewater treatment plant provides secondary wastewater treatment to meet surface water discharge quality standards. Its effluent may be further treated by tertiary wastewater treatment at the water reuse facility. (Zoltay, et.al. 2010) Fig.1 : Schematic representation of the integrated watershed management model

HYDRO 2014 International

BMP‟s- Best Management Practices, SW – Surface Water GW- Ground Water WTP – Water Treatment Plant P use – Potable use NP use – Non Potable use ASR – Aquifer Storage and Recharge WWTP – Waste Water Treatment Plan 3. BACKGROUND The river Godavari is the second largest river in the country and the largest in Southern India. It raises in the Sahyadri hills at an altitude of about 1067 m near Triambakeswar in the Nasik district of Maharashtra State and flows across the Deccan plateau from the Western Ghats to Eastern Ghats. Rising in the Western Ghats about 80 km from the shore of the Arabian sea, it flows for a total length of about 1465 km in a general SouthEastern direction through the States of Maharashtra and Andhra Pradesh before joining the Bay of Bengal at about 97 km south of Rajahmundry in Andhra Pradesh. The major tributaries joining the Godavari are the Pravara, the Purna, the Manjra, the Maner, the Pranhita, the Penganga, the Wardha, the Wainganga, the Indravati and the Sabari. The Godavari basin extends over an area of 312813 km2, which is nearly 10% of the total geographical area of the country. The basin comprises areas in the States of Maharashtra, Madhya Pradesh, Chhattisgarh, Andhra Pradesh, Karnataka and Orissa. The State-wise distribution of the areas is given in table below: Table 1: Distribution of Godavari river Sr. No.

Name of the state

1. 2. 3. 4.

Maharashtra Madhya Pradesh Chhattisgarh Andhra Pradesh

MANIT Bhopal

Drainage are (km2) 152199 26168 39087 73201

Percentage of the total basin drainage area 48.6 8.4 12.5 23.4

Page 64

International Journal of Engineering Research Issue Special3 5. 6.

Karnataka Orissa Total

4406 17752 312813

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

1.4 5.7 100.0

Except for the hills forming the watershed around the basin, the entire drainage basin of the river Godavari comprises of undulating country, a series of ridges and valleys interspersed with low hill ranges. Large flat areas which are characteristic of the Indo-Gangetic plains are scarce except in the delta. The Sahyadri ranges of Western Ghats form the Western edge of the basin. The interior of the basin is a plateau divided into a series of valleys sloping generally towards East. The Eastern Ghats, which form the Eastern boundary, are not so well defined as the Sahyadri range on the West. The Northern boundary of the basin comprises of tablelands with varying elevation. Large stretches of plains interspersed by hill ranges lie to the South. Important tributaries of Godavari is given the following table : (Integrated Hydrological Data Book.,2006) Table 2: Tributaries of Godavari river Sr. No.

Name of the river

Elevation of source

Length of tributary (km)

Catchment Area (sq.km.)

1

Upper Godavari Pravara Purna Manjira Middle Godavari Maner Penganga Wardha Pranhita Lower Godavari Indravati Sabari

1,067

675

33502

Average annual Rainfall (mm) 770

1,050 838 823 323

208 373 724 328

6537 15579 30844 17205

606 797 846 955

533 686 777 640 107

225 676 483 721 462

13106 23898 24087 61093 24869

932 960 1055 1363 1208

914 1,372

535 418

41665 20427

1588 1433

2 3 4 5 6 7 8 9 10 11 12

The water resources potential in Godavari basin has been assessed to be 110.54 km3.The utilisable surface water is about 76.3 km3 ,the replenish able ground water is about 45 km3. There is a vast potential for irrigation development and hydropower generation in the basin. Prior to Independence only a few irrigation projects were constructed in Godavari basin. Important among these are Godavari delta system (with Dowlaiswaram weir as head works), Nizamsagar reservoir, Kadana dam and Pravara dam. After independence, under various five year plans a large number of multipurpose and irrigation projects have been taken up. Themost important among them are the Jaikwadi, Sriramsagar, Kadam, Upper Indravati, Singur and Godavari Barrage (by modernising the existing gated weir at Dowlaiswaram). Since the mid1960's, the Central Water Commission is conducting hydrological observations in Godavari basin. Hydrological observation stations have been established on main Godavari River as well as on all the important tributaries. During the year 2008-09,

HYDRO 2014 International

hydrological observations at 48 stations have been under operation. Out of these, 7 stations are on the main Godavari and the remaining 41 are on its tributaries. In addition to gauge and discharge observations, sediment load at 16 stations and water quality monitoring at 18 stations are also being done. There are 32 water quality measurement sites on the basin and as many as 25 of them are for sediment measurements also. In addition, there are 24 gauge discharge observation stations in the basin. 4. IMPLEMENTATION OF MANAGEMENT OPTIONS The different ways by which the available water resources can be managed is by the effective application of judicious methods. This is where the management options come into picture. Once the above model is applied we get the following management options which are listed in the table : (Zoltay, et.al.,2010) Table 3: Management options Module Storm water run off Usage of land

Supply of treatment

water

Demand management

Wastewater treatment

Aquifer storage Inter basin transfer

and

Management options More bio retention units should be installed Forest land and cover should be preserved More land should be purchased depending on need Surface water pumping Groundwater pumping Treatment of water Surface storage Repair of leakages in the distribution system Increasing revenues for water and wastewater services Secondary treatment Reuse by treating with tertiary methods Distribution system for non potable water Repair in filtration into collection system Replenish ground water with water from reservoirs Import potable water Export waste water

5. RESULTS OF WATER SHED MANAGEMENT MODEL The main storage capacity is in groundwater aquifers, which were used through ASR and bio retention units. Another interesting aspect of these results is that both bio retention units and ASR were recommended even though they serve similar functions of recharging groundwater.The utilization of the bio retention facility and the ASR facility highlights the need to increase the ground water recharge in the basin. ASR is more effective and versatile than the bio retention units in terms of source of recharge water and the quantity of water flow. Although the repair of leaks in distribution infrastructure is increasingly common, repairing sewer pipes to prevent the infiltration of groundwater is generally considered too costly because of the deeper and larger diameter pipes. 6. CONCLUSION An integrated watershed management optimization model to support informed decision making was introduced and used to evaluate a wide range of management options including land-use

MANIT Bhopal

Page 65

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

management to simultaneously address numerous watershed management objectives, which are traditionally modelled independently. The model demonstrated that with an increasing diversity of management options, net benefits of watershed management can increase. In addition, our results indicated that demand management through price changes and the repair of leakage in water distribution and wastewater collection systems are effective management options as they were selected in all scenarios where they were available. The recommendation for the joint implementation of ASR and bio retention units demonstrated that complex interactions among components of a watershed necessitate the evaluation of management options within a systems framework in order to realize the full impact of management decisions and to enable informed decision making. REFERENCES: i. Integrated Hydrological Data Book, Water Planning & Projects Wing Central Water Commission, New Delhi, September, 2006, pp 15-16 ii. Jamieson, D. G., and Fedra, K..,The ‗Waterware‘ Decisionsupport System For River-Basin Planning. 1: Conceptual Design .,1996., pp 163–175 iii. UNEPA Publication,. A Quick Guide To Developing Watershed Plans To Restore And Protect Our Waters, May 2013 iv. Viktoria I. Zoltay, Richard M. Vogel,Paul H. Kirshen,Kirk S. Westphal., Integrated Watershed Management Modeling: Generic Optimization Model Applied to the Ipswich River Basin, Journal Of Water Resources Planning And Management ., September/October 2010, Pp 566-575 v. Yates, D, Sieber, J., Purkey, D, and Huber-Lee, A.,WEAP21-A demand-, priority-, and preference-driven water planning model, Part 1: Model characteristics. Water Int., 2005., pp 487–500

Runoff and Sediment Yield Modeling of an Agricultural Hilly Watershed Using Wepp Model 1

2

3

Saroj Das , Laxmi Narayan Sethi and R. K. Singh 1. M. Tech. Student, Department of Agricultural Engineering, Triguna Sen School of Technology, Assam University, Silchar-788011 2. Associate Professor, Department of Agricultural Engineering, Triguna Sen School of Technology, Assam University, Silchar-788011 3. Principal Scientist & Head, Agricultural Engineering Division, ICAR Research Complex for NEH Region, Barapani (Umiam), Meghalaya-793 103 Email:[email protected] ABSTRACT: Soil erosion rates caused by water are highest in agro systems located in hilly or mountainous regions of Asia, Africa and Southern America, especially in less developed countries. Each year about 10 million ha of cropland are lost due to soil erosion, thus reducing the cropland available for food production. The loss of cropland is a serious problem. So, a good management practice to protect the soil from erosion to sustain long-term productivity is imperative for meeting the world‟s future demand for food and fiber. Thus, the present study was undertaken to develop the best management practices for a small

HYDRO 2014 International

hilly watershed (Mawpun, Meghalaya) in North Eastern of India. The watershed covers around 57.17 ha and falls under high rainfall and high land slope conditions. For quantification of runoff, sediment yield from areas of different land uses and conservation practices of the watershed a physically based Water Erosion Prediction Project (WEPP) model was used. The WEPP model was calibrated using meteorological data (2002 to 2004) and most sensitive soil related parameters (namely, rill erodibility, interrill erodibility, effective hydraulic conductivity and critical shear stress) of the small treated watershed (Mawpun watershed) and validated using data of 2005 and 2006 monsoon season. The performance of the model was also evaluated by estimating the daily runoff and sediment yield using the monsoon season data of different years. Coefficient of determination (R2) (0.72–0.96), Nash–Sutcliffe simulation model efficiency (0.71–0.95), and percent deviation values (16.4-21.2) indicate resonable simulation accuracy of runoff from the watershed. High value of coefficient of determination (R2) (0.73–0.94), Nash–Sutcliffe simulation model efficiency (0.55–0.89) and percent deviation values (16.1– 19.3) for sediment yield indicate that the WEPP model can be successfully used in the Mawpun watershed, India. Keywords: Runoff, Sediment yield, Watershed Management, WEPP Model.

1. INTRODUCTION: Land and water are the most precious natural resources, the importance of which in human civilization needs no elaboration. The overexploitation of these natural resources causes natural imbalance of the ecosystem and environment degradation. Soil erosion is one of the main reasons for degradation of soil and water quality ultimately adversely affecting the environment. About 99.7% of the food consumed by human beings comes from the land (Pimentel and Pimentel, 2003) and about 1964.4 million ha area which is 12% of the world‟s total land surface suffers from degradation problems (Koohafkan, 2000). Therefore, to combat the problem of resource degradation and ecological imbalance, appropriate management practices were the most efficient factor for long term agricultural sustainability. With these facts in mind the present study was conducted to evaluate the WEPP model for quantification of runoff and sediment yield from areas under different land uses and conservation practices.

2. MATERIAL AND METHODS: 2.1 Study area: The study site (Mawpun Watershed) is located at 250 41‟ N latitude, 910 55‟ E longitudes and at an altitude of 1010 m in RiBhoi district of Meghalaya state of India. The location of the study site is shown in Figure 3.1.The study area is a part of the eastern Himalayan range is made up mostly of Precambrian metamorphic and igneous rocks. The study area is mainly hilly

MANIT Bhopal

Page 66

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

with steep slope that ranges between 0 to 30 % and the maximum slope of some hilly portion is nearly 100%.

2.5 Model performance evaluation: The hydrological model was evaluated through a pair wise comparison of the observed and simulated data to determine the closeness of their match. Split sample calibration approach was adopted for model‟s performance evaluation. Five-year‟ data set pertaining to 2002 through 2006 was split into two parts. The data of 2002-2004 were used for model calibration and that of 2005-2006 for model validation. The manual calibration based on trial-and-error procedure (Sorooshian and Gupta, 1995) was used in the study. Singh et al. (2011) reported that soil related parameters namely; rill erodibility, interrill erodibility, effective hydraulic conductivity and critical shear stress were most sensitive in Meghalaya conditions. Therefore, only these parameters were considered for calibration. The calibrated values of these parameters reported by Singh et al. (2011) were taken as base value and fine tuned for the Mawpun watershed.

2.2 Meteorological and hydrological data:

The weather data such as daily rainfall, maximum and minimum temperature, morning and evening relative humidity, wind speed, pan evaporation and sun shine hours for a period of 5 years (2002–2006) were collected from the Agricultural Engineering Division, ICAR Research Complex for North East Hill Region and analyzed for making the model input files. The observed hydrological data such and daily sediment yield for the periods of five years (2002 to 2006) were collected from the Agricultural Engineering division, ICAR Research Complex for NEH Region and analyzed for making model input file. 2.3 Topographic data and soil properties: Topographic information pertaining to the Mawpun watershed in the form boundary map, contour map, drainage map, soil map, and land use/land cover maps were collected from the Agricultural Engineering Division, ICAR Research Complex for North Eastern Hill Region, Barapani and used for delineation of watershed. Physical and chemical properties of soil for the study watershed were collected from Agricultural Engineering Division, Indian Council of Agricultural Research Complex for NEH Region.

Figure-1: Location map of Mawpun watershed

2.4 WEPP model: The USDA – WEPP (Water Erosion Prediction Project) Hillslope is a physically based, distributed parameters model based on fundamentals of stochastic weather generation, infiltration theory, hydrology, soil physics, plant science, hydraulics and erosion mechanics. Date, amount, intensity and duration of rainfall, minimum and maximum temperatures, wind velocity and direction at 8 and 14 h of the day, daily values of radiation and dew point temperatures for the period of 2002–2006 were used as input to create climate input files for WEPP model using Break Point Climatic Data Generator (BPCDG). The delineation of watershed using WEPP model was presented in Fig. 2. Slope and soil files were created using slope and soil file builder within the WEPP interface. The management input file was built using file builder within the model interface.

HYDRO 2014 International

Figure-2: Delineated hillslopes and channels of the Mawpun Watershed using WEPP model (Martinec and Rango, 1989), Nash and Sutcliffe (1970) simulation coefficient (ENS) and coefficient of determination (R2) were determined. Performance of the model was evaluated for runoff as well as sediment yield simulations. The underprediction/over-prediction by the model within or equal to ±25%

MANIT Bhopal

Page 67

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

of observed values were considered acceptable level of accuracy for the simulations as suggested by Bingner et al. (1989).

3. RESULTS AND DISCUSSIONS: 3.1 Simulation of runoff and sediment yield: The daily observed runoff and sediment yield hydrographs for the calibration (May–October) 2002 to 2004 and the validation periods (May–October) 2005 and 2006 are shown in Figure-3 through Figure-7 and Figure-8 through Figure-12, respectively. It is observed that the trend of the simulated values closely matches the trend of the measured values for calibration periods and validation periods. However, the measured daily runoff and sediment yield of higher magnitude is under-predicted by the model during simulations for calibration periods and validation periods. Based on the goodness-of-fit test statistics (Table-1, Table-2, Table-3 and Table-4), it can be concluded that the WEPP model simulates daily runoff from the Mawpun watershed with acceptable accuracy.

Figure-5: Observed and simulated daily runoff hydrograph of Mawpun watershed during model calibration for the period of May to October 2004.

Figure-6: Observed and simulated daily runoff hydrograph of Mawpun watershed during model validation for the period of May to October 2005.

Figure-3: Observed and simulated daily runoff hydrograph of Mawpun watershed during model calibration for the period of May to October 2002. Figure-7: Observed and simulated daily runoff hydrograph of Mawpun watershed during model validation for the period of May to October 2006.

Figure-4: Observed and simulated daily runoff hydrograph of Mawpun watershed during model calibration for the period of May to October 2003. Figure-8: Observed and simulated daily sediment yield of Mawpun watershed during model calibration for the period of May to October 2002.

HYDRO 2014 International

MANIT Bhopal

Page 68

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 parameter

Obs erve d 4.16

Sim Obser ulate ved d 2004 4.87 4.24

Simulat ed

Std.Dev.

Obser Sim ved ulate 20 2 d 02 0 5.81 5.00 0 5.04 3 4.70

7.49

6.29

5.08

4.93

Maximum

33.2

20.1

46.3

30.1

30.3

25.0

Total

465.5

512. 2 93

442.3

90

437. 2 93

382.2

No ofevents %Dv

541. 0 90

102

102

Mean

Figure-9: Observed and simulated daily sediment yield of Mawpun watershed during model calibration for the period of May to October 2003.

16 0. .2 60 0. 70

R2 ENS

1 0. 7. 92 0. 0 8 8

4.91

-15.7 0.76 0.74

Table-2: Goodness-of-fit statistics of observed and simulated daily runoff simulation during validation periods 2005 and 2006 (May to October).

Statisti cal parame ter Mean Std.De v. Maxim um Total

Figure-10: Observed and simulated daily sediment yield of Mawpun watershed during model calibration for the period of May to October 2004.

No ofevent %Dv s R2

Sediment yield(t/ha) Ob ser ved 0.2 0 0.3 1 2

Simulated

Observ Simu ed lated 2003

Observed

0.24

0.16

0.19

0.19

0.27

0.31

0.29

1.11

1.99

1.65

22.7

17.2

20.1

90

93

2002

2004

-16.4

-16.8

0.81

0.79

0. 2 0.31 0. 2 2 2.0 1. 8 2 16.8 12 9. 102 16 0 -16.7 2 0.74

0.55

0.76

0.73

19. 5 90

ENS

93

S i m u l a t e d

Table-3: Goodness-of-fit statistics of observed and simulated daily sediment yield simulation during calibration periods 2002 through

2004 (May to October). Figure-11: Observed and simulated daily sediment yield of Mawpun watershed during model validation for the period of May to October 2005.

Stati stical para mete r Mea n Std. Dev. Maxi mum Total No ofeve %Dv nts R2

Figure-12: Observed and simulated daily sediment yield of Mawpun watershed during model validation for the period of May to October 2006. Table-1: Goodness-of-fit statistics of observed and simulated daily runoff simulation during calibration periods 2002 through 2004 (May to October). Statistical

Runoff (mm) Observe d

Simulated

Observed

2005

2006

Simula ted

3.31

3.90

2.42

2.83

4.36

3.76

4.06

3.76

32.2

18.0

23.1

17.0

337.7

398.3

247.5

289.5

105

105

102

102

ENS

-17.9

-17.0

0.67

0.80

0.73

0.80

Table-4: Goodness-of-fit statistics of observed and simulated daily sediment yield simulation during validation periods 2005 and 2006 (May to October). Statistical

Sediment yield(t/ha)

Runoff (mm)

HYDRO 2014 International

MANIT Bhopal

Page 69

International Journal of Engineering Research Issue Special3 parameter

Obse rved

Mean Std.Dev. Maximum Total No ofevents %Dv R2 ENS

0.11 0.17 1 11.5 105

Simulated 2005 0.13 0.16 0.7 13.9 105 -20.9 0.69 0.62

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 v. Pimentel, D., Pimentel, M., 2003. World population, food, natural resources and survival. World Future, Vol. 59(3-4); 145-167. vi. Singh, R.K., Panda, R.K., Satapathy, K.K., Ngachan, S.V., 2011. Simulation of runoff and sediment yield from a hilly watershed in the eastern Himalaya, India using the WEPP model. Journal of Hydrology, Vol. 405(3-4); 261-276. vii. Sorooshian, S., Gupta, V.K., 1995. Model calibration. In: Singh, V.P. (Ed.), Computer Models of Watershed Hydrology. Water Resources Publication, Highlands Ranch, Colorado, USA; 23–68.

Obser Sim ved 2006ulat ed 0.09 0.1 10.1 0.17 60.9 1.00 4 9.4 11. 2102 102 -19.1 0.80 0.57

Prioritization of a Watershed Based on Spatially Distributed Parameters

4. CONCLUSSIONS: 1

In the present study, we tested the WEPP model for its efficacy to predict runoff and sediment yield in high rainfall and steep slope conditions of eastern Himalaya. The model was used to develop vegetative and structural control measures to enhance agricultural sustainability in the Mawpun watershed. Based on results of the study the following conclusions were drawn: 1. The WEPP model simulates runoff and sediment yield satisfactorily in high rainfall and high slope conditions of Meghalaya with Nash–Sutcliffe coefficients > 0.50 and percent deviations < ± 25.0. Comparison between WEPP–simulated and measured values of runoff and sediment yield revealed that the model tends to under-predict the values of higher magnitude. 2. Toposequential cropping on hill slope with graded bunding and terracing at appropriate locations reduced the sediment yield by 52%. 3. Crops cultivation in mild sloped and valley lands with graded bunding, crop cultivation in bench terraces in medium to high slope up to 30%, horticultural fruit crops from 30 to 60% slope and forest or timber farming on land slope above 60% yielded sediment at the rate of 9.4 t/ha. 4. Thus topo-sequential land use reinforced with graded bunding and terraces at appropriate locations will bring the sediment yield within the safe limit enhancing the sustainability and profitability of agricultural system in hilly ecosystem. 5.

REFERENCES:

i. Bingner, R.L., Murphee, C.E., Mutchler, C.K., 1989. Comparison of sediment yield models on various watershed in Mississippi.Trans, ASAE, Vol. 32 (2); 529–534. ii. Koohafkan, A.P., 2000. Land resources potential and sustainable land management- An overview. Lead paper of the International conference on Land Resource Management for Food, Employment and Environmental Security during November 9-13, New Delhi(India); 1-22. iii. Martinec J, Rango A (1989) Merits of statistical criteria for the performance of hydrologic models. Water Resour Bull AWRA, Vol. 25; 421– 432. iv. Nash, J.E., Sutcliffe, J.V., 1970. River flow forecasting through conceptual models Part 1-A discussion of principals. J. Hydrol. 10 (3), 282– 290.

HYDRO 2014 International

C. D. Mishra1, R.K. Jaiswal2, A. K. Nema1 Institute of Agriculture Sciences, Banaras Hindu University Varanasi (U.P.) -221005 2 National Institute of Hydrology, Regional Center, Bhopal (M.P.) – 462001 Email: [email protected]

Abstract: Identification of erosion prone and runoff generation areas of a watershed is essential for the effective and efficient implementation of best management practices for conserving the natural resource in favour of sustainable development. In this study, an effort has been made to identify critical erosion-prone areas of the Nagwan watershed (89.44 km2) of Upper Damodar Valley situated in Hazaribagh District in Jharkhand state India, using the spatially distributed parameters responsible for hazard of erosion. A geographical information system and remote sensing was used for generating these parameters including slope factor, soil erodibility factor of Universal Soil Loss Equation (USLE), stream power index, sediment transport index and curve number (CN) value, topographic wetness index for water conservation. Using supervised classification method with a maximum likelihood (ML) technique was applied to three multi-spectral bands to generate the land use/cover map from IRS-P6 (LISS-IV) satellite data and found six land use classes such as agricultural land (55.78 km2), dense forest (1.47 km2), open forest (11.63 km2), barren land (0.25 km2), water body (1.26 km2), shrubs land (3.46 km2) and built up land (4.76 km2). The soil erodibility factor map was prepared from the soil map, and K factor values from a soil survey data. The Watershed priorities have been divided in four categorizes namely very high, high, moderate, and low priority. From the analysis, 13.45 km2 and 22.81 km2 have been found under very high and high priority classes respectively where immediate attention for soil and water conservation measures are required. Keywords: GIS, remote sensing, wetness index, stream power index, sediment transport index, Watershed prioritization 1. INTRODUCTION Watershed is an ideal unit for management of natural resources that also supports land and water resource management for achieving sustainable development. The significant factor for the planning and development of a watershed are its physiography, drainage, geomorphology, soil, land use/land cover and available

MANIT Bhopal

Page 70

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

water resources. The concept of watershed management recognizes inter-relationship among land use, soil, water and the linked between uplands and downstream areas (Tideman, 1996). The deterioration occurs generally in terms of forest loss and land degradation by soil erosion. Among several factors, the major one is deforestation followed by unsuitable agricultural practices. Watershed characteristics, such as land use/land cover, slope, and soil attributes, affect hydrologic and water quality processes and hence regulate sediment and chemical concentration (Basnyat et al. 2000). Knowledge of the basic hydrologic processes occurring in watersheds give a better understanding of land use impacts on soil and water resources. Change in land use/land cover is considered as an important hydrologic factor affecting storm runoff generation and sediment yield (Calder 1992; Naef et al. 2002; Bakker et al. 2005). This is especially true for humid and sub-humid subtropical areas in India which are affected by heavy monsoon rains during four to five rainy months (Sharma et al. 2001). With reference to nonpoint source (NPS) pollution, the critical areas are those areas where either soil erosion exceeds the soil loss tolerance limit or where the maximum improvement in the quality of water resources can be attained with the minimum capital investment through best management practices (Mass et al. 1985). Land and water are the two basic natural resources for the survival of living systems. These two resources have been interacting with each other in various phases of their respective cycles. The future of the nation depends largely on the effective utilization, management and development of these resources in an integrated and comprehensive manner. Soil erosion has been accepted as a serious problem arising from agricultural intensification, land degradation and possibly due to global climatic change (Yang et al.,2003). Accelerated soil erosion has been globally recognized as a serious problem since people took up agriculture (Renschler et al., 1999). In India, annual soil erosion (displacement of soil) rate is about 5334 million tones out of which about 1572 million tones is carried away by the river systems into the sea and 9% of total annual soil erosion i.e. about 480 million tones is deposited in the various reservoirs reducing their carrying capacity (Dhruva Narayan and Ram Babu,1983). Under Indian conditions, an average soil loss value of 16.4 t/ha-yr (Narayana 1993) may be considered as the limit for identifying critical watershed areas (Singh et al. 1992). Satellite based remote sensing technology meets both the requirements of reliability and speed and is an ideal tool for generating spatial information needs. However, the use of remote sensing technology involves large amount of spatial data management and requires an efficient system to handle such data. Thus, blending of remote sensing and GIS technologies has proved to be an efficient tool and have been successfully used by various investigators for water resources development and management projects as well as for watershed characterization and prioritization (Chalam et al. 1996; Chaudhary and Sharma 1998; Kumar et al. 2001;Ali and Singh 2002; Singh et al. 2003; Pandey et al. 2004; Suresh et al. 2004). A few more studies are

HYDRO 2014 International

reported where remotely sensed data had been used for the assessment of soil degradation to devise cost effective methods for soil conservation (Jain and Kothyari 2000; Jain et al. 2001; Baba and Yusof 2001; Fistikoglu and Harmancioglu 2002; Sekhar and Rao 2002; Chowdary et al. 2004; Pandey et al. 2007). Digital elevation models (DEMs) are already widely used and play an increasing important role in geomorphology, hydrology, soil erosion and many related geoanalysis fields (Moore et al., 1991; Goodchild et al., 1993; Wise, 2000). Topography is a firstorder control on spatial variation of hydrological conditions. It affects the spatial distribution of soil moisture, and groundwater flow often follows surface topography (Burt and Butcher, 1986; Seibert et al., 1997; Rodhe and Seibert, 1999; Zinko et al., 2005). The TWI is usually calculated from gridded elevation data. Different algorithms are used for these calculations; the main differences are the way the accumulated upslope area is routed downwards, how creeks are represented, and which measure of slope is used (Quinn et al., 1995; Wolock and McCabe, 1995; Tarboton, 1997; Guntner et al., 2004). The topographic wetness index (TWI) has been used to describe the spatial soil moisture patterns and zones of saturation or variable sources for runoff generation is obtained (Beven and Kirkby, 1979; Wilson and Gallant, 2000) and also used to study spatial scale effects on hydrological processes (Beven et al., 1988; Famiglietti and Wood, 1991; Sivapalan and Wood, 1987; Siviapalan et al., 1990) moreover to identify hydrological flow paths for geochemical modelling (Robson et al., 1992) as well as to characterize biological processes such as annual net primary production (White and Running, 1994), vegetation patterns (Moore et al., 1993; Zinko et al., 2005), and forest site quality (Holmgren, 1994a). The locations of higher TWI host more favorable conditions for landslide formation (Conoscenti et al., 2008). The stream power index could be used to identify the erosive effects of concentrated surface runoff (Wilson and Gallant, 2000), to identify suitable locations for soil conservation measures and reduce the effect of concentrated surface runoff. The sediment transport index accounts for the effect of topography on erosion. The two-dimensional catchment area is used instead of the one-dimensional slope length factor as in the Universal Soil Loss Equation. 2. Description of the study area Nagwan watershed (89.44 km2) is located the Upper Damodar Valley, situated in Hazaribagh district of Jharkhand, India, the second most seriously eroded area in the world (EI-swaify et al. 1982), was selected for the study. The watershed lies between 85016′41″ and 85023′50″ E longitudes and between 23059′33″ and 2405′37″ N latitudes. Location map of the study area is shown in Figure 1. The test watershed is just 7 km from the soil conservation department of Damodar Valley Corporation (DVC) at Hazaribagh, Jharkhand; is well connected by road/rail network. Geologically, the area is quite complex, having rocks of varying composition. The soils of the area are mainly of clay loam and silty loam type. The topography of the watershed is

MANIT Bhopal

Page 71

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

undulating and maximum and the minimum elevations of the area are 667 m and 560 m, respectively. The area experiences sub-humid sub-tropical monsoon type of climate, characterized by hot summers (40◦C) and mild winters (4◦C). The watershed receives an average annual rainfall of 1256 mm, out of which more than 80% rainfall contributes during monsoon season (June–October). The average storm intensity, by considering storms of more than 30 min duration, is about 10 cm/hr. The daily mean relative humidity varies from a minimum of 40% in the month of April to a maximum of 85% in the month of July. The main agricultural crops grown during kharif season are paddy and maize and in rabi season are wheat, gram and mustard. The agriculture is mostly rainfed as only 20% irrigation is available in the area through sources other than rain and the cropping intensity is also quite low at 98%. The irrigation is received mainly by wells. Prevalence of conventional cultivation practices, characterized by conventional tillage or no tillage; low fertilizer/manure consumption and local varieties of the crops is mainly responsible for the low crop productivity in the area. All this information on the test area was obtained through secondary sources such as Directorate of Economics and Statistics, Ministry of Agriculture; Directorate of Census (data Center); DVC, Hazaribagh officials and Sadar block office of Hazaribagh district.

3.2 Generation of GIS data base For the generation of GIS data base of their spatial distribution different thematic maps such as base map, digital elevation model map, delineation of watersheds, soil group map, topographic wetness index map, stream power index map, sediment transport index map and land use map are prepared with the help of GIS based software ILWIS (3.6). A base map has been generated by digitizing the Survey of India (SOI) toposheet as reference map for all other purposes. The watershed covered by 1:50,000 scale SOI topographic maps NO.72H8 and 73E5. The watershed boundary was marked on the basis of the contours and the drainage lines available on the SOI topographic map and also using the procedure described by Jenson and Domingue (1988). 3. 2. 1 Slope map Generation of slope map, the contour map and point elevation map of study area has been used. Using the GIS based software ILWIS (3.6), the slope map for the region is generated. 3. 2. 2 Digital elevation model (DEM) The contour map (20 m interval) and spot height map of the area are merged together and a composite map having information about contours as well as spot height is formed. This combined map is further interpolated at 20-metre pixel resolution using map interpolation function available in Integrated Land and Water Information System (ILWIS) to generate a DEM of the area. Slope map was calculated using contour line map using script function available in ILWIS 3.6. 3. 2. 3 Soil erodibility factor (k) map The soil maps of the study area in the scale of 1:250,000 were traced, scanned and exported to ILWIS 3.6..The scanned maps were loaded in ILWIS 3.6. and georeferenced. Boundaries of different soil textures as per the soil conservation service soil classification system were digitized and the polygons representing various soil categories were assigned with different colours for identification. This information is then transferred on to the base map for preparation of the soil map and assign the K factor values from a Soil Survey data which is given in table 1.

Figure 1. Location map of Nagwan watershed. 3. MATERIALS AND METHODS 3.1 Data used

Table: 1. Soil texture of Nagwan watershed

Topographic maps at 1 : 50 000 scale from the Survey of India, Calcutta and soil resources data from Damodar Valley Corporation (DVC), Hazaribagh were used in this study for digitization of contour lines, construction of Digital Elevation Model (DEM). IRS-P6 (LISS IV ) satellite data having sensor scenes 23.5 m resolution(Path-105 and Row-55), with pass dates of 22 December 2012, were used for land-use/land-cover classification maps. soil map collected from National Bureau of Soil Survey and Land use Planning (NBSS&LUP), Government of India for identification of soil types of the study area.

HYDRO 2014 International

MANIT Bhopal

Map unit* 16

32

Taxonomy* Fine, mixed, hyperthermic Typic Haplustalfs Loamy, mixed, hyperthermic Lithic Ustorthents Fine loamy, mixed, hyperthermic Typic Paleustalfs Fine-loamy, mixed, hyperthermic Typic Rhodustalfs

K value 0.19

0.33

Page 72

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

*Department of Agriculture & Cane Development, Govt. of Jharkhand

(Burroughet al., 1998). The sediment transport index is defined by the equation below.

3. 2. 4 Land use map LU/LC map was developed by supervised classification techniques with maximum likelihood algorithm were used for the classification of digital data of an IRS-P6 (LISS IV ) satellite in which an area or group of pixels that belongs to one or more categories of specific land use and land cover was classified. The land uses were classified into five classes namely agriculture, water, dense forest, fallow land and urban settlement and assign the standard curve curve number (CN) value numbersfor the Indian conditions(ministry of agricultural, Govt. of india 1972).

(3) 3. 3 Priority assessment For the determination of priority of the critical erosion-prone areas in the watershed values of the parameters are normalized in a standard scale such as 0 to 1. The following equation has been used to normalize all the parameters on the 0 to 1.

3. 2. 5 Topographic Wetness Index Map The topographic wetness index (TWI), also known as the compound topographic index (CTI), is a steady state wetness index. It is commonly used to quantify topographic control on hydrological processes (Sorensen, 2006) The index is a function of both the slope and the upstream contributing area per unit width orthogonal to the flow direction. The index was designed for hill slope catenas. Accumulation numbers in flat areas will be very large, so TWI will not be a relevant variable. The index is highly correlated with several soil attributes such as horizon depth, silt percentage, organic matter content, and phosphorus (Moore, 1993) wetness index map prepared by using ILWIS 3.6 software with DEM raster map. The WI is defined as

(4) Where,

is the Normalized value of a parameter for

parameter, (1),

ia the Upper value in the standard scale

is the Lower value in the standard scale (0),

is

the Maximum value of the parameters,

is the Minimum

value of the parameters respectively and

is the Observed

value of parameters for parameter. After computing the normalized values of different parameters and then getting average of parameters for the final priority. After determining the final priority critical area it has been grouped in four classes of priority namely very high, high, moderate and low on the basis of priority ranking. 4. RESULTS AND DISCUSSION

(1)

4. 1 Development of thematic map

where As is the contributing area draining to the grid cell per unit length of a side of the grid cell (m2/m) and β is the slope angle of the cell (degrees). Slope values of zero were substituted with a value of 0.001 to avoid returning an undefined index value.

The thematic map of Nagwan watershed has been prepared using satellite image, toposheets and soil map in GIS. These are discussed below:

3. 2. 6 Stream power index (SPI)

4. 2. 1 Slope factor

Using ILWIS 3.6 software with raster map of wetness index generate the stream power index map. it is reflect the erosive power of the stream terrain (moore, 1993). it is defined as:

The factors of slope steepness (S) are in the present study area varied from 0.06 to 1.0 as shown in a Figure 2.

4. 1. 2 Topographic wetness index (TWI) (2) 3. 2. 7 Sediment transport index (STI) The Sediment Transport Index characterizes the process of erosion and deposition. it reflect erosive power of the overland flow. Unlike the length-slope factor in the Universal Soil Loss Equation (USLE) it is applicable to three-dimensional surface

HYDRO 2014 International

In the analysis found maximum area 3.88 km2 with value of TWI is 11.28. The maximum, minimum, average and standard deviation of TWI is 21.68,5.74, 13.72 and 4.27 respectively. The DEM and flow accumulation map have been used as inputs and STI map was prepared in ILWIS (3.6) for the watershed as shown an in Figure 3.

MANIT Bhopal

Page 73

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Supervised classification techniques with maximum likelihood classifier were used for the land use classification with average accuracy 89.98 %, average reliability 85.09 % and overall Accuracy 90.78 %. Seven major land use categories namely agriculture land (with & without crop and grass land), barren land, builtup land, dense forest, open forest, scrubs, water bodies were identified and then assign CN value . The land use map of the watershed is shown in Figure 7 and the land use details are shown in Table 2.

Figure 2. Slope map of the Nagwan watershed

Figure 3. TWI map of the Nagwan watershed

4. 1. 3 Stream power index The stream power index are calculated by using the eq. 2, the value are varies 0.70 to 50. SPI map was generated by ILWIS (3.6) for the watershed and shown an in Figure 4. 4. 1. 4 Soil erodibility (K) factor The soil map of the catchment area was used to prepare the digitized soil map. The predominant soil textural classes were clay loam and silty loam type, found in the watershed. Soil group of the study area shown in Figure 5 and soil erodibility value given by Table 1.

Figure 6. STI map of the Nagwan watershed

Figure 7. Land use map of nagwan watershed.

Table 2. Land use pattern of nagwan watershed. Land use

Area in km2

Agricultural land (with crop) Agricultural land (with no crop) Barren land Built up land Dense forest Grass land Open forest Shrubs Water body

27.40

Curve number 95

28.39

95

0.26 4.74 1.47 10.84 11.63 3.46 1.26

85 91 58 79 60 64 100

4. 2. Final priority map

Figure 4. SPI map of the Nagwan watershed

Figure 5. Soil erodibility map of Nagwan watershed.

4. 1. 5 Sediment transport index The sediment transport index was calculated for watersheds using the Eqn.3. These values ranged from 0.03 to 5. STI map was prepared in ILWIS (3.6) for the watershed as shown an in Figure 6.

Not all watershed contribute erosion and at same rate. the identification of erosion prone area within the watershed which contribute maximum sediment yield obviously should determine our priority to go forward appropriates conservation management strategy for maximum benefit. Also prioritization is required for proper planning and management of natural resources for catchment area treatment plan in the watershed. Determination of priority for the watersheds have been determined and normalized and give weight. The final priorities of spatially based for watershed are determined and priorities of critical erosion prone area for watersheds are grouped in four categories as shown in Table 3 and spatially depicted in Figure 8

4. 1. 6 Land use/ land cover based on curve number

HYDRO 2014 International

MANIT Bhopal

Page 74

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 8. Final priority map of the Nagwan watershed Table 3. Final priority of Nagwan watershed Priority category Low Moderate High Very high

Area (km2) 19.83 33.36 22.82 13.45

5. CONCLUSIONS The compound indices such as topographic wetness, stream power and sediment index these indices can be used to derive spatilly meaningful parameterisations of a landscape like potential for erosion. In land use classification the maximum area comes under agricultural land (62%) with minimum in barren land (< 1%). The use of GIS and remote sensing data enabled the determination of the spatial distribution paramets ( slope map, soil erodibility map, topographic wetness index map, stream power index map, sediment transport index map and land use map) and prioritization of watersheds was done. The watershed prioritization indicated that the critical erosion area under high (25.50%) and very high (15%) priority class where requires immediate attention for soil conservation treatment. Hence, remote sensing and GIS technology can be used as an alternative to conventional method of soil loss estimation and subsequent prioritization of spatilly erosion prone area of watershed for implementing soil conservation practices. The best management practices proposed for nagwan watersheds are; afforestation, trenching, bunding, stone wall fencing, brushwood check dams, earthen check dams, gabian structures and masonry structures. REFERENCES i. Ali S, Singh R (2002) Morphological And Hydrological Investigation In Hirakud Catchment For Watershed Management Planning. Journal of Soil water Conservation India 1(4): 246–256 ii. Baba SMJ, Yusof KW (2001) Modelling Soil Erosion In Tropical Environments Using Remote Sensing and Geographical Information Systems. Hydrol Sci Journal 46(1): 191–198. iii. Bakker MM, Govers G, Kosmas C, Vanacker V, Van Oost K, Rounsevell M, (2005) Soil erosion as a driver of land-use change. Agric Ecosyst Environ 105: 467–481

HYDRO 2014 International

iv. Basnyat P, Teeter LD, Lockaby BG, Flynn KM (2000) The use of remote sensing and GIS in watershed level analyses of non-point source pollution problems. For Ecological Management 128: 65–73 v. Beven KJ, Kirkby MJ (1979) A Physically Based, Variable Contributing Area Model of Basin Hydrology. Hydrological Sciences Bulletin 24: 43–69 vi. Beven KJ, Wood EF, Sivapalan M (1988) On Hydrological Heterogeneity – Catchment Morphology And Catchment Response. Journal of Hydrology 100: 353–375 vii. Burrough PA, McDonnell RA. (1998) Principal of Geographical Information System. New York: Oxford University press viii. Burt T, Butcher D (1986) Stimulation From Simulation – A Teaching Model of Hill slope Hydrology for Use on Microcomputers. Journal Geogr Higher Educ 10: 23–39 ix. Calder IR (1992) Hydrologic effects of land use change, In: Maidment DR (ed) Handbook of Hydrology. pp13.1–13.5 x. Chalam BNS, Krishnaveni M, Karmegam M (1996) Correlation of runoff with geomorphic parameters. Journal of Applied Hydrology. IX(3–4): 24–31 xi. Chaudhary RS, Sharma PD (1998) Erosion hazard assessment and treatment prioritization of Giri River catchment, NorthWestern Himalayas. Indian Journal of Soil Conservation 26(1): 6–11 xii. Chowdary VM, Ramakrishnan D, Srivastava YK, Chandran V, Jeyaram A (2009) Integrated Water Resource Development Plan for Sustainable Management of Mayurakshi Watershed, India using Remote Sensing and GIS. Water Resource Management 23: 1581–1602 xiii. Chowdary VM, Yatindranath Y, Kar S, Adiga S (2004) Modeling of Non-Point Source Pollution In Watershed Using Remote Sensing and GIS. Journal of the Indian Society of Remote Sensing 32(1): 59–73 xiv. Conosecnti C, Maggio CD, Rotigliano E (2008) GIS Analysis To Assess Landslide Susceptibility In A Fluvial Basin of NW Sicily (Italy). Geomorphology 94: 325-339 xv. Dhruva Narayana VV (1993).Soil and Water Conservation Research in India. pp 146–151 xvi. Famiglietti JS, Wood EF (1991) Evapotranspiration and Runoff From Large Land Areas – Land Surface Hydrology For Atmospheric GeneralCirculation Models. Surv Geophys 12: 179–204 xvii. Fistikoglu O, Harmancioglu NB (2002) Integration of GIS with USLE in Assessment of Soil Erosion. Water Resour Manage 16: 447–467 xviii. Goodchild MF, Parks BO, Steyaert LT (1993) Geographic Information System and Environmental Modeling [M]. Oxford University Press: Oxford. Gorge (NW Turkey). Natural Hazards 46: 323-351 xix. Holmgren P (1994) Topographic and Geochemical Influence on The Forest Site Quality, With Respect To Pinus Sylvestris and Picea Abies In Sweden. Scand Journal of Forest Research 9: 75–82 xx. Jain MK, Kothyari UC (2000) Estimation of Soil Erosion and Sediment Yield Using GIS. Journal of Hydrological Science. 45(5): 771–786 xxi. Jain SK, Kumar S, Varghese J (2001) Estimation of Soil Erosion for A Himalayan Watershed Using GIS Technique. Water Resource Management 15: 41–54 xxii. Jenson SK, Dominque JO (1988) Extracting Topographic Structures From Digital Elevation Data For GIS Analysis. Photogrammetric Engineering And Remote Sensing 54(11): 1593–1600 xxiii. Kaur R, Singh O, Srinivasan R, Das SN, Mishra K (2004) Comparison of A Subjective and A Physical Approach For Identification of Priority Areas For Soil and Water Management In A Watershed: A Case Study of Nagwan Watershed In Hazaribagh District of Jharkhand, India. Environ Model Assess 9 (2): 115–127 xxiv. Kiran VSS, Srivastava YK (2012) Check Dam Construction by Prioritization of Micro Watershed, using Morphometric Analysis as a Perspective of Remote Sensing and GIS for Simlapal Block, Bankura, W.B. Bonfring International Journal of Industrial Engineering and Management Science. Special Issue 1, Vol. 2 xxv. Kumar R, Lohani, AK, Kumar S, Chatterjee C, Nema R K (2001) GIS based morphometric analysis of Ajay river basin up to sarth gauging site of South Bihar. Journal of Applied Hydrology XIV(4): 45–54 xxvi. Mass RP, Smolen MD, Still DA (1985) Selecting critical areas for nonpoint source pollution control. Journal of Soil Water Conservation. 40 (1): 68–71 xxvii. Mishra Ashok, Kar S, Singh VP (2007) Prioritizing Structural Management by Quantifying the Effect of Land Use and Land Cover on

MANIT Bhopal

Page 75

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Watershed Runoff and Sediment Yield. Water Resource Management 21: 1899–1913 xxviii. Moore ID, Gessler PE, Nielsen GA, Petersen GA (1993) Terrain Attributes: Estimation Methods And Scale Effects.". In Jakeman AJ; Beck MB; Mcaleer M, Modelling Change In Environmental Systems. London: Wiley. pp189 - 214 xxix. Moore ID, Grayson RB, Ladson AR (1991) Digital Terrain Modeling: A Review of Hydrological, Geomorphological and Biological Applications. Journal of Hydrological Processes 5: 3–30 xxx. Moore ID, Norton TW, Williams JE (1993) Modelling Environmental Heterogeneity In Forested Landscapes. Journal of Hydrology 150: 717–747 xxxi. Naef F, Scherrer S, Weiler M (2002) A process based assessment of the potential to reduce flood runoff by land use change. Journal of Hydrology 267: 74–79 xxxii. Narayana DVV, Rambabu (1983) Estimation of soil erosion in India. Journal of Irrigation and Drainage Engineering 109(4): 419-434 xxxiii. Pandey A, Chowdary VM, Mal BC, (2004) Morphological analysis and watershed management using GIS. Hydrological Journal of India 27(3– 4): 71–84 xxxiv. Pandey A, Chowdary, VM, Mal BC, (2007) Identification of Critical Erosion Prone Areas In the Small Agricultural Watershed Using USLE, GIS and Remote Sensing. Water Resource Management 21(4): 729–746 xxxv. Quinn PF, Beven KJ, Lamb R (1995) The ln(a/tan beta) index: How To Calculate It And How To Use It Within The TOPMODEL Framework. journal of Hydrological Processes 9: 161–182 xxxvi. Renschler CS, Mannaerts C, Diekkrüeger B, (1999) Evaluating Spatial and Temporal Variability in Soil Erosion Risk—Rainfall Erosivity and Soil Loss Ratios in Andalusia, Spain Catena 34(3–4): 209–225 xxxvii. Robson A, Beven K, Neal C (1992) Towards Identifying Sources of Subsurface Flow: A Comparison Of Components Identified By A Physically Based Runoff Model and Those Determined By Chemical Mixing Techniques. Journal of Hydrological Processes 6: 199–214 xxxviii. Rodhe A, Seibert J (1999) Wetland Occurrence In Relation to Topography: A Test of Topographic Indices as Moisture Indicators. Agricultural Forest Meteorology 98–99, 325–340 xxxix. Seibert J, Bishop KH, Nyberg L (1997) A Test of TOPMODEL‘s Ability to Predict Spatially Distributed Groundwater Levels. Journal of Hydrological Processes 11: 1131–1144 xl. Sekhar KR, Rao BV (2002) Evaluation of Sediment Yield By Using Remote Sensing and GIS: A Case Study From The Phulang Vagu Watershed, Nizamaba District (A.P.). International Journal of Remote Sensing 23(20): 4449–4509 xli. Sharma T, Satya Kiran PV, Singh TP, Trivedi AV, Navalgund RR (2001) Hydrologic response of a watershed to land use changes: a remote sensing and GIS approach. International Journal of Remote Sensing 22 (11): 2095–2108 xlii. Shinde Vipul, Tiwari KN, Singh Manjushree (2010) Prioritization of micro watersheds on the basis of soil erosion hazard using remote sensing and geographic information system. International Journal of Water Resources and Environmental Engineering 2(3): 130-136. http://www.academicjournals.org/ijwree xliii. Singh G, Babu R, Pratap N, Bhushan LS, Abrol IP (1992) Soil erosion rate in India. Journal of Soil and Water Conservation 47(1): 97–99 xliv. Singh RK, Bhatt CM, Prasad VH (2003) Morphological Study Of A Watershed Using Remote Sensing and GIS Techniques. Journal of Hydrology 26(1–2): 55–66 xlv. Sivapalan M, Wood EF (1987) A Multidimensional Model of Non stationary Space-Time Rainfall at the Catchment Scale. Water Resource Research 23: 1289–1299 xlvi. Sivapalan M, Wood EF, Beven KJ (1990) On Hydrologic Similarity 3 A Dimensionless Flood Frequency Model Using A Generalized Geomorphologic Unit Hydrograph And Partial Area Runoff Generation. Water Resource Research 26: 43–58 xlvii. Sorensen R, Zinko U, Seibert J (2006) On the Calculation of The Topographic Wetness Index: Evaluation of Different Methods Based on Field Observations. Hydrology and Earth System Sciences 10: 101–112 xlviii. Sorensen R, Zinko U, Seibert J (2006) On The Calculation of The Topographic Wetness Index: Evaluation of Different Methods Based On Field Observations. Hydrology And Earth System Sciences 10: 101–112 xlix. Suresh M, Sudhakar S, Tiwari KN, Chowdary VM (2004) Prioritization Of Watersheds Using Morphometric Parameters and

HYDRO 2014 International

Assessment of Surface Water Potential Using Remote Sensing. Journal of Indian Society Remote Sensing 32(3): 249–259 l. Tarboton DG (1997) A New Method For The Determination of Flow Directions and Upslope Areas In Grid Digital Elevation Models, Water Resource Research 33: 309–319 li. Tideman EM (1996) Watershed management, guidelines for Indian conditions. New Delhi, Omega Scientific Publ lii. Tripathi MP, Raghuwanshi NS, Rao GP (2006) Effect of watershed subdivision on simulation of water balance components. Journal of Hydrological Processes 20 (5): 1137–1156 liii. Whelan MJ, Gandolfi C (2002) Modelling of Spatial Controls on Denitrification At The Landscape Scale. Journal of Hydrological Processes 16: 1437–1450 liv. White JD, Running SW (1994) Testing Scale-Dependent Assumptions In Regional Ecosystem Simulations. Journal of Vegetation Science 5: 687–702 lv. Wilson JP, Gallant JC (2000) Digital terrain analysis, In Wilson JP, Gallant JC (eds)Terrain Analysis. John Wiley & Sons, New York 1-27 lvi. Wise S, (2000) Assessing the Quality For Hydrological Applications of Digital Elevation Models lvii. Derived From Contours. Journal of Hydrological Processes 14: 1909–1929 lviii. Wolock DM, McCabe GJ (1995) Comparison of Single and Multiple Flow Direction Algorithms For Computing Topographic Parameters In Topmodel. Water Resource Research 31: 1315–1324 lix. Yang D, Kanae S, Oki T, Koikel T, Musiake T (2003) Global potential soil erosion with reference to land use and climate change. Journal of Hydrological Processes 17(14): 2913–2928 lx. Zinko, U., Seibert, J., Dynesius, M., and Nilsson, C. (2005). Plant Species Numbers Predicted by A Topography Based Groundwater-Flow Index. Journal of Ecosystems. 8: 430–441.

Minimization of Conveyance Losses for Nashik Left Bank Canal [NLBC] by Closed Conduit Irrigation [CCI] Gayatri R. Gadekar1, Dr. Sunil Kute2 Dr. N. J. Sathe3, 1

ME Hydraulics, Civil Engineering, Sinhgad College of Engineering, Pune, University of Pune. 2 Professor, Civil Engineering, K. K. Wagh Institute of Engineering and Research, Nashik, University of Pune. 3 Assistant Professor, Civil Engineering, Sinhgad College of Engineering,Pune, University of Pune. E-mail:[email protected] 2 , [email protected] [email protected]

MANIT Bhopal

Page 76

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

ABSTRACT:The present paper focuses on the minimization of conveyance losses for Nashik Left Bank Canal [NLBC] originating from Gangapur dam of Nashik District of Maharashtra state located at 20° 38‟ Lattitude and 73° 19‟ Longitude. This is an unlined canal of 64 km stretch having design discharge 8.92 cumecss. NLBC has conveyance losses of about 57% and 55% in rabi and hot weather season, respectively. To minimize these conveyance losses of NLBC, Closed Conduit Irrigation [CCI] system has been suggested and analysed in this paper. This CCI system will consist of a conduit line of 1.82m diameter of Glass Fibre Reinforced Pipe [GFRP] running as an open channel i.e. under atmospheric pressure for total 64 km length of the canal with longitudinal slope of 1:4000. The CCI system of NLBC with free board of 0.5m has 3.02m of head losses for the entire length of the canal. For the Full Supply Depth [y] of 1.32m in GFRP of , the Froude number [Fr] of flow is 0.4149; which indicates Subcritical flow for CCI. The CCI for NLBC will save of about 15.55 Mm3 of irrigation water which constitutes a part of conveyance loss for the present Open Canal Irrigation [OCI] system of NLBC for the entire canal length.

2.

CASE STUDY OF NASHIK LEFT BANK CANAL [NLBC]

2.1 Study area Nashik district of Maharashtra state is one of the leading districts in the field of agriculture. The new experiments and use of advanced technology have empowered the farmers to increase export of agro based products. Gangapur Dam is most important and the oldest earthen dam in Nashik. It was constructed in 1965 on Godavari River. Two canals namely Nashik Right Bank Canal [NRBC] and Nashik Left Bank Canal [NLBC] take off from the dam. The GRBC is closed due to high civilization in the area. The present paper focuses on the case study of Nashik Left Bank Canal of Nashik district, Maharashtra state which is located at 20° 38‟ Lattitude and 73° 19‟ Longitude. The reach of this canal is 64 km which is running open to atmosphere. The alignment of canal and its command area is shown in Figure 1.

Keywords- Open Canal Irrigation [OCI], Conveyance losses, Hydraulic Design of Conduit, Closed Conduit Irrigation [CCI], Glass Fibre Reinforced Pipe [GFRP]. 1.

INTRODUCTION

1.1 Open canals are used to convey the water from storage reservoir to the agricultural land for irrigation. Water has to travel from its head to fulfill the needs of agriculture; irrigation channels with poor maintenance causes heavy losses during its conveyance phase. It is observed that the losses due to evaporation, infiltration, percolation and water thefts in open canal reduce the efficiency and yield of irrigation. Therefore, it is necessary to check these conveyance losses in case of irrigation canals. Discharge of water through the canals is utilized for irrigation purposes only. During its passage from canal head up to the agriculture land, there are various types of losses occurring; these losses are termed as conveyance losses. Major amount of irrigation water is lost during this conveyance phase.

Figure 1: Command Area of Nashik Left Bank Canal [NLBC] Source: Nasik Irrigation Departmen

1.2 Many researchers have tried to quantify these conveyance losses. Kolhe, P. S. (2012), in his paper has suggested the Pressurized Pipe Distribution Network [PDN] for Nagthana-II for optimal utilization of Irrigation Water. Ghazaw,Y.M. (2010) has developed the design charts and computer programme to facilitate the design of optimal water loss section. Burt, C.M. et.al. (2008) have given the solution for reduction in canal seepage by in place compaction of canal banks and bed. Swamee, P.K. et.al. (2002) have given the minimum water loss canal sections that have been obtained using the explicit equations for seepage loss and evaporation equation for flowing channels.

HYDRO 2014 International

MANIT Bhopal

TABLE I: General Information of NLBC [6] Sr. No.

Description

1

Cross-Section

2 3 4 5

Shape Canal Bed Level [CBL] Design Discharge Chainage [Location]

Data 2.44 m X 2.44 m Trapezoidal 589.94 m 8.92 cumecss 801.83 m

6

Bed Width

3m

7

Bed Gradient

1:4000

8 9 10 11 12 13

Length Full Supply Depth Type Canal Top Width Depth of Canal

64 km 1.65 m Unlined 4.67m 2m

Side Slopes

1:0.5

Page 77

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

2.2 Design of NLBC

2.3 Crop water requirement for NLBC

The data for the case study is collected from Nashik Irrigation Department [NID]. The general information of NLBC is given in Table-I wherein Table -II represents the crop pattern and crop water requirement details for NLBC.

In Table II, the yearly crop water requirement is calculated as V= 24.5 Mm3. This is the crop water requirement for base period of 72 days. Therefore, Discharge in cumecs corresponding to volume of 24.5 Mm3 = [24.5 * 106] / [72 *24*60*60] = 3.96 cumecs. From Table II, it is clear that the crop water requirement in cumecs for NLBC is 3.96 cumecs. 2.4 Conveyance losses and Diameter of Conduit of NLBC The conveyance losses for NLBC are calculated by applying the general water budget equation to the open canal for rabi and hot weather season. The values of the conveyance losses and efficiency for the rabi and hot weather season are represented in Table – III. TABLE II: Crop Pattern and Crop Water Requirement Details for NLBC [6] Season

Crop Pattern

Area [Ha]

Rabi [54 Days]

Grapes Sugarcane Vegetables Wheat Others

965.10 137.04 86.02 81.03 304.03

Water Requirement Mm3 Cumecs 5.76 0.93 4.201 0.68 1.28 0.21 0.888 0.14 3.641 0.59

Grapes

1125.07

7.88

1.27

Sugarcane

162.15

0.85

0.14

Hot Weather [18 Days]

A= 2860.44 Ha

V= 24.5 Mm3

W= 3.96 Cumecs

It can be seen from the Table III, that there are huge conveyance losses for the NLBC. NLBC has yearly conveyance loss of 15.55 Mm3, in which rabi season has the conveyance loss of 10.69 Mm3 and conveyance loss of 4.862 Mm3 has observed in hot weather season. It is clear from the Table III that the conveyance loss for NLBC is more than 55 %, which is very huge. The efficiency is calculated from the details of conveyance losses for rabi and hot weather season. The efficiency of NLBC is 43.02 % for rabi season wherein 44.71% for hot weather season. S r. N o.

HYDRO 2014 International

MANIT Bhopal

Seas on

Area under Crop [Ha]

No. of Days water suppli ed

Quan tity of water suppli ed at head of canal [Mm3 ]

Quan tity of water used [Mm3 ]

Conve yance Losse s Mm3

Efficiency [%] %

Page 78

International Journal of Engineering Research Issue Special3 1

Yea rly

2860. 44

72

27.55

12.00

15.55

56.44

2

Rab i

1573. 22

54

18.76 1

8.073

10.69

56.98

3

Hot Wea ther

1287. 22

18

8.793

3.931

4.862

55.29

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 43 .5 6 43 .0 2 44 .7 1

= 1.82 – 0.5 = 1.32 m

Table III: Actual Details of Conveyance Losses and Efficiency for NLBC [6] Therefore, the actual discharge [QActual] required in NLBC can be calculated by considering the designed discharge [QD] and efficiency of NLBC. Figure 2: Closed Circular Conduit [GRP]

∴ Actual Discharge [QActual] = Designed discharge [QD] X Efficiency [ ] = 8.92 X 0.4356 = 3.886cumecs.

2.6 Velocity and Type of Flow of NLBC

This discharge is to be supplied to the area of 2860.44ha which is under crop.

1) For closed conduit irrigation of GRP 1.82m,

From Table I, the longitudinal slope[S] of the canal is 1:4000. The conduit to be used for NLBC irrigation has to be designed for the Actual Discharge [QActual] of 3.86 cumecs. The conduit which will be used for NLBC‟s Closed Conduit Irrigation [CCI] needs to be durable and strong. Hence, for NLBC, the Glass Fibre Reinforced Pipe [GRP] is recommended as it has working life of about 70 years, its C value is 140, it is light weight and the Glass Fibres structure increases its strength to a great extent [7]. The diameter [D] of GRP conduit section is obtained with the relation of discharge [Q], area [A] and velocity [V]. For the velocity of flow, Chezy‟s formula is used.

Velocity [V] of flow through GRP 1.82m [5] = C√RS V = C √ {[D/4] S} V = 140 √ {[1.82/4] * [1/4000]} V = 1.493 m/s Froude Number [Fr] of flow for GRP ϕ1.82m [5] = V/ √ [gy] Fr = 1.493 / √ [9.81 * 1.32] Fr = 0.4149 < 1 Subcritical flow. 2) For an open canal flow in NLBC,

Therefore, Actual Discharge [Qactual] = Area [A] X Velocity [V] = {[П / 4] * D2} X {C X √[R*S]} Substituting the values of Qactual [3.886 cumecs], C of GRP [140] and longitudinal slope [1:4000], and simplifying above equation, the diameter of GRP for NLBC is calculated which comes out to be 1.82m. This is the diameter of equivalent closed conduit section for NLBC for supplying the discharge of 3.886cumecs. Diameter of Equivalent GRP for CCI of NLBC = 1.82m Figure 3: Open canal cross-section of NLBC

2.5 Freeboard for CCI of NLBC But, the closed conduit irrigation [CCI] which is suggested in this paper is an open channel flow i.e. the flow inside the conduit will be running under atmospheric pressure. Hence sufficient free board should be available in a GRP of 1.82m diameter. The free board for the discharge range of 1-5 cumecs is assumed as 0.5 m. Full supply depth [y] through conduit of 1.82m = Diameter of GRP – Freeboard

HYDRO 2014 International

Velocity [V] of flow through open NLBC [5] = C√RS = C √ [A/T] S = 40 √ [6.0866/4.38] * [1/4000] = 0.7456 m/s Froude Number [Fr] of flow for open channel NLBC [5] = V/ √ [gy] = 0.7456 / √ [9.81 * 1.65]

MANIT Bhopal

Page 79

International Journal of Engineering Research Issue Special3 = 0.3058 < 1

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Hot Weather [18 days]

Subcritical flow.

For open canal NLBC, the velocity is 0.7456 m/s whereas for GRP 1.82m, it is 1.493 m/s. The Froude Number for open channel NLBC is 0.3058 whereas for GRP 1.82m it is 0.4149.

3.

4.862

3.13

3.31

1592.08

ADVANTAGES OF CLOSED CONDUIT IRRIGATION [CCI] OVER OPEN CANAL IRRIGATION [OCI] FOR NLBC

2.7 Head losses of NLBC for GRP In a Closed Conduit Flow through ϕ 1.82m GRP, there will be the Head losses. 1) For the head loss due to friction [5], hf = [fLV2] / [2gD] Friction factor for GRP ϕ1.82m [f] = 2.13 X 10-3 [7] Now, hf = [2.13 X 10-3 * 64000* 1.49532] / [2*9.81*1.82] = 8.535m Friction loss per meter [hf] =8.535/6400 = 1.33 X 10-4m Considering the total length of the canal i.e. 64 km, the friction head loss is very less. 2) Head loss at entry of GRP ϕ1.82m [5] = 0.5 [V2/2g] = 0.5 [1.49532 / (2*9.81)] = 0.057m 3) Head loss at exit of GRP ϕ1.82m [5] = [V2/2g] = [1.49532 / (2*9.81)] = 0.114m

NLBC has conveyance losses of about 15.55 Mm3. Due to the conversion of open canal into closed conduit section; these losses of water in each season will be minimized. This can be considered as the saving of water. Thus, the water saved can be utilized for improving duty. II. As CCI increases the duty of water by 2.29 cumecs and 3.31 cumecs for rabi and hot weather season,respectively. Hence, more area can be brought under irrigation for NLBC. III. Conveyance losses have resulted into decreased efficiency of canal ranging from 57% in rabi and 55% in hot weather season. Hence, the use of CCI will save 15.55 Mm3 of water, thus increasing the efficiency of NLBC. IV. NLBC sites have problems like breeding of mosquitos, fly nuisance, water logging and salinity which can be stopped if CCI system is implemented I.

4. CONCLUSIONS 4) Head loss at entry of GRP ϕ1.82m for each branch [5] = 0.5 [V2/2g] = 0.5 [1.49532 / (2*9.81)] = 0.057m Head loss at entry of GRP ϕ1.82m for 50 branches = 50 X 0.057 = 2.85 m Therefore, total head lost in GRP ϕ1.82m is calculated as, HLoss = 1.33 X 10-4 + 0.057 + 0.114+2.85 HLoss = 3.02m The head losses are calculated for NLBC‟s CCI by considering the loss at entry and exit of conduit, friction losses and loss at entry of each branch. The head loss is 3.02m for the entire 64km stretch of NLBC. 2.8 Saving in water of NLBC by CCI Due to conversion of OCI into CCI, conveyance losses of 15.55 Mm3 are saved, which can be used for improving the duty of water. The following table IV shows the details of improvement in the duty. Table IV: Details of Improvement of duty Extra Extra Water Saved water land that Season made can be Mm3 Cumecs available irrigated [cumecs] [ha] Rabi [54 2083.2 10.69 2.29 2.29 days]

HYDRO 2014 International

It is revealed from the hydraulic analysis, that the conversion of open canal into circular closed conduit is technically feasible and there is impact of water saving of 10.69 Mm3 for rabi season and 4.862 mm3 for hot weather season for improving irrigation potential by reducing the conveyance losses. In addition to saving in water, there is 50% increase in the velocity of flow because of increased C-value of GRP. A case study of Nashik Left Bank Canal [NLBC] of length 64 km shows that 57% losses during rabi season and 55% of conveyance losses during hot weather can be stopped by adopting this system. Thus, the net saving of 15.55 Mm3 can be achieved by adopting CCI. The capital cost of such conversion is justified on the basis of water saving of 15.55 Mm3 for the 64 km stretch of NLBC and increased irrigation potential of 2083.2 ha and 1592.08 Ha for Rabi and Hot Weather season respectively. Hence, it is recommended to use CCI in place OCI to save the valuable water. ACKNOWLEDGEMENT A paper of this nature calls for intellectual nourishment, professional help and encouragement from many quarters. I would like to extend my sincere gratitude towards the Nashik Irrigation Department (NID) and Graphite India Ltd. for providing me with the necessary authentic data required for the paper.

MANIT Bhopal

Page 80

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

REFERENCES i. Kolhe P.S. (2012) ―Optimal Utilization of Irrigation Water by Use of Pipe Distribution Network (PDN) Instead Of Canal Distribution Network (CDN) In Command Area‖, India Water Week 2012, New Delhi. ii. Ghazaw Y. M. (2010), ―Design Charts of Optimal Canal Section for Minimum Water Loss.‖ Journal of Engineering and Computer Sciences, Qassim University, Vol. 3, No. 2, pp. 73-95 iii. Burt C. M. et. al. (Nov 2008) ―Canal Seepage Reduction by Soil Compaction‖, IA Technical Conference, ITRC Paper No. P 08-002. iv. Prabhata K. Swamee, Govinda C. Mishra, Bhagu R. Chahar (2002), ―Design of Minimum Water-Loss Canal Sections‖, Journal of Hydraulic Research, Vol. 40, 2002, No. 2. v. Garg S. K., (2005), ―Irrigation Engineering and Hydraulic Structures‖ 19th Edition, Khanna Publishers, Delhi, India. Pp 1141,1162. vi. Annual Report (June 2013): ―Annual Water Account of Major and Medium Projects‖, Nashik Irrigation Department. Pp. 7 vii. IS 12709: 2009, ―Glass Fibre Reinforced Plastics (GRP) Pipes, Joints and Fittings for Use for Potable Water Supply — Specification.‖

methods which can give reasonably good accuracy. In view of the recent development in data acquisitions and techniques to model soil water crop interaction, selection of appropriate model has become very important step. The objective of the study is to review all the methods available to estimate first reference evapotranspiration based on climate. For estimating reference evapotranspiration (ETref) various empirical methods, radiation based equations and methods based on radiation as well as dynamic factors are discussed. The paper suggests points to be considered for selection of appropriate method. ASCE Standardized PM Equation and dual crop coefficient provide precise estimation of ET under varied climates. Keywords:Evapotranspiration,Reference Penman-Monteith

evapotranspiration,

1.0 INTRODUCTION:

BIOGRAPHIES Ms. Gayatri R. Gadekar is pursuing her post graduation in Hydraulics from Sinhgad College of Engineering. Her research area includes water resources engineering.

Dr.Sunil Kute is currently Professor of Civil Engineering. Also, he is Chairman, Board of Studies (Civil Engineering) and member of Academic Council and Senate of University of Pune .He has experience of 23 years in teaching, administration and research. He is Ph.D. guide of University of Pune and North Maharashtra University .His 60 research papers are published in journals and conferences .His research areas are structural engineering and water resources engineering .Currently, 6 students are pursuing Ph.D. under his guidance. Dr.N. J. Sathe is currently M. E. Hydraulics coordinator in civil engineering department of Sinhgad College of Engineering, Pune. . Also, he is Chairman of Geoinformatics and Engineering Geology subjects of University of Pune. He is member of Board of Studies of Shivaji University. He has experience of 15 years in teaching and research. He is Ph.D. guide of University of Pune. His 37 research papers are published in journals and conferences. His research areas are Geoinformatics, Engineering Geology and Water Resources Engineering.

The irrigated agriculture uses large chunk of water, thus a big responsibility lies with irrigation managers to efficiently use the water. The large quantity of water is lost as evaporation and transpiration from the fields. Evaporation and transpiration usually happen at the same time and is hard to separate the two processes. To match the irrigation supply with demand, estimation of the evapotranspiration is required to be done with appropriate methods which can give reasonably good accuracy. FAO presented two publications to describe various model for estimating crop water requirements (Doorenbos and Pruitt, 1977; Allen et al., 1998). In view of the recent development in data acquisitions and techniques to model soil water crop interaction selection of appropriate model needs the understanding of capabilities and limitations of each available model. This paper reviews most of the widely used methods available to estimate reference evapotranspiration based on climate data. The paper also suggests points to be considered for selection of appropriate method. 2.0 EVAPOTRANSPIRATION:

Methods for Estimation of Crop Evapotranspiration Using Climate Data: A Review Gopal H. Bhatti 1, H.M. Patel2 Research Scholar and Associate Professor, Civil Engineering Department, Faculty of Technology & Engg, The M.S University of Baroda, Vadodara. 390 001, Gujarat, India. 2 Head and Professor, Civil Engineering Department, Faculty of Technology & Engg, The M.S University of Baroda, Vadodara. 390 001, Gujarat, India. Email: [email protected], [email protected] 1

ABSTRACT: As water being the limited resource, its optimum utilization is of great concern in irrigated agricultural sector as it is the largest user in most part of the world. To match the irrigation supply with demand, estimation of the evapotranspiration is required to be done with appropriate

HYDRO 2014 International

Evapotranspiration is the combined process through which water is lost by evaporation from the soil surface and from the crop by transpiration. The crops require a fixed quantity of water to meet the water losses through evapotranspiration for bumper crop production under standard conditions. The crop evapotranspiration (ETc) under standard conditions refers to crops that are disease-free, well fertilized and are grown in large fields under optimum soil water with excellent management and environmental conditions, so as to attain full production under the given climatic conditions Allen et al. (1998). ETc measurement is not easy and requires sophisticated, expensive equipment and trained research personnel with varied range of systems. Lanthaler (2004) reported measuring evapotranspiration using lysimeter. Evapotranspiration data could be obtained from varied range of measurement systems which included lysimeters, eddy covariance, Bowen ratio, scintillometry, sap flow, satellite-based remote sensing, direct modeling and soil water balance such as gravimetric, neutron

MANIT Bhopal

Page 81

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

probes, electromagnetic types of soil sensors, time domain reflectometry etc. Phene et al.,(1990); Cammalleri et al. (2010); Allen et al., (2011); and Evett et al., (2012). Direct measurement techniques are not feasible for estimating evapotranspiration in large irrigated area. Mostly they are used for research purposes by trained personnel. Evapotranspiration is generally estimated by using different methods which requires measurements of climatological parameters.

especially in the higher latitudes. Radiation method would be more reliable than Blaney Criddle in equatorial regions, on small islands, or at high altitudes even if measured sunshine or cloudiness data were available (Doorenbos and Pruitt, 1977). The empirical and temperature based methods have been used for estimating evapotranspiration for longer periods i.e. monthly or weekly. 4.0 RADIATION METHODS:

3.0 EMPIRICAL METHODS:

AND

TEMPERATURE

BASED

3.1 Pan Evaporation method Evaporation pan provided measurement of integrated effect of temperature, radiation, wind and humidity on evaporation from a particular open water surface. Evaporation pan data were utilized to convert evaporation from free-water surface with pan coefficient to estimate potential evapotranspiration (Allen et al, 1998). Incorrect accounting for pan environment and local climate could cause errors in estimation of crop water use upto plus or minus 40 percent (Cuenca 1989). However pan evaporation has been one of the widely used methods due to simplicity and minimum data requirements. 3.2 Temperature based methods Hedke (1924) developed a method for estimating valley consumptive use based on “heat available” defined as degreedays (number of days multiplied to temperature). Blaney and Morin (1942); Lowry and Johnson (1942) developed a method for roughly calculating seasonal consumptive use. Blaney-Morin term included relative humidity term which was useful index for measuring vapour transport component of evaporation process. Lowry and Johnson method was developed based only on temperature. Thornthwaite (1948) developed a method with an assumption of an exponential relationship existing between mean monthly consumptive use and mean monthly temperature. The formula did not take into account the wind effect which could be an important factor at many places. Blaney – Criddle (1950 and 1962) developed method for areas where available climatic data covered air temperature data only. The mean air temperature was considered to be a good measure of solar radiation. It was considered one of the popular procedures for estimating potential evapotranspiration due to its simplicity and readily available temperature data. In this method monthly consumptive use crop coefficient k had to be developed for each and every crop under the climatic condition of particular area. Phelan (1962) developed a procedure for adjusting monthly k values as a function of air temperature which is known as SCS Blaney Criddle method. Doorenbos and Pruitt (1977) suggested including other meteorological variables by using specific data or general estimates of sunshine hours, relative humidity and wind speed to have an improved estimate of potential evapotranspiration which is known as FAO Blaney-Criddle method. Blaney Criddle method had a limitation of selecting percent of daytime hours instead of solar radiation as an index of solar energy. It is observed that daytime hours obtained from sunshine tables did not properly accounted for solar angle effects

HYDRO 2014 International

Evapotranspiration occurs only when energy is available and hence estimation of solar radiation can give better estimation of ET by using Energy Balance equation which includes Rn (radiation from sun and sky), G (heat to ground), H (heat to air). Makkink (1957) proposed a formula for estimating ET from air temperature and sunshine or cloudiness or solar radiation. The Makkink equation was the base of the subsequent FAO 24 Radiation method. Turc (1961) developed a formula based on ten-day mean air temperature and solar radiation. The Turc equation had limitation to be applied only if Tmean > 10 . Jensen-Haise (1963); Hargreaves-Samani (1985) developed a relationships between temperature and solar radiation using the observations of consumptive use of water. In spite of sufficient energy available, ET could be less due to aerodynamic resistance in form of Wind speed and Humidity as for the atmosphere‟s ability to remove water vapour, an “Aerodynamic” strength also plays a crucial role. 5.0 COMBINATION METHODS: Penman (1948, 1963) utilized Bowen ratio principle and derived a “combination equation” by coalescing two terms, one (radiation) term which was for the energy required to uphold evaporation from open water surface and second (wind and humidity) term for the atmosphere‟s ability to remove water vapour, an “aerodynamic” strength. Penman formula could be used for estimation of potential evapotranspiration by using a reflection coefficient (r) value of 0.25 for most crops. Monteith (1965, 1981) extended Penman‟s basic concept to plants and cropped areas by introducing resistance factors, including surface resistance and aerodynamic resistance by clearly identifying the reliance of transpiration on canopy controls known as Penman-Monteith evapotranspiration equation. Priestly and Taylor (1972) proposed a well- known simplification of Penman‟s equation for humid environments where the aerodynamic term was put at a constant value (0.26) of the energy term. Doorenbos and Pruitt (1975, 1977) proposed a modified Penman method with a revised wind function term and an adjustment for mean climatic data for estimating reasonably accurately the reference crop ET by giving tables and graphs to facilitate computation. Wright (1982) modified the original Penman equation and adapted 1982 Kimberly-Penman equation. Kizer et al., (1990) developed hourly evapotranspiration prediction model by calibrating the Penman equation for an alfalfa reference crop. Allen et al., (1998) used the equation on hourly basis with the rs term having a constant value of 70 s m-1 throughout the day and night. They recommended FAO-56 Penman Monteith method as the sole

MANIT Bhopal

Page 82

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

standard method for determining reference evapotranspiration in all climates, especially when there was availability of data. Allen, (2000) developed REF-ET program which provided standardized reference evapotranspiration calculations in different time steps for more than 15 methods commonly used such as Pan Evaporation, Temperature methods, Radiation methods, Combination methods. Allen (2002) compared the seasonal ET obtained by reference evapotranspiration estimated by ASCE standardized Penman-Monteith with 1982 Kimberly Penman and found the differences to be low. Walter et al., (2005) developed a standardized reference evapotranspiration equation which could be applied to two types of reference surfaces alfalfa and clipped grass for daily and hourly calculation time step. The ASCE Standardized Reference Evapotranspiration Equation based on FAO-56 PenmanMonteith equation was developed by ASCE-EWRI task committee with aforesaid purpose. The equation is also recognized as ASCE-EWRI standardized Penman-Monteith equation. Allen et al. (2006) reviewed the functioning of FAOPM method using surface resistance parameter rs = 70 sm-1 in hourly time step while using a constant rs = 50 sm-1 during day and rs = 200sm-1 during night for hourly period. The various widely used equations discussed above are depicted in Table 1. Values for Cn and Cd in FAO-PM and ASCE-EWRI standardized PM equations are given in Table 2. 6.0 COMPARISON METHODOLOGIES

STUDIES

OF

Table1. Equation and Measured data required for ET o prediction for various methods. Name of Prediction Method

Equation

Data used

Empirical and Temperature Methods Hedke (1924)

Heat available = Temp x days

FEW

Many comparison studies have been carried out worldwide regarding the functioning of various methods to estimate reference ET. Each method has its own strengths and weakness under the particular set of conditions. Here only few studies have been discussed to just give a brief idea about their functioning. Hatfield and Allen (1996) compared ET estimates under deficient water supplies with Priestly-Taylor and PenmanMonteith equations. Penman-Monteith gave more consistent results, while Priestly-Taylor overestimated ETc. Dodds et al., (2005) reviewed various methodologies to estimate ETref. (i) Evaporation Class-A pan tended to be 7-8% higher than the locally calibrated ETo values for evaporation rates < 10mm day1 and for values > 10mm day-1the pan overestimated the values by upto 30%. (ii) Two methods of Penman combination Equation with certain variation in it were compared with lysimeter. a). Kohler-Parmele variation was with a purpose of calculating the long wave radiation from the soil-plant system using the air temperature instead of evaporating surface temperature, b) Morton gave an iterative variation of the Penman equation to calculate a suitable evaporating surface temperature; where both methods performed well. Berengena and Gavilan (2005) compared measured ETo using lysimeter with estimated ETref in a highly advective semi arid environment. They found that locally adjusted Penman and ASCE-PM gave the best results; followed by FAO-PM. Hargreaves equation under predicted for high ET values and the Priestly-Taylor equation was found to be too sensitive to advection and the values improved only after the application of correction of the Jury and Tanner. Er. Raki et al.(2010) compared three empirical methods Makkink, Priestley-Taylor and Hargreaves-Samani for

HYDRO 2014 International

computing reference evapotranspiration (ETo ) to those with FAO Penman-Monteith in semi arid climate. Hargreaves equation tended to under estimate ETo upto twenty percent for daily periods. Makkink and Priestly& Taylor methods clearly under estimated the values of ETo during dry periods in comparison to FAO-PM model, since values of α = 1.26 and Cm = 0.61 that used are suitable for humid conditions. Artificial Neural Networks (ANNs) could be a useful tool to estimate reference evapotranspiration as a function of climatic elements Kumar et al., 2002; Jothiprakash et al., 2002. Chauhan and Shrivastava, (2012) reported that ANNs performance when compared with lysimeter measured values were better than those obtained from Penman-Monteith method for estimation of ET ref. Ojha and Bhakar (2012) carried out the comparison between daily ETref estimated by Penman Monteith (PM) method and that of estimated by ANNs and found the ANNs results encouraging.

T

Blaney and Morin (1942)

PET = rf(0.45 Ta+8)(520 – R1.31)/ 100

T,SS,RH

Lowry and Johnson (1942)

CU = 0.00185 HE+ 10.4

T

Thornthwaite (1948)

T,SS

Blaney and Criddle (1945,1962)

T,SS

SCS-Blaney Criddle Phelan(1962)

T,SS ;

US Weather Bureau Class A pan

RH,E,W

FAO-Blaney Criddle Doorenbos & Pruitt (1977)

T,SS,RH,W

Temperature and Radiation Methods FAO radiation (Makkink, 1957)

T,SS,RH,W,Rs

Turc(1961)

T,RH,Rs,

Jensen and Haise (1963)

T, Rs

Hargreaves and Samani (1985)

T, Rs,/(SS1,Ra)

MANIT Bhopal

Page 83

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 1

= mean air temperature (o F and

).

Combination Methods

-1

extraterrestrial radiation (mm d ) ,

Penman (1948,1963)

T,SS,RH,W,Rs

Penman-Monteith method (Monteith 1965)

T, RH, Rn

Priestly Taylor(1972)

T, RH, Rn

and

C),

=

= maximum and

minimum daily air temperature difference. -2

o

= evaporative

-1

latent heat flux (MJ m day ), = slope of saturated vapour o -1 pressure curve ( kPa C ), Rn= net radiation flux (MJ m-2 day-1), G = sensible heat flux into the soil (MJ m-2 d-1), = psychrometric constant ( kPa o C-1), = vapour transport of flux (mm d-1).

= density of air ( kg m-3), -1 o

= specific heat of

-1

Modified Penman method, Doorenbos and Pruitt (1975,1977)

T, W, Rn

moisture ( J kg C ), VPD = vapour pressure deficit, = canopy surface resistance and aerodynamic resistance ( sm-1). W = temperature related weighting factor, = wind related

1982 Kimberly Penman Method, Wright (1982)

T, RH, W, Rn

Penman equation for hourly ET for alfalfa, Kizer et al.,(1990)

T, RH, W, Rn

function, = difference between saturation vapour pressure at mean air temperature and the mean actual vapour pressure of air (both in mbar), c = adjustment factor to compensate for the effect of day & night weather conditions. ETr = reference evapotranspiration (MJ m-2d-1), = wind function. LE = mean hourly latent heat flux (Wm-2), U2 = wind speed at 2m (km h-1), = coefficients. = saturation vapour

;

pressure (k Pa), FAO-56 PenmanMonteith Method, Allen et al.,(1998)

T, RH, W, Rn

ASCE-EWRI standardized -PM method, Walter et al., (2005)

T RH, W, Rn

and = numerator constants and denominator constants respectively that change with reference type and calculation time step . Table 2. Values for Cn and Cd in Equation for the FAO-PM and ASCE-EWRI standardized PM equations (as reported in Allen et al., (1998) and ASCE-EWRI (2005))

T = Temperature, SS = Sun shine hours, RH = Relative Humidity, W = Wind, E = Evaporation, Rs= Solar Radiation, Rn = Net Radiation.. PET= Potential evapotranspiration (mm day-1), Ta= Mean monthly temperature in o C, R= Mean monthly Relative humidity, rf = ratio of monthly to annual radiation. CU= Annual consumptive use (inches), HE= Effective heat, in degree days above 32o F. e = unadjusted potential ET (cm/month)( month of 30 days each and 12 hrs daytime), t= mean air temperature(o C), I = annual or seasonal heat index, α= an empirical exponent. = monthly consumptive use factor, T = mean monthly temperature (o F), p = monthly per cent of total daytime hrs of the year. ET= Seasonal crop water requirements (inches), = monthly Blaney Criddle coefficient , = monthly consumptive use factor , = mean temperature for month i, (o F). ETo= Reference evapotranspiration (mm day-1), Kp= Pan coefficient, Epan = Pan evaporation (mm day-1). , b = climatic calibration coefficients , = mean daily percentage of total annual daytime hours, = mean daily temperature in o C over the month considered. = adjustment factor depending on mean humidity and daytime wind conditions, W = function of the temperature & altitude, Rs= solar radiation (mm day-1). = coefficient depending mean relative humidity, Rs= solar radiation (MJ m-2 day-1), = latent heat of vaporization (MJ kg-

HYDRO 2014 International

= mean actual vapour pressure (k Pa),

Method

Calculation time step

Cn

Cd

FAO-PM (ETo) &

24-h

900

0.34c

Hourly

37

0.24/0.96a

24-h

1600

0.38

Hourly

66

0.25/1.7a

ASCE-PM (ETo) ASCE-PM (ETr)b

a

The first value for daytime periods (when Rn>0) and the second value is for night time. b ETr is reference ET from 0.5m tall alfalfa. c The Cd= 0.34 is now recommended to be changed to 0.24 for daytime and 0.96 for night time for hourly or shorter time steps. 7.0 DISCUSSION Irrigation is supplied to compensate the moisture deficit in soil occurred due to evapotranspiration. Hence precise estimation of ET is very much required. The factors affecting potential ET are radiation, temperature, relative humidity and wind speed. The measurement techniques just provide the point value of moisture content and it cannot be used to estimate the crop water requirement of large irrigated area with varied climate. The

MANIT Bhopal

Page 84

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

empirical and temperature based methods performed suitably under specific climatic and agronomic conditions for which they were originally developed and could not be used under different conditions, other than that for which they were developed. Transferring these to other regions led to either under/over estimation causing substantial errors. The radiation methods which considered the radiant energy provides better estimates in humid climate but were less precise in advective conditions in arid and semi arid climates, and hence it needed adjustment or correction. The combination methods take into account the radiant energy term as well as aerodynamic term the ability to remove water vapour hence it improved upon the ET estimation. FAO-PM was considered the sole standard method in case all the climate data are available. ASCE-PM method was standardized for different reference crops and also for different calculation time step. The ASCE- PM standardized reference ET equation is widely accepted for precise estimation of ET. This method can provide important tool for developing decision support system for irrigation scheduling. The relationship of ET and climate parameters is complex and hence many researchers have resorted to data modelling such as ANN technique. REFERENCES: i. Allen, R. (2002). Evapotranspiration: The FAO 56 Dual Crop Coefficient Method and Accuracy of predictions for Project - wide Evapotranspiration. International meeting on Advances in Drip/Micro Irrigation. ii. Allen, R. G. (2000). Using the FAO-56 dual crop coefficient method over an irrigated region as part of an evapotranspiration intercomparison study. Journal of Hydrology, 27-41. iii. Allen, R. G., Pereira, L. S., Howell, T. A., & Jensen, M. E. (2011). Evapotranspiration information reporting: I. Factors governing measurement accuracy. Agricultural Water Management, 98, 899-920. iv. Allen, R. G., Pereira, L. S., Raes, D., & Smith, M. (1998). Crop evapotranspiration - Guidelines for computing crop water requirements - FAO Irrigation and Drainage paper 56. Rome: United Nations Food and Agriculture Organization. v. Allen, R. G., Pruitt, W. O., Wright, J. L., Howell, T. A., Ventura, F., Snyder, R., . . . Elliott, R. (2006). A recommendation on standardized surface resistance for hourly calculation of reference ETo FAO56 Penman-Monteith method. Agricultural Water Management, 81, 122. vi. Berengena, J., & Gavilan, P. (2005). Reference Evapotranspiration Estimation in a Highly Advective Semiarid Environment. Journal of Irrigation and Drainage Engineering, 147-163. vii. Blaney, H. F., & Criddle, W. D. (1950). Determining Water Requirements in Irrigated Areas from Climatological and Irrigation Data. In: Jensen, M. E. Historical evolution of ET estimating methods, A Century of progress. CSU/ARS Evapotranspiration Workshop, Fort Collins, CO, 12- Mar-2010, p. 4. viii. Blaney, H. F., & Criddle, W. D. (1962). Determining Consumptive Use and Water Requirements. In Jensen, M. E. Historical evolution of ET estimating methods, A Century of progress. CSU/ARS Workshop Evapotranspiration Workshop, Fort Collins, CO, 12-Mar-2010, p. 4. ix. Blaney, H. F., & Morin, K. V. (1942). Evaporation and consumptive use of water formulas. . American Geophysics Union Transaction, 76-82. x. Cammalleri, C., Agnese, C., Ciraolo, G., Minacapilli, M., Provenzano, G., & Rallo, G. (2010). Actual evapotransporation assessment by means of a coupled energy/hydrologic balance model: Validation over an olive grove by means of scintillometry and measurements of soil water contents. Journal of Hydrology 392, 70-82. xi. Chauhan, S., & Shrivastava, R. K. (2012). Estimating Reference Evapotranspiration using Neural computing technique. Journal of Indian Water Resources Society, Vol. 32, No. 1-2,, 22-32.

HYDRO 2014 International

xii. Cuenca, R. H. (1989). Irrigation System Design an Engineering Approach, p.122. New Jersey: Prentice-Hall, Englewood Cliffs. xiii. Dodds, P. E., Meyer, W. S., & Barton, A. (April, 2005). A Review of Methods to Estimate Irrigated Reference Crop Evapotranspiration across Australia. Griffith: Cooperative Research Centre for Irrigation Futures, Technical Report No.04/05, . xiv. Doorenbos, J., & Pruitt, W. O. (1975). Guidelines for predicting Crop Water Requirements, FAO- 24. In: Michael, A. M. Irrigation theory and practice second edition, 2008. p. 497. Cochin: Vikas Publishing House, New Delhi. xv. Doorenbos, J., & Pruitt, W. O. (1977). Guidelines for predicting Crop Water Requirements, FAO- 24 (Revised). In: Michael, A. M. Irrigation theory and practice second. p.497 edition, 2008. Cochin: Vikas Publishing House Pvt. Ltd. New Delhi. xvi. Er-Raki, S., Chehbouni, A., Khabba, S., Simonneaux, V., Jarlan, L., Ouldbba, A., . . . Allen, R. (2010). Assessment of reference evapotranspiration methods in semi-arid regions: Can weather forecast data be used as alternate of ground meteorological parameters? Journal of Arid Environments, 74, 1587-1596. xvii. Evett, S. R., Schwartz, R. C., Howell, T. A., Baumhardt, R. L., & Copeland, K. S. (2012). Can weighing lysimeter ET represent surrounding field ET well enough to test flux station measurements of daily and sub-daily ET? Advances in Water Resources, Article in Press. xviii. Hargreaves, G. H., & Samani, Z. A. (1985). Reference crop evaporation from temperature. . Applied Engineering in Agriculture 1(2), 96-99. xix. Hatfield, J. L., & Allen, R. G. (1996). Evapotranspiration estimates under deficient water supplies. Journal of Irrigation and Drainage Engineering, 301-308. xx. Hedke, C. R. (1924). Consumptive use of water by crops. In: Jensen, M. E. Historical Evolution of ET estimating methods, A century of progress. CSU/ARS Evapotranspiration Workshop, Fort Collins, CO, 12- Mar- 2010, 1-17. xxi. Jothiprakash, V., Ramachandran, M. R., & Shanmuganathan, P. (2002). Artificial neural network model for estimation of REF-ET. Journal-CV 83(2), 17-20. xxii. Kizer, M. A., Elliott, R. L., & Stone, J. F. (1990). Hourly ET Model calibration with Eddy Flux and Energy Balance Data. Journal of Irrigation and Drainage Engineering Vol. 116, No. 2, 172-182. xxiii. Kumar, M., Raghuwanshi, N. S., Singh, R., Wallender, W. W., & Pruitt, W. O. (2002). Estimating evapotranspiration using artificial neural network. Journal of Irrigationand Drainage Division., 224-233. xxiv. Lanthaler, C. (2004). Lysimeter Stations and Soil Hydrology Measuring sites in Europe – Purpose, Equipment, Research results, Future Developments. Graz: Department for Water Resources Management, Hydrogeology and Geophysics. xxv. Lowry, R. L., & Johnson, A. F. (1942). Consumptive use of water for agriculture. Transaction American Society of Civil Engineers 107, 1243-1302. xxvi. Makkink, G. F. (1957). Testing the Penman Formula by Means of Lysimeters. In Jensen, M. E. Historical evolution of ET estimating methods, A Century of progress. CSU/ARS Evapotranspiration Workshop, Fort Collins, CO, 12- Mar-2010, p. 8. xxvii. Montieth, J. L. (1965). Evaporation and Environment. In: Michael, A. M. Irrigation theory and practice second edition, 2008. p. 499. Cochin: Vikas Publishing House Pvt. Ltd. New Delhi. xxviii. Montieth, J. L. (1981). Evaporation and Surface Temperature. Quarterly Journal of Royal Meteorological Society 107, 127. xxix. Ojha, S., & Bhakar, S. (2012). Estimation of evapotranspiration for wheat crop using artificial neural network. Journal of Indian Water Resources Society, Vol. 32, No. 1-2,, 13-21. xxx. Penman, H. L. (1948). Natural evaporation from open water, bare soil, and grass. Proceedings Royal Society of London, A193, (pp. 120-146). xxxi. Penman, H. L. (1963). Vegetation and hydrology. In: Farhani, H. J., Howell, T. A., Shuttleworth, W. J. and Bausch, W. C. Evapotranspiration: Progress in measurement and modeling in agriculture. American Society of Agricultural and Biological Engineers. Vol. 50(5), 2007, p. 1627. xxxii. Phelan, J. T. (1962). Estimating monthly "k" values for the Blaney-Criddle formula. In: Jensen, M. E. Historical evolution of ET

MANIT Bhopal

Page 85

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

estimating methods, A Century of progress. CSU/ARS Evapotranspiration Workshop, Fort Collins, Co, 12-Mar-2010, p.4. xxxiii. Phene, C. J., Reginato, R. J., Itier, B., & Tanner, B. R. (1990). Sensing Irrigation needs. In: Farhani, H. J., Howell, T. A., Shuttleworth, W. J. and Bausch, W. C. Evapotranspiration: Progress in measurement and modeling in agriculture. American Society of Agricutural and Biological Engineers. Vol. 50(5). 2007, p.1629. xxxiv. Priestly, C. B., & Taylor, R. J. (1972). On the assessment of surface heat flux and evaporation using large-scale parameters. . Monthly Weather Review, 81-92. xxxv. Thornthwaite, C. W. (1948). An approach towards a rational classification of climate. The Geographical Review, Vol.38, N0.1, 55-94. xxxvi. Turc, L. (1961). Estimation of irrigation water requirements, potential ET : A simple climatic formula evolved up todate.In: Jensen, M. E. Historical evolution of ET estimating methods, A Century of progress. CSU/ARS Evapotranspiration Workshop, Fort Collins, CO, 12-Mar-2010, p.8. xxxvii. Walter, I. A., Allen, R. G., Elliott, R., Itenfisu, D., Brown, P., Jensen, M. E., . . . Wright, J. L. (2005). The ASCE Standardized Reference Evapotranspiration Equation. USA: Task Committee on Standardization of Reference Evapotranspiration, Environmental and Water Resources Institute of the American Society of Civil Engineers. xxxviii. Wright, J. L. (1982). New evapotranspiration crop coefficients. . Journal of Irrigation and Drainage Division. 108 (IR1), 5774.

Estimation of Deep Percolation from Rice Paddy Field Using Lysimeter Experiments on Sandy Loam Soil Hatiye, Samuel D.1, K.S.Hari Prasad 2, C.S.P. Ojha 2 and G.S. Kaushika 3 1,3 Ph.D. Scholar, Civil Engineering Department, IIT Roorkee, 247667 Roorkee, India; 2Professor Department of Civil Engineering, IIT Roorkee, 247667 Roorkee, India. ABSTRACT: In this study, variation and characteristics of deep percolation from irrigated rice paddy field using drainage type lysimeter set up has been presented. The water intensive lowland rice paddy has been grown from July to November 2013. Water balance components including irrigation size, rainfall, soil moisture and deep percolation were monitored on daily bases. It has been observed that quite a large volume of water is returned as deep percolation flow as physically demonstrated from twin lysimeter measurements. We employed a simple tipping bucket type water balance model to validate the experimental data. The deep percolation monitored on daily bases does not agree with the model computed value, however it agrees well on an extended time interval in an order of seven days (weekly bases). On average more than 80% of the input volume of water goes on the account of deep percolation in non puddled, continuously irrigated rice field. 1

Corresponding Author: email: [email protected]; phone +918266802124

HYDRO 2014 International

Our study proves that locally constructed lysimeters could effectively be utilized in water balance study of a cropped area when used in combination with root zone soil moisture monitoring devices and can contribute to the further water resources management of an irrigated field. We deduce from this study that deep percolation process is one of the most important factors lowering surface method irrigation efficiency in general and rice paddy fields in particular in course textured soils. We recommend, revisit of irrigation scheduling options besides the already practiced water saving options in water intensive crops for better utilization of water resources. Key words: Deep percolation, Lysimeter experiment, Rice paddy, Root zone depletion, Water balance model 1. INTRODUCTION Deep percolation phenomena from frequently irrigated fields such as paddy rice seriously diminish irrigation efficiency, jeopardise proper water management and minimize water productivity. This is quite sound in coarse textured soils where water holding capacity is relatively less. Seepage and percolation losses of water are major reasons behind the poor water productivity in wetland rice (Patil et al. 2011). Percolation loss of water from irrigated field is not only reducing irrigation efficiency but also becoming a haphazard to an environment by carrying agriculture-based chemicals to the surrounding water bodies, especially to the groundwater aquifer systems (Tafteh and Sepaskhah 2012). Various studies were conducted to estimate deep percolation from irrigated fields. Large volume of deep percolation loss could exist during the continuous flooding operation of rice paddy, even in under puddled conditions (Kukal and Aggarwal 2002; Bouman et al. 2007; Yadav et al. 2011). Bouman and et al. (2007) reported that around 70% of input water could go for percolation loss when groundwater depth is equal to or more than 2m. Yadav et al. (2011) observed that, about 81% of water added was drained beyond the root zone (0–60 cm) from continuously flooded rice field. Many factors influence percolation phenomena through the bottom of the crop root zone. Ponding size, water table depth, evapotranspiration, antecedent soil moisture condition, soil texture and structure characteristics, shrinkage behaviour of soil and biotic activities in soil root zone, irrigation size and time, climatic condition, crop type and characteristics, water management and agronomic practices, puddling intensity and depth, etc… (Kukal and Aggarwal 2002; Bouman 2007; Bethune et al. 2008; Selle et al. 2011). Sizable efforts have been made so far to reduce deep percolation from rice fields: alternate wetting and drying (AWD) ( de Vries, et al. 2010; Bouman et al. 2007; Belder et al. 2004; ), aerobic rice (Nie et al. 2012), delayed application of continuous flooding (Dunn and Gaydon 2011), puddling (Kuakal and Aggarwal 2002; Kukal and Sidhu 2004). However, consideration and effort to reduce deep percolation under non puddled rice paddy field was not dealt significantly. There are various ways available to quantify and estimate deep percolation. Drainage type lysimeters are considered to be the most important facilities, at field level, to measure percolation.

MANIT Bhopal

Page 86

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

However, lysimeters are criticized to be costly to install, maintain and operate; and so they often are used singly such that adequate replication of measurements is not possible (Evett et al. 2012; Bethune et al. 2008; Hillel 2004). Apart from the direct measurement of percolation using lysimeter set up, various models have also been developed to estimate deep percolation from agricultural areas. Deep percolation, the water that passes below the crop root zone, is usually calculated based on the conventional water balance equation (Peng et al. 2012; Bethune et al. 2008; Huang et al. 2003). Estimation of deep percolation from rice paddy has not commonly been determined using drainage type lysimeters. The reason may be due to the fact that drainage volume at the bottom of water intensive crops is quite large which may not be easy for continuous monitoring. This particular study aims, therefore, to quantify deep percolation in continuously monitored irrigated fields of paddy rice to understand and characterize deep percolation. The experimental data is planned to be evaluated using simple water balance model after FAO. In the course of the exercise, we try to examine the major influencing factors of deep percolation from cropped area employing drainage type lysimeters in sandy loam soils. 2.

MATERIALS AND METHODS

The study site is located in the Utterakhand state of India, a field experimental plot situated in Department of Civil Engineering, IIT Roorkee in the geometric grid of 77 o53‟52” East Longitude and 29o52‟00‟‟ at an average altitude of 274m above mean sea level. The area experiences hot summer season with monsoon rainfall and cold winter. The monthly average maximum temperature of the study area is recorded in the range of 19.33 (January) to 37.73oC (May) and monthly average minimum temperature in the range of 7.2(January) to 25.6 oC (July) according to the data from National Institute of Hydrology (NIH) at Roorkee. The average relative humidity runs from 52.2% (May) to 89.7% (January). The average annual daily sunshine duration is 7.7hrs. The average annual rainfall of Roorkee is 1060 mm out of which almost 80% is recorded during the monsoon season (June to September).

experimental conditions have been maintained inside and outside the lysimeters throughout the growing period of the crop. Twenty one days old seedlings were transplanted in a soaked field. Basal doses of zinc sulphate, superphosphate and urea (N fertilizer) were applied in two equal instalments during transplanting and 6 weeks after transplanting. Weed control has been undertaken manually by hand removing all the weeds from field three times during the growth period of the crop. Irrigation water size of 20mm to 100mm has been applied to the paddy field during the growth stages except the final late stages when the crop was matured to harvest. The soil physical and hydraulic characteristics have been determined in the laboratory for three representative spots of the plot and replicate depths from 0 to 140cm following standard procedures. The soil physical properties determined are indicated in the table below (table 1). Irrigation water was applied for a specific area by measuring discharge and calculating time required to provide a predetermined depth of water. The soil moisture status was monitored by using soil moisture probe (Profile Probe-2; Delta T Devices, Cambridge) through access tubes installed both inside and outside the lysimeters. The profile probe sensor which is connected to HH2 meter provides soil moisture content data at 10, 20, 30, 40, 60 and 100cm depths. It enables to measure the soil moisture content in volumetric bases for different soil types ranging from clayey to sandy soils with accuracy between +0.04 (after soil specific calibration) and +0.06(after generalized soil calibration in normal soils). The soil moisture was measured on daily bases and before and after irrigation or rainfall whenever these events took place. Deep percolation was measured twice in a day at bottom of the lysimeters early in the morning (07:00 a.m.) and evening (around 07:00 p.m.). The lysimeter rim was kept 10cm above the ground to avoid run-on or runoff. Tipping buckets in access caisson hall were used to collect the drainage water. Climatic data (temperature, relative humidity, pan evaporation, wind speed and rainfall) for the growth period of the crop was obtained from nearby metrological station, National Institute of Hydrology (NIH), India located at distance of 0.8 kilometres from the experimental site. Table 1. Soil physical characteristics of the experimental plot

The field experiment consisted of growing paddy rice ((Oryza Sativa L.) , var. Surbati Basmati) from July 23 (day of transplanting) to 02 November (day of harvest) of the 2013 kharif season. The area of lysimeters is 1m2 having a depth of 1.5m repacked soil monolith of the experimental field. The construction of the lysimeters took place in 2007 and hence they are considered to replicate the surrounding root zone soil environment. The soil monolith is a repacked soil material consisting of the upper 1.3m filled with a sandy loam textured soil, moderately homogeneous throughout the profile, characterized by an organic content of 1.1 to 1.2%. The bottom 0.08m was filled with a very course gravel of size more than 3cm diameter overlain by 0.12m thick gravel of about 2cm in diameter. This bottom arrangement allows drainage towards imbedded perforated pipes which carry percolating water towards tipping buckets (Shankar 2007) (Fig 1). The same

HYDRO 2014 International

2 3

Soil depth (below GL2),cm

Bulk density( g/cm3)

San d (%)

Silt (%)

Cl ay ( % )

Soil Class (USDA)3

Satura ted Water conten t, θsat

1.58

Particl e densit y (g/cm3 ) 2.55

0-30

73. 40

22.7 0

Sandy Loam

0.38

30-60

1.55

2.57

66. 89

28.3 9

Sandy Loam

0.40

60-80

1.54

2.56

68. 57

26.5 4

Sandy Loam

0.40

80-100

1.54

2.58

69. 10

26.5 4

2. 9 6 4. 0 1 4. 3 3 3. 8

Sandy Loam

0.40

Ground level USDA=United States Dep‟t of Agriculture

MANIT Bhopal

Page 87

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

4 100-140

1.59

2.62

68. 01

27.3 8

4. 5 8

Sandy Loam

0.39

when the root depth has grown deeper, water contents measured at deeper depths has been taken into account in computation of soil moisture deficit besides the soil water content in the surface layer. Consequently, the deep percolation is computed as:-

DPi  10R j  ( i 1   i )  Pi  I i  ETci  Ri

3. MODEL DESCRIPTION The soil water balance is the concept, derived from the law of conservation of mass, used in quite many studies dealing with water flow in the soil root zone, solute transport, groundwater flow and recharge, etc.…. It is dealing with quantification and analysis of each inflow and outflow components while accounting for storage in the system environment (Kim, et al. 2009; Chien and Fang 2012; Peng et al. 2012). A FAO based simple tipping bucket soil water balance model (Allen et al. 1998) is used in this study to test the validity of field experimental observation. The lysimeter water balance can be given by (Hillel 2004):-

Di  Di 1  Pi  I i  ETci  DPi  Ri (1)

(4) Irrigation and precipitation are usually inputs to the field and obtained from actual field measurements. Evapotranspiration could be calculated from various models. Among the various methods developed so far, the FAO Penman Monthieth approach has been applied in this study. Evapotranspiration for standard conditions, Etc, is estimated by incorporating a crop coefficient, Kc,

ETc  K c  ETo (5) Where ETc for standard conditions assumes hypothetical conditions where there is no short of water, actively growing disease free crops in an extensive area. However, this imaginary condition seldom occurs in a practical field condition. Detailed procedures to estimate Kc and attached parameters are given by FAO paper 56 (Allen et al., 1998), Rallo et al., 2012)

Where D (mm) = root zone moisture depletion; P (mm) = precipitation; I(mm) = applied irrigation; ETc (mm) = actual evapotranspiration; DP (mm) = deep percolation of water moving out of the root zone; Ri(mm) is surface runoff ; i and i-1 are, respectively, considered to be the current and previous time steps (days in this study). The soil moisture deficit in the root zone is obtained from monitored water contents at respective depths. It is usually referenced with the field capacity of a given soil and may be given by:-

Di  10 R j   fc   i 

Figure-1. Lysimeter set up details (All dimensions are in mm)

(2) Where θi is the soil moisture content (%) in the root zone depth Rj (m) at the end of day i ; θfc is the soil moisture content at field capacity (%). The deep percolation is computed taking into account the root growth of the crops. The field observed root length has been interpolated for each day of the crop growth period and used as an input in the computation of the soil water balance model. In particular, the root growth has an effect on the soil moisture deficit as portrayed in the following equation.

Di  Di 1  10R j  ( fc   i )  R j  ( fc   i 1 ) Di  Di 1  10R j  ( i 1   i )

Runoff component of the water balance in lysimeter studies is often neglected since it is either minimal or controlled in such a way that there exists no run-on and run-off. If the top level of the lysimeter is constructed a slightly above the ground elevation, surface water inflow or outflow could be eliminated. However, in certain torrential storms it is advisable to consider runoff from a lysimeter since water could overflow the lysimeter rim. Runon in our experimental site did not occur since the field surrounding is constructed of earthen bunds covered with plastic sheets. Therefore, surface runoff in our experimental field has been considered only when rainfall magnitude overflows above the lysimeter rim level according to the following algorithm:

(3) Where

is the average root depth (m) in the time interval i and

i-1 and other terms are as defined earlier. If the depth of root zone is small, < 10cm, as in the early growth stages of the crops, the soil moisture content on the top layer is considered;

HYDRO 2014 International

(6) Where Ri= runoff generated (mm); P i= rainfall (mm) and Lrh = the lysimeter rim height measured from ground surface inside the lysimeter (mm).

MANIT Bhopal

Page 88

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

4. RESULTS AND DISCUSSIONS 4.1. Diurnal Deep Percolation Figure 2 shows the measured deep percolation for the day and night times. During the period of the experimental run, the observed deep percolation on day time (measured around 7:00 p.m. in the evening) is lower than the deep percolation occurred during night time (measured around 7:00 a.m. in the morning). Although we could clearly observe such variation of day and night time percolation which took place due to the effect of evapotranspiration during day time, the comparison of evapotranspiration with deep percolation shows poor correlation. The correlation coefficient between daily deep percolation and actual evapotranspiration is nearly 0.13 (not shown here). The less dependence of percolation on ET refers that deep percolation is more dependent on some other factors such as input water volume (Selle et al. 2011; Bethune et al. 2008; Ochoa et al. 2007; Smith et al. 2005), soil hydraulic characteristics (Smith et al. 2005), final infiltration rate (Selle et al. 2011; Bethune, et al. 2008), Groundwater depth (Bouman and et.al. 2007; Bethune, et al. 2008), antecedent root zone soil moisture condition (Ochoa et al, 2007), irrigation management techniques (Smith et al. 2005); crop type and cropping pattern (Smith et al. 2005). The input depth of water, antecedent soil moisture conditions, groundwater depth and irrigation management techniques have eventually influenced the deep percolation in the experimental field.During the crop period, the deep percolation event was observed to follow the input water pattern. Occurrence of intense storms caused high deep percolation than event irrigations (fig. 4). Irrigation could be controlled to minimize deep percolation but it is hardly possible to manage percolation from storm rainfall. The process of puddling would enhance the soil water retention capacity (Kuakal and Aggarwal 2002; Kukal and Sidhu 2004). However, effectiveness of this technique in ensuring lateral flow through side bunds and deep percolation thereof is being debatable. The antecedent soil moisture condition is obviously another factor which could characterize the deep percolation. Whenever, the soil is at or above field capacity, the input water added would contribute to deep percolation balance. Since, the wetting event in this particular study was frequent (fig 4); the soil was remained near field capacity for most of the growing period and hence large deep percolation. Generally, the deep percolation showed a decreasing trend from the monsoon season (JulySeptember) to late season stage of the crop season (OctoberNovember). The decreasing trend would be due to the coupled effects of reduced irrigation sizes, frequency and the ending of monsoon rainfall season.The performance of the two lysimeters in metering deep percolation has also been investigated. It has been seen that the observed amount of deep percolation from both lysimeters is fairly similar showing the repacked soil monolith exhibit the same property in both lysimeters particularly during the non-storm periods. During

HYDRO 2014 International

Figure 2. Deep percolation at lysimeter 1(L1) for day (broken) and night times (solid) lines storm periods, the lysimeters were observed to demonstrate variations in allowing percolation (fig. 2 and 3). This may be due to the fact that the lysimeters depict differences in preferential flow which is significant during rainy days. 4.2. Model Percolation

Predicted

and

Measured

Deep

The model predicted and measured deep percolation is shown below (fig. 6) for various time steps. The deep percolation computed using the simple water balance model on daily time step poorly agrees with the field measured daily deep percolation. This would be due to the inherent nature of the model in which it assumes the deep percolation to occur on the day of event irrigation or rainfall. However, in practical field situations deep percolation could take place starting on the day of triggered irrigation or rainfall occurrence and in the next consecutive days (Liu et al. 2006; Peng et al. 2012). Peng et al. 2012, has indicated that percolation would cease after seven days (a weekly time step). Liu et al. (2006) has shown that deep percolation would follow a sort of power law function. Apart from that, till the percolating water finds way out to tipping buckets in to the

Figure 3. Relation of Deep Percolation (DP) Measured in the two lysimeters

MANIT Bhopal

Page 89

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 4. Input (Irrigation-IRR and Rainfall) water and deep percolation atmosphere, there is a time lag between incidence of irrigation or rainfall and drainage of water. This time lag could not be perfectly one day as assumed in the simple tipping bucket model. In actual situations, construction of lysimeters could only be done for specific depth of the root zone, mostly considering the maximum depth of root lengths of major crops in an area. Therefore, whenever the root zone of a particular crop is less than the outlet level of the lysimeter, we would expect certain effects of storage which reasonably cause time lag for percolation to occur. The important thing is that the ability of the lysimeter set up in monitoring deep percolation beyond the root zone.

The statistics with regard to the lumped time step deep percolation is shown in the table below (table 2). Deep percolation computed on weekly (7 days) time step showed very good agreement with the measured cumulative deep percolation. This shows that consideration of smaller time steps (in order of few days or less than a day) would yield erroneous results particularly in computing the deep percolation component of the water balance from drainage type lysimeters. In fact, the storage effect of the lysimeter monolith could not be disregarded. However, it is only possible to construct drainage type lysimeters whose outlets are located at certain fixed position below the root zone (usually below 1m depth from ground level). Table 2. Statistical parameters for measured and computed deep percolation Time interval, days

C O D

C A O R V E

1

0 . 1 1 3

1 . 0. 0 06 2

0 . 6 9

0 . 0. 3 11 7

0 . 9

0 0. . 01 3

After observation of the deviations between model predicted and field observed deep percolation besides the temporal characteristics of measured percolation, we extended time interval from daily time step to 5, 7 and 10 days interval to apply the water balance. The results of this time lumping exercise, commencing from the day of transplanting to crop harvest, showed that there is a good L1=Lysimeter 1; L2 =lysimeter 2; cum = cumulative; ETA=Actual Evapotranspiration

Figure 5. Cumulative Water Balance Components during the crop period

5

agreement between measured and predicted deep percolation values.

7

HYDRO 2014 International

MANIT Bhopal

Page 90

International Journal of Engineering Research Issue Special3

10

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

1

2

0 . 7 3

0 0. . 04 2 4

COD=Coefficient of determination; COV= Coefficient of Variation; ARE=Average relative error The statistics shows that there occurs a good agreement between model predicted and measured values of deep percolation when applied on extended time steps. Thus we deduce from these results that locally constructed drainage type lysimeters could provide detailed information in characterizing deep percolation phenomena in an irrigated farm. Deep percolation measurements could be undertaken in the time intervals of 5 to 10 days in further investigation of deep percolation researches unless drainage outlets are installed at very shallow depths which may, however, not be practicable.

(a) Time Interval = 1day Interval = 5days

(b) Time

(c) Time Interval = 7days Interval = 10days

(d) Time

Deep percolation from rice field has been investigated. The deep percolation varies mainly in response to the input water depth and frequency of application/occurrence when groundwater table is assumed deep. Intense and continuous storms particularly caused high percolation rate and depth owing to the saturated antecedent moisture conditions during and after these incidences. Evapotranspiration is observed to have some influence on deep percolation as daily measurements reveal, although there is a weak correlation between evapotranspiration and deep percolation. The FAO based simple tipping bucket water balance model poorly simulates the daily deep percolation measured at drainage type lysimeters. However, the model better predicts the cumulative deep percolation on lumped time step of the order of 7 days (weekly time interval). Overall, in this study it has been investigated that deep percolation is the most important process in the water balance of irrigated paddy field diminishing irrigation efficiency. Comparable volumes of deep percolation from rice cultivated areas have been reported earlier, even under puddled root zone conditions. Therefore, it is advisable to seek alternative irrigation scheduling strategies to minimize deep percolation and hence increase irrigation efficiency and further enhance the water resource utilization of a region. REFERENCES

Figure 6. (a-d) Measured (solid lines) and model predicted (Dots) deep percolation The overall share of deep percolation in the water balance is quite high. We observed that there occurred above 80% of the volume of water input goes as deep percolation. The total amount of input water during the growing season was 3078.1mm and the total measured deep percolation was 2506.5mm (fig. 5) while the model computed deep percolation was 2646.60mm. This shows that quite a significant volume of water is percolated during frequent irrigation of the paddy growth period, although it could be quite possible to reduce the amount of water input by appropriate irrigation scheduling.

i. Allen RG, Pereira LS, Raes D, Smith M, (1998) Crop evapotranspiration: Guidelines for Computing Crop Water Requirements. Food and Agriculture Organization of the United Nations, Rome. ii. Belder P., Bouman BAM., Cabangon R, Lu G, Quilang, EJP, Li YH, Spiertz, JHJ., Tuong, TP (2004) Effect of water-saving irrigation on rice yield and water use in typical lowland conditions in Asia. Agricutural Water Management 65 (3), 193–210. iii. Bethune MG., Selle B, Wang QJ (2008) Understanding and predicting deep percolation under surface irrigation. Water Resources Research, 44, W12430, doi: 10.1029/2007WR006380. iv. Bouman BAM, Feng L, Tuong TP, Lu G, Wang H, Feng Y (2007). Exploring options to grow rice using less water in northern China using a modelling approach II. Quantifying yield, water balance components, and water productivity. Agricultural Water Management 88, 13-33. v. Bouman BAM, Lampayan RM, Tuong, TP (2007) Water Management in Irrigated Rice: Coping with Water Scarcity. International Rice Research Institute, Los Banos. vi. Chien CP, Fang WT (2012) Modelling irrigation return flow for the return flow reuse system in paddy fields. Paddy Water Environment 10,187196. vii. de Vries ME, Rodenburg J, Bado BV, Sow A, Leffelaar PA, Giller KE (2010). Rice production with less irrigation water is possible in a Sahelian environment. Field Crops Research 116, 154–164. viii. Dunn W, Gaydon DS (2011) Rice growth, yield and water productivity responses to irrigation scheduling prior to the delayed application of continuous flooding in south-east Australia. Agricultural Water Management 98, 1799 -1807. ix. Evett SR, Schwartz RC, Howell TA, Baumhardt, RL, Copeland, KS (2012) Can weighing lysimeter ET represent surrounding field ET well enough to test flux station measurements of daily and sub-daily ET? Advances in Water Resources 50, 79–90. x. Hillel D (2004) Introduction to Environmental Soil Physics. Elsevier Academic Press, Amsterdam xi. Kim HK, Jang TI, Im SJ, Park SW (2009) Estimation of irrigation return flow from paddy fields considering the soil moisture. Agricultural Water Management 96 , 875-882. xii. Kukal SS, Aggarwal GC (2002) Percolation losses of water in relation to puddling intensity and depth in a sandy loam rice (Oryza sativa) field. Agricultural Water Management 57, 49-59.

5. CONCLUSIONS

HYDRO 2014 International

MANIT Bhopal

Page 91

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

xiii. Kukal SS, Sidhu AS (2004) Percolation losses of water in relation to pre-puddling tillage and puddling intensity in a puddled sandy loam rice (Oryza sativa L.) field. Soil & Tillage Research 78, 1-8. xiv. Liu Y, Pereira LS, Fernando RM (2006) Fluxes through the bottom boundary of the root zone in silty soils: Parametric approaches to estimate groundwater contribution and percolation. Agricultural Water Management 84 , 27– 40. xv. Nie L, Peng S Chen, M, Shah F., Huang JK., Cui K., Xiang, J (2012). Aerobic rice for water-saving agriculture. A review. Agronomy and Sustainable Development 32, 411-418. xvi. Ochoa CG, Fernald AG, Guldan SJ, Shukla MK. (2007) Deep Percolation and its Effects on Shallow Groundwater Level Rise Following Flood Irrigation. American Society of Agricultural and Biological Engineers ISSN 0001−2351. Vol. 50(1): 73−81. xvii. Patil MD, Das BS, Bhadoria PBS, (2011) A simple bund plugging technique for improving water productivity in wetland rice. Soil & Tillage Research 112, 66–75. xviii. Peng W, Song X., Han D, Zhang Y, Zhang B (2012) Determination of evaporation, transpiration and deep percolation of summer corn and winter wheat after irrigation. Agricultural Water Management 105, 32- 37 xix. Rallo G, Agnese C, Minacapilli M, Provenzano G (2012) Comparison of SWAP and FAO Agro-Hydrological Models to Schedule Irrigation of Wine Grapes. ASCE Journal, Irrigation and Drainage Engineering 138:581-591. xx. Selle B, Minasny B, Bethune M, Thayalakumaran T, Chandra S (2011) Applicability of Richards' equation models to predict deep percolation under surface irrigation. Geoderma 160, 569–578 xxi. Shankar V (2007) Modelling of Moisture uptake by plants: a Ph.D, Thesis. IIT Roorkee, Department of Civil Engineering, Roorkee, India. xxii. Smith RJ, Raine SR. Minkevich J (2005) Irrigation application efficiency and deep drainage under surface irrigated cotton. Agricultural Water Management 71,117-30. xxiii. Tafteh A., Sepaskhah AR (2012) Application of HYDRUS-1D model for simulating water and nitrate leaching from continuous and alternate furrow irrigated rapeseed and maize fields. Agricultural Water Management 113, 1929. xxiv. Yadav S, Li T., Humphreys E., Gill G, Kukal SS ( 2011). Evaluation and application of ORYZA2000 for irrigation scheduling of puddled transplanted rice in North West India. Field Crops Research, 122 104–117.

Reservoir Modelling in Bearma Basin by Using Mike Basin Shikha Sachan1*, T. Thomas2, R.M. Singh3, Pushpendra Kumar4 1 Department of Farm Engineering, Banaras Hindu University Varanasi 221005, India * E-mail : [email protected] ABSTRACT: MIKE BASIN is an integrate water resource management and planning computer model that integrates GIS with water resource modelling (DHI, 2006). The Bundelkhand region in Central India has been in the grip of severe drought in the last decade mainly due to poor, limited and untimely rainfall and its high variability coupled with improper water resources development and management. Bearma river is one of the important tributary of river Ken lies completely in Madhya Pradesh. In Bearma basin, Irrigation planning and management has been carried out for drought year (2002). Study has been conducted and analysed under two different Scenarios, (1) : without provision of reservoir in the Bearma basin (2) : with provision of reservoir in the Bearma basin

HYDRO 2014 International

In the first scenario, all demands of water users on 10 daily basis from july 1 to November 10 are fulfilled through river whereas in second scenario all demands of users on 10 daily basis are fulfilled by river as well as from the reservoir RS directly connected through user WU7. In the present study, the irrigation management in the command of Bearma basin has been carried out from reservoir releases. In this study “rule curve reservoir” method was used for addition of reservoir in Bearma basin. Irrigation demands for soybean crop during the monsoon period (June to October) on a 10- daily basis for all users namely WU1, WU2, WU3, WU4, WU5, WU6 and WU7 existing in sub-basins namely SW1, SW2, SW3, SW4, SW5, SW6 and SW7 have been computed by using CROPWAT. It can be seen that in scenario (1) there is no provision of reservoir in the basin, user WU7 used maximum water as 125.55 MCM and deficit is also maximum in this sub-basin with 88.48 MCM. In scenario (2) with provision of reservoir in basin, it can be seen that that reservoir RS has used maximum water of 218.05 MCM and deficit of 42.41 MCM also occurs. The performance is more noticeable that demand deficits have greatly reduced from 88.48 MCM to 42.41 MCM for WU7 by construction of reservoir. It can be appreciated that all the users that have not been connected to the reservoir are facing deficits of varying magnitudes under drought situation. Therefore, it will be prudent to explore additional sites for reservoirs on different locations so that the deficits can be minimised to the minimum extent possible. Key words : Bearma basin, MIKE BASIN, rule curve method. The Bundelkhand region was once known for its large natural resources, abundant water resources including perennial streams, large number of traditional tanks and rich forests. However, large scale exploitation of all these resources has made the area to be the poorest by which pressure on water resources in the Bearma basin is likely to increase dramatically in the near future as a result of high population growth. It is required to protect rivers from degradation caused by hydrological conditions (Cui et. al., 2010). However, the water demand is increasing whereas water resources are expected to decrease because of climate warming and the same or decreasing precipitation (Bates et. al., 2008).Climatic variability, changes and uneven distribution of resources create water shortages and interrupt the usual water linked activities posing serious threat to nature, quality of life and economy (Hisdal and Tallaksen, 2003). The recurrent droughts in the last decade had led to large scale migration from the Bundelkhand due to non-availability of water for domestic and agricultural activities. The low stream flows are indicative of rainfall situation (Galkate et. al., 2010). In fact, drought is estimated to be the most costly natural disaster in the world, wide range of detrimental effects associated with precipitation deficits include: decreased crop yields, increased wildfires, death of cattle and wildlife, water shortages, and rising food prices (Witt, 1997) and the most complex and least understood of all natural hazards, affecting more people than any other hazard (Wilhite, 2000). Drought impacts the poorer economies to a larger extent and may cause fatalities as compared to developed countries. A drought is an extended period when a region notes a deficiency in its water (Beran and

MANIT Bhopal

Page 92

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Rodier, 1985). The consequences of drought vary greatly depending on its location, timing, extent and the type of society or societal sector impacted by the drought (Gleick, 1993).The different types of droughts have each their own specific spatiotemporal characteristics (Peters et al., 2006; Tallaksen et al., 2009). Different types of drought are meteorological, hydrological, agricultural and socio-economic (Hisdal and Tallaksen, 2003). Meteorological drought simply refers to the atmospheric conditions that result in the absence or reduction of precipitation and since its definition only relies on rainfall. Due to its reliance on plant and soil conditions, agricultural drought usually has a lag time in response to precipitation changes (Park et al., 2005), and the impact depends greatly on the timing of the drought in relation to crop growth. In light of the grim scenario in the region, the Bearma basin is a major tributary of the Ken river system, the life line of Bundelkhand has been selected to study the drought scenario in the recent years and for planning to cope up with such situation in future through reservoir modeling. The monsoon rainfall is the only possible source of irrigation in Bundelkhand region of semi-arid Central India. A continuous spell of poor rainfall in combination with high temperature in successive years hinders water availability and imparts stress on ground water resources leading to severe drought in many parts during both, the monsoon and the non-monsoon seasons. Therefore, in present study irrigation model has been simulated especially for drought year 2002, under two different scenarios (without provision of dam in the main stream and with provision of dam in the stream) has been carried out to analyse irrigation deficit for soybean crops. The model was simulated from observed flow by preparing Bearma basin model in MIKE BASIN software. Methods Irrigation Management Planning for Bearma basin For determination of suitable sites for construction of reservoirs in study area, one location has been identified selected on the main river. To develop drought mitigation strategies through scientific planning of water resources and management, MIKE BASIN model has been developed. In the present study, the irrigation management in the command of Bearma basin has been carried out from reservoir releases; therefore, reservoir, irrigation nodes and transfer of water through channel have been specified.

Figure 1. Schematic representation of reservoirs and the various water users drawing water from reservoir as well as from the river (Scenario-2)

Description of MIKE BASIN Model Rivers and their main tributaries are represented by a network consisting of branches and nodes in the model. The model requires the entire catchment to be segmented into a series of sub catchments. The river system is represented in model by digitized river network which can be generated directly on the computer screen in Arc Map (DHI, 2003). A nodal representation of case study of Attanagalu Oya Basin, Sri Lanka was prepared using MIKE BASIN to estimate stream flow at each node, (K. R. J. Perera et al., 2010). Reservoir MIKE BASIN can accommodate multiple multi-purpose reservoir system and individual reservoir to simulate the performance of specified operating policies using associated operating rule curves. In present study rule curve reservoir method was used for addition of reservoir in Bearma basin. Rule curve reservoir regards a single physical storage and all users are drawing water from the same storage. Reservoir properties The reservoir characteristics, operating rules, upstream and downstream connections to users and control nodes are specified in the reservoir properties dialog. The level-area-volume table is used to compute reservoir volume at any level in reservoir. Reservoir operation properties The most common operating rule is the rule curve (standard reservoir method). Rule curves define the desired storage volumes, water levels, and releases at any time as a function of existing water level. Present study has been carried out using rule curve method. Channels The channels are the segments that connect water users, irrigation nodes and hydropower nodes to a river or a reservoir. In the present study the channel segment was used for connecting water users and reservoirs. Simulation MIKE BASIN Model has been simulated for drought year (2002). In first case, the model simulated after setting up all water users „without any reservoir‟ and in second case the model is simulated after setting up „reservoir‟ and water users. The output time series contain water used, demand deficit, stored volume in reservoir, water levels in reservoir, and channel flows at given time span assigned during simulation. The schematic representation of the reservoir and the various water users drawing water from reservoir as well as from the river (Scenario-2) is given in Results and Discussion

HYDRO 2014 International

MANIT Bhopal

Page 93

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 11

323

31.70

157.70

Irrigation Management Planning for Bearma basin The development of irrigation management plan depends on its scientific resource action plans and their proper implementation. Reservoir Characteristics For application of MIKE BASIN model, reservoir properties such as reduction level, high flood level, dead storage level, bed level and reduction factor are required which were determined using GIS and is given in Table 2. By using DEM of the study area and the drainage pattern of the catchment, one suitable site has been selected for construction of reservoir and incorporated in the analysis. Reduction level for reservoir has been fixed between high flood level and dead storage level, from where a specified reduction from demand user node was applied by the software, high flood level for a reservoir has been fixed and area below that has been extracted and histogram has been used to determine area elevation capacity Table. Dead Storage Level has also been fixed for the reservoir because below DSL water cannot use for irrigation. Table 2. Reservoir characteristics for reservoir RS S.No.

Reservoir Properties

1

Reduction Level

316m

2

Reduction Fraction

0.9

3

Dead Storage Level

314m

4

Bed Level

306 m

5

High Flood Level

323m

The cone formula (Murthy, 1968) has been used to compute the capacities between two successive levels, which in turn gave the cumulative capacities at different levels of reservoir. The area elevation capacity (AEC) table for reservoir RS has been given below in Table 3. The total command area of each user with respect to total command area in each sub-basin is given in table 4. The water demands for soybean crop for each of the water users on ten daily basis is given in Table 5. Table 3. Area Elevation Capacity Table for reservoir RS Sr. No.

Reduction Level (m)

1 2 3 4 5 6 7 8 9 10

306 309 312 315 317 318 320 321 322 323

HYDRO 2014 International

Cumulativ e Area (km2) 0.51 1.42 2.95 5.88 9.88 12.58 19.43 23.42 27.53 31.53

Cumulativ e Capacity (MCM) 1.25 3.96 10.34 22.92 38.37 49.57 81.29 102.68 128.13 157.64

Table 4. Total command area of each water user with respect to total command area in each sub-basin Sr. No.

Sub-basin SW1

Water Users WU1

Total Command Area (km2) 284.19

1 2 3 4 5 6 7

SW2 SW3 SW4 SW5 SW6 SW7

WU2 WU3 WU4 WU5 WU6 WU7

54.31 121.26 121.16 356.13 25.95 707.91

Table 5. Water demands of all users for soybean crop in Bearma basin in different ten daily period Date 01071995 10071995 20071995 31071995 10081995 20081995 31091995 10091995 20091995 31101995 10111995 20111995

WU1 WU2 WU3 WU4 WU5 WU6 WU7 (m3/sec) (m3/sec) (m3/sec) (m3/sec) (m3/sec) (m3/sec) (m3/sec) 7.57 1.45 0.00 0.00 0.00 0.73 19.91

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.35

1.03

0.31

8.36

4.77

0.91

0.00

2.15

6.31

0.00

0.00

7.63

1.46

7.33

0.00

0.00

1.05

28.59

10.26

1.96

5.09

2.93

8.61

1.38

37.53

13.91

2.66

5.94

0.00

0.00

0.61

16.71

8.52

1.63

5.04

0.77

2.27

0.90

24.42

12.63

2.41

0.94

6.30

18.51

1.35

36.79

3.45

0.66

0.79

0.00

0.00

1.20

32.61

1.32

0.25

0.72

3.94

11.58

0.84

22.94

0.00

0.00

0.00

2.20

6.47

0.41

11.31

Simulation of model under scenario-1(without reservoir) MIKE BASIN model is simulated for all sub-basins and water users without any reservoir during the period of July 1 to November 10 for the drought year, 2002. The analysis has been carried out to obtain the used water and deficit volume in different 10-days period at all seven sub-basins. The water used by different users and their deficit for different 10- days period have been presented in Table 6. From the analysis it can be seen

MANIT Bhopal

Page 94

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

that user WU7 used maximum water as 125.55 MCM and deficit is also maximum in this sub-basin with 88.48 MCM, and WU1 user faces maximum deficit of 62.07 MCM without any water being used. Amongst all the users, user WU3 faces a minimum deficit as 2.64 MCM. Other water users WU2, WU4 and WU6 also face low demand deficits of 11.86 MCM, 16.61 MCM and 7.86 MCM. Therefore it is imperative to harness the excess surface water by constructing reservoirs of larger capacity at suitable locations in the basin to meet the all demands of various water users, so as to provide buffer storage during drought situation.

with provision of reservoir. Table 6. Used water and Demand deficit water for all users without reservoir

Dat e

De ca de

WU1

WU2

WU3

WU4

WU5

W WU7 U 6 Used D Wate efi r cit (MC ( M) M C M )

Us ed W ate r (M C M)

De fici t (M C M)

Us ed W ate r (M C M)

De fici t (M C M)

Us ed W ate r (M C M)

De fici t (M C M)

Us ed W ate r (M C M)

De fici t (M C M)

Used Wate r (MC M)

D efi cit ( M C M )

Use d Wat er (MC M)

De fici t (M C M)

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.00

0. 00

0.00

0.0 0

0.00

0. 00

Jul y1D

0.0 0

6.5 4

0.0 0

1.2 5

0.0 0

0.0 0

0.0 0

0.0 0

0.00

0. 00

0.00

0.6 3

0.27

16 .9 4

Jul y2D

0.0 0

0.6 5

0.0 0

0.1 3

0.0 0

0.0 0

0.0 0

0.0 0

0.00

0. 00

0.00

0.0 6

0.09

1. 63

Jul y3D

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.3 3

0.98

0. 00

0.00

0.2 9

2.47

5. 47

Au g1D

0.0 0

4.1 2

0.0 0

0.7 9

0.0 0

0.0 0

0.0 0

1.8 9

3.46

2. 08

0.00

0.0 3

0.12

0. 60

Au g2D

0.0 0

6.5 9

0.0 0

1.2 6

3.6 9

2.6 4

0.0 0

0.0 0

0.00

0. 00

0.00

0.9 1

11.9 4

12 .7 7

Au g3D

0.0 0

9.7 5

0.0 0

1.8 6

4.8 4

0.0 0

0.0 0

2.7 8

8.18

0. 00

0.00

1.3 1

35.6 7

0. 00

Se p1D

0.0 0

12. 02

0.0 0

2.3 0

5.1 3

0.0 0

0.0 0

0.0 0

0.00

0. 00

0.00

0.5 3

14.4 4

0. 00

Se p2D

0.0 0

7.3 6

0.0 0

1.4 1

4.3 5

0.0 0

0.0 0

0.6 7

1.96

0. 00

0.00

0.7 8

21.1 0

0. 00

Se p3D

0.0 0

10. 91

0.0 0

2.0 8

0.8 1

0.0 0

0.0 0

5.4 4

15.4 7

0. 53

0.00

1.1 7

23.6 4

8. 15

Oc t1D

0.0 0

2.9 8

0.0 0

0.5 7

0.6 8

0.0 0

0.0 0

0.0 0

0.00

0. 00

0.00

1.0 4

5.60

22 .5 8

Oc t2D

0.0 0

1.1 4

0.0 0

0.2 2

0.6 2

0.0 0

0.0 0

3.4 0

4.95

5. 06

0.00

0.7 3

3.14

16 .6 8

Oc t3D

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

2.0 9

5.99

0. 16

0.00

0.3 9

7.09

3. 66

No v1D

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.0 0

0.00

0. 00

0.00

0.0 0

0.00

0. 00

To tal

0.0 0

62. 07

0.0 0

11. 86

20. 13

2.6 4

0.0 0

16. 61

40.9 9

7. 83

0.00

7.8 6

125. 55

88 .4 8

Simulation of model under scenario-2 (with reservoir) In this analysis, one reservoir has been suggested and simulation in MIKE basin has been conducted considering supply from this reservoir also. After simulation run, with reservoir during the period of July 1 to November 10 for all these seven users, the model provides used water, deficit water, reservoir volume, reservoir level. Analysis has been carried out to obtain the used water and deficit volume in different 10-days periods. In the analysis, the water user WU7 was directly drawing water for meeting their demands from reservoir RS whereas the remaining users WU1, WU2, WU3, WU4, WU5 and WU6 were simultaneously withdrawing water from the river directly to meet their demand requirements. The result of the water used and deficit for user SW7 has been presented in Table 7. From the analysis it can be seen that reservoir RS has used maximum water of 193.95 MCM and deficit of 66.51 MCM also occurs. It can be appreciated that all the users that have not been connected to the reservoir are facing deficits of varying magnitudes in drought years and therefore, it will be prudent to explore additional sites for reservoirs on different locations of the main Bearma river so that the deficits can be minimised to the minimum extent possible.

01 – 07200 2 1007200 2 2007200 2 3107200 2 1008200 2 2008200 2 3108200 2 1009200 2 2009200 2 3009200 2 1010200 2 2010200 2 3110200 2 1011200 2

Comparison of performance between scenario (1) and scenario (2) Initially, when the planning is carried out for a „no reservoir condition‟, it can be seen that all users face higher demand deficit in varying extents. The total demand deficit is 197.35 MCM in the drought year. The provision for constructing reservoirs helps to drastically reduce the demand deficit. The comparison of the demand deficit for water user WU7 drawing water through reservoir RS3 reveals that there is the demand deficit of 42.41 MCM after the provision of the reservoir RS3. However, when we compare the performance between Scenario1 (no reservoir) and Scenario-2 (with reservoir), it is seen that the maximum demand deficit of 88.48 MCM for WU7 with no reservoir scenario drastically gets reduced to 42.41 MCM with the provision of reservoir RS3. The comparison of demand deficit for both scenarios clearly demonstrates that the provision of reservoir RS3 with the basin has greatly helped to reduce the impact of drought as can be seen by the significant reduction in demand deficit with the reservoir supplies for water user. The performance is more noticeable because the demand deficits have greatly reduced from 88.48 MCM to 42.41 MCM for WU7

HYDRO 2014 International

ble 7. Water used-demand deficit for user WU7 directly connected through the reservoir RS for drought year Date

01-07-

MANIT Bhopal

Decade

Used Water (MCM) 0.00

Deficit (MCM) 0.00

Stored Volume (MCM) 1.25

Reservoir Level (m) 306.00

Page 95

International Journal of Engineering Research Issue Special3 2002 10-072002 20-072002 31-072002 10-082002 20-082002 31-082002 10-092002 20-092002 30-092002 10-102002 20-102002 31-102002 10-112002 Total

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

July-1D

0.00

17.20

1.35

306.11

July-2D

0.00

1.72

2.06

306.90

July-3D

0.00

7.95

4.27

309.15

Aug-1D

0.00

0.72

10.75

312.10

Aug-2D

9.88

14.82

157.64

323.00

Aug-3D

35.67

0.00

157.64

323.00

Sep-1D

14.44

0.00

157.64

323.00

Sep-2D

21.10

0.00

157.64

323.00

Sep-3D

31.79

0.00

156.72

322.97

Oct-1D

28.18

0.00

140.60

322.42

Oct-2D

28.18

0.00

122.46

321.78

Oct-3D

30.99

0.00

109.84

321.28

Nov-1D

17.84

0.00

96.82

320.73

218.05

42.41

From the results obtained, it is concluded that the demands are not fully satisfied and there were demand deficit under both scenarios for drought year. In second scenario one reservoir was planned and water was drawn from the reservoir as well as from the river and the analysis performed. The model was run to see the performance of the model and its ability to cope up during droughts. The model run in Scenario-1 shows that the demand deficits have increased significantly in all of the sub-basins as the supply in the river was very less. The maximum deficit was observed in sub-basin SW7.This indicates that the gravity of the situation magnifies as seen by the abrupt increase in the demand deficit in a drought year. Subsequently, the planning was carried out with the provision of one reservoir and model run in a drought years. Here it can be observed that the demand deficit has reduced considerably to 42.41 MCM which was aiming to achieve under such study. This study clearly demonstrates that planning for drought mitigation can be carried out by constructing small reservoir in the sub-basin to cater the increased demand during periods of intermittent dry spells during drought years. References i. Bates, B. C., Z. W. Kundzewics, S. Wu, J. P. Palutikof. 2008. Climate Change and Water Technical Paper of the Intergovernmental Panel on Climate Change. IPCC Secretariat, Geneva, 210. ii. Beran, M.A. and Rodier, J. A. 1985. Hydrological Aspects of Drought: a Contribution to the International Hydrological Programme, World Meteorological Organization, Studies and reports on hydrology 39, Paris. iii. Carter, D.B., Mather, J.R. 1966. Climatic classification for environmental biology. Publications in Climatology, Laboratory of Climatology 4 (19). iv. Cui, B. X. Li, K. Zhang. 2010. Classification of hydrological conditions to access water allocation schemes for Lake Baiyangdian in North China. Journal of Hydrology, 385:247-56,

HYDRO 2014 International

v. DHI. 2008. MIKE BASIN User Manual, Water and Environment, Inc. and Council of Governments. vi. Galkate, R.V., Thomas, T., Pandey, R.P., Singh, S. and Jaiswal, R.K. 2010. Drought Study in Chhindwara District of Madhya Pradesh, India. Third International Conference on Hydrology and Watershed Management (ICHWAM-2010), February 3-6, 2010, JNTU, Hyderabad, India. vii. Gleick, P.H. 1993. Water in Crisis: A Guide to the World‘s Fresh Water Resources. Oxford University Press, New York, NY. viii. Hisdal, H. and Tallaksen, L. M. 2003. Estimation of Regional Meteorological and Hydrological Drought Characteristics: A case study for Denmark, Journal of Hydrology, 281,230-247. ix. K. R. J. Perera,N.T.S. Wijesekera. 2012.Potential on the use of GIS Watershed Modelling for River Basin Planning – Case Study of Attanagalu Oya Basin, Sri Lanka, Vol. No.04, pp(13-22). x. Murtthy, B.N. 1968. ―Capacity survey of storage reservoirs‖, Central board of irrigation and power, publication no. 89. xi. Park, S., Feddema, J.J., Egbert, S.L. 2005. MODIS land surface temperature composite data and their relationships with climatic water budget factors in the central Great Plains. International Journal of Remote Sensing 26 (6), 1127–1144. xii. Perera, K. R. J, & Wijesekera, N.T.S. 2010.Identification of the Spatial Variability of Runoff Coefficients of Three Wet Zone Watersheds of Sri Lanka for Efficient River Basin Planning. ASCE: EWRI Conference on International Perspective on Current and Future State of Water Resources and the Environment. xiii. Wilhite, D.A. 2000. Drought: A Global Assessment (2 volumes, 51 chapters, 700 pages). Hazards and Disasters: A Series of Definitive Major Works (7 volume series), Routledge Publishers. xiv. Witt, J.L. 1997. National Mitigation Strategy: Partnerships for Building Safer Communities. Federal Emergency Management Agency, p. 2. xv. Yodre, R.E., Odhiambo, L.O., Wright, W.C. 2005. Evaluation of methods for estimating daily reference crop evapotranspiration at a site in the humid southeast United States.American Society of Agricultural Engineers ISSN 0883-8542,Vol.21(2).pp.197-202.

Replacement of Field Channels with Pressurized Irrigation Systems: in Ssp Command Area Mrs Sahita I. Waikhom1, Monali Patel2, Dr P.G Agnihotri3 Asst. Professor, CED, Dr. S. & S. S. G.G.E.C, Surat-395001, Gujarat, India 2 M.E Water Resources & Mgmt., Dr. S. & S. S. G.G.E.C, Surat395001, Gujarat, India 3 Asso. Professor, CED, S.V.N.I.T, Surat-395007, Gujarat, India 1 [email protected] 2 [email protected] 3 [email protected] 1

ABSTRACT: To irrigate the entire command area of SSP through conventional flow irrigation is no possible. There is Strong need for efficient and cost effective use of limited delta to cover the entire command area where optimization of water use is the prime consideration. It has been recognized that use of modern irrigation methods like drip and sprinkler irrigation is the only alternative using Pressurized Irrigation Network System (PINS). This is primarily, a pipe network carrying required discharge at adequate pressure, finally delivering it to the attached MIS network. Design of this network is suitably framed incorporating features of water distribution under the Canal Command Area (CCA). Pressurized Irrigation Network

MANIT Bhopal

Page 96

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

System (PINS) is used as a substitute for sub-minors and field channels in an open canal network.

2. NECESSITY OF PRESSURIZED NETWORK SYSTEM (PINS)

In present study, the design of PINS using Spreadsheet with multiple outlets at emitters with two alternatives i.e., 24 hrs power supply and 8 hrs power supply is carried out. The study area is selected in agro-climatic zone no.6 of SSP Command area which is situated at village Rampur. Rampur village is served by Dholka direct minor canal. Analysis is carried out for both the alternatives using Darcy-Weisbach formula and diameter of PINS pipe, connecting pipes, storage, pumping requirements and number of filters is computed using spreadsheet.



Adoption of PINS with MIS in the VSAs in the SSP area can assure water availability to each farmer and uneven distribution and tail end problems can be overcome. It is envisaged that where the Narmada water has reached but the sub-minors are yet to be constructed – is the most preferable situation where such pilot projects can be attempted. Keywords- PINS, Sub-minors, Command area, Sardar Sarovar Project (SSP), Conventional Irrigation.





  

To make the Micro-Irrigation System (MIS) adoption technically viable in the canal command areas, a pressurized water conduit system act as bridge by drawing water from the canal, storing in a place. To minimize the land acquisition problem. Not possible to irrigate the entire command area of SSP through conventional flow irrigation. Strong need for efficient and cost effective use of limited delta to cover the entire command area Limited availability of water - optimization of water use is the prime consideration Adverse soil characteristics in certain areas - low application of water is imperative Flood/ flow irrigation not desirable to problematic areas. To restrict unregulated water lifting from canals. Conjunctive management of pipe distribution with ground water. To improve overall farm efficiency.

2.1 Objective

1. INTRODUCTION Water is one of the most critical inputs for agriculture which consumes more than 80% of the water resources of the country (Sen, 2012). Agriculture is the largest user of water, which consumes more than 80% of the country‟s exploitable water resources. The overall development of the agriculture sector and the intended growth rate in GDP is largely dependent on the judicious use of the available water resources. While the irrigation projects (major and medium)have contributed to the development of water resources, the conventional methods of water conveyance and irrigation, being highly inefficient, has led not only to wastage of water but also to several ecological problems like water logging, salinization and soil degradation making productive agricultural lands unproductive (MoA, 2006). There is a strong need for efficient and cost effective use of limited delta to cover the entire command area of SSP. It has been recognized that use of modern irrigation methods like drip and sprinkler irrigation is the only alternative using Pressurized Irrigation Network System (PINS) (Carlos, 2009). Pressurized Irrigation Network System (PINS) is substitute arrangement for sub-minors and field channels in an open canal network (SSNNL, 2009). This is primarily, a pipe network carrying required discharge at adequate pressure, finally delivering it to the attached MIS network. Design of this network is suitably framed incorporating features of water distribution under the Canal Command Area (CCA). In present study, the design of PINS using Spreadsheet with multiple outlets at emitters with two alternatives i.e. 24 hours power supply and 8 hours power supply is carried out. Sardar Sarovar Project (SSP) is one of the major irrigation projects in Gujarat state of India. The main thrust of command development activities is on the empowerment of beneficiary farmers in sustainable water resource management (SSNNL, 2009).

HYDRO 2014 International

 

IRRIGATION

The objective of the study is to Design Pressurized Irrigation Network System (PINS) with 8 hours & 24 hours power supply to make the Micro-Irrigation System (MIS) adoption technically viable in the canal command area. 3. SSP COMMAND AREA Sardar Sarovar Project (SSP) is one of the major irrigation projects of Gujarat state of India. Sardar Sarovar (Narmada) Project Phase –IIA covers Culturable Command Area (C.C.A of 20, 42, 39 Ha) between Mahi and Surashtra Branch Canal off– taking from Narmada Main Canal. The study area is selected in agro-climatic zone no.6 of SSP Command area which is situated at village Rampur.

MANIT Bhopal

Page 97

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 1 Location of Study area in Gujarat

(iii) Design of PINS pipe

Rampur Village is served by Dholka Direct Minor through Dholka branch canal located at Dholka taluka, Ahmedabaddistrict. The Dholka Minor is off-taking @ Ch. 52140 m of Dholka Branch Canal having C.C.A. of 789.82 Ha. Out of which C.C.A. of Rampur is 124 ha. Thus C.C.A of Dholka minor is divided in 19 chaks. It is divided into two chak; chak 1 & chak 2 with areas as 27.74 & 32.09 C.C.A. respectively, each of which is further divided into 4 sub-chaks. Darcy-Weishbach formula is used to carry out analysis to decide diameter of PINS pipe, connecting pipes, storage, pumping requirements and number of filters required by using spreadsheet. Discharge has been computed using basic discharge co-efficient (BDC) taken as 0.65 for agro-climatic zone VI. Presently the farmers of the proposed project area have limited irrigation facilities. There is only one bore well in the proposed area of the study area. The power supply can be made available by Uttar Gujarat Vij Company Limited (UGVCL). 3.1 Data Requirement The data needed to carry out design are meteorological data, region map, index map, soil map & salient features of Dholka direct minor. 4. METHODOLOGY FOR DESIGN OF PINS For pressurized pipe network, three types of pipes like Polyvinyl Chloride (PVC), High Density Polyethylene (HDPE) and Fiber Reinforced Pipe (FRP) can be used.

They carry water at an adequate pressure, to deliver it to the attached MIS network. Here in study for pressurized flow HDPE pipe is preferred. Design discharge Q = (6 / n) x [BDC x CCA 1] /2 (8 hrs) Q = BDC x CCA1/ 2 (24hrs)

(iv) Pumping Efforts Pump HP = (Q x H) / (75 x) Where, Q = design discharge in lps H = Pressure head in m; n = no. of sub-chaks (v) Filters Capacity of media filter (m3/ hr) PINS pipe X 3.6 (8 hrs) Capacity of media filter (m3/ hr) PINS pipe X 3.6 (24 hrs)

= Design discharge of = Design discharge of

5. OUTCOME As per above steps design is carried out and result is obtained as shown below(Table-1 & 2) for both alternatives along with schematic (Fig. 2 & 3) For 8 hours Power Supply PINS Pipe designed: Table 1 Design of PINS Pipe (8hrs Power Supply)

For design of pressurized irrigation network system components like connecting pipes, storage facility, PINS pipe, pumps, filters, and intake well and pump house are required. Same design can also be prepared for the regions which face severe water scarcity and areas where natural water bodies exist can be identified and PINS can be adopted there. The design for PINS at Rampur village is carried out by following steps. In the distribution design of PINS, storage well is considered in the start of command area and pump house close to well. PINS design for all chak area is prepare using Spreadsheet for 8 hours & 24 hours power supply. (i) Connecting Pipes

Cha k No

1

2

CCA (ha)

27.7 4

32.9 6

Discharg e (lps)

Sub cha k No

Designe d

Available

Pipe OD (mm )

1

117.3

127.6

140

2

177.15

182.6

200

3

165.8

182.6

200

4

149.9

164.2

180

1

148.86

164.2

180

2

148.86

164.2

180

3

183.8

205.4

225

4

187.3

205.4

225

Pipe Inside Dia (mm)

13.52

15.93

Connecting pipe is an arrangement necessary to connect the source of water to the storage with the intake well i.e. Initial point of PINS. In our case for non-pressurized gravity flow we prefer PVC pipe. For this, generally low pressure gravity mains of PE 80 class of PN 2.5 (2.5 kg/cm2) would be sufficient. (ii) Storage Facility Facility is required for 8 hrs power supply. For practical purpose, 1 day storage facility is to be designed.

HYDRO 2014 International

MANIT Bhopal

Page 98

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 performance and improvement in efficiencies of the irrigation systems, it is necessary to adopt a self-sustainable system. A modernization of canal command area is, therefore, necessary through micro irrigation system. To bring more area under irrigation it has become extremely necessary to introduce new irrigation techniques like micro irrigation system for economizing the use of water and increase productivity per unit of water. Micro irrigation system need be promoted in a holistic manner involving appropriate methods like “PRESSURIZED IRRIGATION NETWORK SYSTEM” (PINS).

Figure 2 Schematic Diagram of PINS (8 hrs Power Supply) (Source: SSNNL, Gandhinagar) For 24 hours Power Supply PINS Pipe designed: Table 2 Design of PINS Pipe (24 hrs Power Supply)

Chak No

1

2

CCA (ha)

27.74

32.96

Discharge (lps)

18.03

The PINS along with MIS will result in many advantages like increase in crop productivity (20-30%), water saving (30-50%), fertilizer savings (approximately 40%) and bringing more area under irrigation with the same quantity of available water, equity in distribution of water, both spatially and temporary.

Designed

Available

1

117.46

127.6

140

Adoption of MIS with PINS in the VSAs in the SSP area can assure water availability to each farmer and uneven distribution and tail end problems can be suitably overcome. In addition to the above tangible financial benefits, the conversion of irrigation method from flooding to MIS and its integration with PINS in SSP will also have other important intangible benefits.

2

152.58

164.2

180

REFERENCES:

3

129.1

145.8

160

4

142.7

145.8

160

1

148.3

164.2

180

2

148.3

164.2

180

3

183.8

205.4

225

4

187.3

205.4

225

Sub chak No

Pipe Inside Dia (mm)

Pipe OD (mm)

21.42

Figure 3: Schematic Diagram of PINS (24 hrs Power Supply) (Source: SSNNL, Gandhinagar) 6. CONCLUSIONS The country is likely to be more water stressed in the coming years. Therefore technologies for water harvesting and storage and technologies for precision water application methods need to be adopted (Mehta, Sharma, Kathuria, 2012). For effective

HYDRO 2014 International

i. Carlos Estrada, César González,Ricardo Aliod, and JaraPano (2009), Improved Pressurized Pipe Network Hydraulic Solver for Applications in Irrigation Systems, American society of Civil Engineering. ii. LakhdarZella, Ahmed Kettab, Gerard Chasseriaux (2006), Design of a Micro-irrigation system based on the control volume method, Biotechnology, Agron. Soc. Environment volume10 iii. Literature from Sub-division Office (FO), SSNNL, Dholka iv. Mamta Mehra, Devesh Sharma, Prachi Kathuria (2012) Groundwater use dynamics: analysing performance of microirrigation system - a case study of Mewat District, Haryana, International Journal of Environmental Sciences Volume 3, no 1. v. Micro-irrigation (drip & Sprinkler irrigation) guidelines (January 2006) by Ministry of Agriculture, Department Of Agricultre (DoA) & Cooperation, Govt. of India. vi. Paper on Pressurized Irrigation System by Sardar Sarovar Narmada Nigam Limied (SSNNL 2009). vii. Sen, Somanth Project Report, (2012), Impact Assessment of Micro Irrigation scheme in Madhya Pradhesh.

Reservoir Operation Based on Real Time Flow Data for Flood Control and Incremental Power Generation Rameshwar Prasad Pathak B-474, Sarita Vihar,New Delhi, 110076,India [email protected] ABSTRACT: The floods are most frequented natural disasters in the world. The water management and flood control shall be on top priority in National Development plan. The monsoon

MANIT Bhopal

Page 99

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Rivers and snow melt rivers have its special features and characteristics. In all cases to meet the annual irrigation and power requirement and also for regulation needs, water storages of varying magnitude are created. The water storages/reservoirs are also instrumental in flood moderation. However providing additional reservoir capacity for flood moderation has advantages and disadvantages, but flood moderation by reservoir operation based on real time flow data is more effective method. The pragmatic approach also results in increased power generation in downstream power projects. The river basin planning, intercepted catchment flow, silt load, available command area are amongst various factors to be considered in deciding principle levels of reservoir which imposes limits on reservoir operation and thereby limiting the flood moderation. The paper also considers the elements of dam safety, which are prime factor in the studies. The submergence of land and property and draw down cultivation are another issues relevant to the subject, while framing reservoir operation rules. During monsoon, the inflows in to and outflows from reservoir are not only unpredictable but are subjected to great variations, creating flood like situations. The real time data collection gives new dimension to solution to flood moderation. The paper founded on case study and literature available on the subject, deals with this solution which will help to control flood and add to power generation by utilizing reservoir capacity optimally. Keywords: Reservoir operation, pragmatic approach, Real time Flow data, Forecast, Predepletion, Flood moderation, incremental generation, submergence 1.0 INTRODUCTION: The rainfall and snowfall are two important elements for growth of life on this planet. But there is tremendous variation in space, time and quantum of rainfall. The less rainfall in an area creates draught condition not only affecting human being but all living creatures, flora and fauna in that area. The rivers drain surface water and even underground stream water to sea. The wishful rainfall in terms of space, time and quantum is still in dreams of scientists except for some very expensive methods in very limited way and very limited weather conditions. Consequently we have to live with it and device methods and means to face such adverse situations, keeping in view human comfort. The excessive rainfall creates flood like situation, inundating the area sometime inhabited or agricultural fields or both or other important establishments. Similarly excessive snowfalls may paralyze life and block roads, streets and cover the affected area. As the snowfall has variation so also snow melt has variation in summer and winter as per intensity of the seasons. Therefore planning has to be done for these different categories of the rivers. In China in one of the instant the extreme summer, causing excessive snow melt flow in the river was followed by consecutive rain storms, which resulted in unprecedented floods in the area. Geographically in India following three categories exist: 1. 2.

Rivers in hilly region of Himalayas/snow melt Rivers in hilly region excluding in category no. 1

HYDRO 2014 International

3.

Rivers in plain region

1.1 Occurrence of Flood In brief the floods and draught are resultant of variation of rainfall/snowfall in space, time and quantum and also variation of intensity of summer and winter seasons. The rain precipitate in the catchment area and through Nullahs and tributaries rain water flows to the main river channel, all along its length. The river bed is lowest contour of the area and mostly in fault zone. Over the years, erosion and cutting by river water has shaped its banks. Within these banks the river channel is located and most of the time river is confined within these banks. The flood like situation is encountered, when river overflows its banks. This so happens that the inflow to rivers exceeds the channel capacity and afflux is above the height of the banks. In this scenario the high afflux level obstruct the flow from Nullahs and tributaries and the levels in Nullahs and tributaries also rises above normal due to back water levels causing flood like situation in those areas also. All this causes submergence of neighboring lands of the rivers, Nullahs and tributaries. The snow melt river will have some additional issues to be considered. The storage reservoir created intercepting this flow can accommodate this additional quantum of water and release it in regulated manner to minimize submergence in the downstream areas. In the event reservoir gets filled up to the level as specified by reservoir rule curve, the gates are opened to release extra inflows and the maximum outflow that would be possible would correspond to the level available above the crest level. If the inflows to the reservoir increase, the reservoir level will also build up to the required afflux above the crest to matching outflow is developed and outflow shall balance the inflow. Incase inflow approaches highest flood the reservoir will touch the maximum water level, by this increased capacity the quantum of downstream flood will be reduced. The time required for opening the gates and also uncertainty of estimation of inflows result in to excessive releases from reservoir causing flood like situation in downstream areas, may it be devoid of precipitation of that magnitude. On these events critics raises eye brows against construction of dams. Whereas in extreme conditions, the floods are inevitable, dams or no dams. Only in most exceptional case, dam break can cause additional flood fury, which is disaster beyond control of technology adopted or operator deployed. As such limits of dam safety must be clearly defined. 2.0 AMBIT OF CONSIDERATION With the foregoing discussions it evolves that various elements are responsible for flood downstream of storage/reservoir. This is complex phenomenon, which needs in depth study by expert of the field. Broadly the inflow depends upon the characteristics of upstream portion of the basin, including direct draining areas. Similarly the downstream portion gets affected in accordance with its characteristics. Therefore the holistic approach shall be adopted, considering complete basin, and all other elements for study of the subject. The reservoirs are control node and its parameters are first elements of control, and decide flexibility in the system. There are other factors which add to accuracy of

MANIT Bhopal

Page 100

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

control, like information about siltation of reservoir and consequent reduction in capacity of reservoir. The information about inflows to the reservoir both in quantum and time will multiply the flexibility of control available at reservoir. Even the information of downstream structure/ channel will enable the control of upstream reservoir to be effective in mitigating adverse situation downstream. This is made possible by real time flow data system. This pragmatic approach will not only reduce the fury of flood but will enable the River Bed Power plant to generate additional power, utilizing flood water. The paper deals with the major factors for monsoon rain fed rivers. For snow melt river study is similar except some additional factors need to be considered. 3.0 RIVER BASIN The river basin is the expanse of land from which all surface water from rain or snow melt drains through a sequence of streams, rivers and lakes, into the sea at the stated single river mouth, estuary or delta. Apart from physiograpy and Geological and geographical features, the land use of the basin land is very important. Govt. have also constituted various agencies responsible for development of Basin. The developmental activities would include water management, planning for reservoirs, canals, power plants. For this developmental work the land use may change calling for rehabilitation and resettlement or environment mitigating measures. For the present study the major River Basin Characteristics can be summarized as follows: 



Description of River basin: The Geographical and Physical Features and Natural Elements of the basin. Physiography, Geology, Geochemistry, Soils, vegetative cover, water resources, etc. Land use: Forestry, Agriculture, Biodiversity, wild life, Aquatic life, Population, Roads, Cities, Existing Structures,





ydro Meteorological features: Climate, temperatures, humidity, rainfall, snowfall, surface and under ground water. Water Management: Hydrological details, river network, irrigation, power, flood control

3.1 Description of River Basin The expanse of river basin from origin of river including its tributaries to confluence to sea is defined by longitude and latitude. The large river basin may be divided into sub basin also. The physical characteristics of the River basin include its location, physiography, soils, climate, surface water and ground water resources, and natural water quality. The geochemistry of the River Basin is based primarily on stream sediment and stream geochemical data. The regional geologic grouping of rocks of similar compositions, porosity, permeability, are of greater importance in stream hydrogeochemistry. The parent materials in which the soils formed, the subsoil in various depth, the major groups of soils in the area, and other details need to be studied. The River basin is a dynamic hydrological system

HYDRO 2014 International

containing interactions between aquifers, streams, reservoirs, floodplains, and estuaries. Because water is transmitted through faults and fractures, each surface water drainage basin or watershed is also a ground water drainage basin or watershed. Surface and ground water are in such close hydraulic interconnection that they can be considered as a single and inseparable system. All these elements are responsible to establish relationship between precipitation and runoff. 3.2 Land use The Description of river basin gives clear understanding of what we are dealing with in terms of physiographic details, which is resultant of nature‟s action, and in case of study of land use we have to deal with what environment and life and nature we have to protect. The urban development and rural areas have encroached in the flood zone. The approach road, bridges, culverts, important structures, etc, are also very sensitive items. These shall be indicated on contour maps clearly indicating populated area, agricultural fields and all other details. The forest area, reserved forest, pilgrimage activities, sanctuary, area important for biodiversity, aquatic life, etc needs to be also indicated in the contour maps for complete basin. The water resource studies must include it while planning a project or operating it. The runoff characteristics changes with such developmental activities. 3.3 Hydro Meteorological features Three main elements of the climate that significantly affect the water availability and present grounds for development, use and conservation of this resource are air temperature, precipitation and evapotranspiration. The orographic features reflect upon these most important climatic events. Depending on variation in climate, the large basin can be divided in different zone for convenience of studies. Its variations are the result of land and sea distribution and closeness, as well as of various orographic features. Considerably more precipitation H occurs in mountainous parts of the basin than in the plains winter temperatures (December to February) are low, while high temperatures occur during the summer season (June – September). Average annual temperatures in the region vary in a wide boundaries depending, in the first place, on elevation. The lowest long-term annual average temperatures at measured points take place on the mountain ridges With regard to air temperatures, it can be roughly assessed that within-the-year variations exhibit a common pattern for majority of the catchments in plains. Dividing lines between these different zones are not sharp, due to different degree of influence of various factors that determine the climate. At high altitude the precipitation falls in form of snow so that relatively long periods with snow cover are common characteristic of the region. Generally there are too few reliable data available about impact on climate changes on flows, large pressure to land use change, lack of non-structural measures. The study should comprise collection and analyses of data at meteorological and hydrological gauging stations at the basin-wide level, evaluate flood characteristics and drought properties in meteorological and hydrological aspects, flow forecasting and climate change. Precipitation amount and its

MANIT Bhopal

Page 101

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

annual distribution varies widely within the basin. It, however, can roughly be asserted that the form of precipitation has a common feature: rainfall of different duration is likely to occur all over the whole catchment where low mountains, hilly terrain and plains dominate. Most precipitation occurs in summer monsoon season and part during autumn monsoon. 3.4 Water Management: Water is renewable source. Hydrologic cycle involves the continuous circulation of water in the Earth-atmosphere system. Of the many processes involved in the water cycle, the most important are evaporation, transpiration, condensation, precipitation, and runoff. Although the total amount of water within the cycle remains essentially constant earth as unit, its distribution among the various processes is continually changing. The various steps involved for hydrologic evaluation of details are as under:  Extension of records  Transferring of records  Statistical analysis of historic records  Hydrological Modeling Hydrology is not an exact science. The meteorological data combined with characteristics of river basin including land use, are fundamentals for analysis of hydrological details and working out equation for rainfall and runoff relationship. A typical water balance analysis will compare meteorological input data to a measured (or transferred) set of flow data within the receiving stream. The precipitation-runoff process is complex as it involves numerous flow routing interactions in the watershed. Additionally, the spatial and temporal characteristics of precipitation also make the prediction of runoff a challenge. Additionally, the spatial and temporal characteristics of precipitation also make the prediction of runoff a challenge to engineers. For sustainable development of the earth water management is challenge to the planners. Its scarcity or abundance both creates havoc in the system. Its spatial and temporal variation leads to storage of water to meet the various requirements of society of irrigation, drinking, industry, power, waterways, sanitation, and many such other requirements. This paper deals with the event when there is heavy precipitation and runoff has flooded the channel. 4.0 STORAGE PLANNING There are very few reservoirs planned for flood control. Even no additional capacity is provided in any of the reservoir for flood storage is provided (exceptions are there). Only temporary flood storage exists between the maximum Water Level and Full Reservoir Level. The spillway capacity is designed to pass the Highest Flood and the corresponding highest water level at crest shall not exceed Maximum Water Level. 5.0 RESERVOIR OPERATION The reservoir capacity is designed to meet various requirements of drinking, industrial use, irrigation, power, downstream releases, etc. and to meet losses by way of evaporation losses,

HYDRO 2014 International

seepage losses, silt load, etc. Reservoir Rule Curve envelopes, the Lower Rule Curve, and Upper Rule Curve, which defines the range of Operating Regimes. Lower Rule Curve defines the operation to match the flow that can be maintained through out the dry season under the lowest hydrologic condition so that the reservoir reaches its minimum operating level. These are the minimum elevations that the reservoir should maintain in order to guarantee to meet the required output. Upper Rule Curve defines the limit of operation with the minimum spills, which exceed the regulating capacity of the reservoir combined with the discharge capacity of the power plant These operations are at the maximum elevations that the reservoir should maintain in order to guarantee to meet the required output and safety of dam. In addition to this the priority is set in which order the various requirements are met. The downstream releases are usually on instructions of tribunal or the court and gets first priority. However drinking is basic need of Human and this gets top priority, and almost at par is the downstream release, Industry use is second priority followed by irrigation. Power release trails behind. At the same time downstream releases are through the river bed power releases. During monsoon period, the unpredicted quantum of rain fall adds to inflows, which add to uncertainty to the operations. Excess inflows give opportunity to increase generation. However power releases are limited to machine discharge capacities. A pragmatic approach shall be adopted using the befitting software. 6.0 REAL TIME FLOW DATA The availablity of Real Time Flow Data, supported by extensive hydro meteorological network add new dimension to the solution to the problem. Imposition of competitive water charges, restriction on water releases to control fluctuation of water levels in downstream, environmental aspect, safety of fast growing urbanization, safety of rail- road transport network, Safety of hydraulic Structure are many such factors which warrants for reservoir operation in close margins and accounting and monitoring of releases from reservoirs. This requires to make correct assessment of inflows and out flows from the reservoirs. For this stream-flow data, real time information on impoundment or variation in impoundment at the reservoir projects, estimation of evaporation losses and monitoring of withdrawal from reservoir are required. This requires a strong Hydro meteorological net work, with proper communication preferably satellite communication system which will remain operative in remote areas and in most adverse condition 6.1 Hydro meteorological net work In this paper it is stressed that the complete river basin planning shall be done and not for a reservoir in isolation or State wise. Hydro meteorological network shall cover the complete river basin from origin to confluence covering tributaries and other major drainage system which has come up with growing urbanization. The cover area of rain gauge station normally depends upon the topographic characteristics of different part of basin, intensity, distribution and rainfall, storm areas, the number of streams draining the catchment area, etc. The river basin characteristics need to be considered, some of which have

MANIT Bhopal

Page 102

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

been discussed in foregoing paragraphs. The spatial distribution of network would be influenced by the setting up of developmental scenario comprising of a number of artificial interceptions of flows by way of hydraulic structures or storage and diversion. Meteorological network shall be equipped with all modern equipments and facilities specially well distributed rain gauges capable of collecting and reporting precipitation in terms of time and quantity of occurrence, with adequate number of observatories to monitor the dew point, wind velocity, temperature, relative humidity, radiation/sun shine hour etc. In order to assess evaporation losses pan evaporation data shall be collected. As wind velocity, temperature, relative humidity, radiation have parametric effects on evaporation, these data shall be collected at each storage sites as well. The silt load characteristics of stream flow are required during operation stage to monitor the effect of storage interception, for assessment of realistic quantity of water stored. The hydrological observation network be also equipped with all modern equipments and facilities for measuring runoff, stream velocity water levels, specifically during flood. The rainfall and runoff equation be evolved considering the watershed characteristics, and be revised on regular basis on developmental activities changing landscape, and comparing the runoff so calculated with measurement at various hydrological observatories. 7.0 FLOOD ABSORPTION BY PREDEPLETION AND INCREMENTAL POWER GENERATION Pre depletion of reservoir is not an element of consideration at design and planning stage. But supported by strong Hydro meteorological net work and communication system, and based on computerized realistic assessment the pre depletion of reservoir can be safely implemented with negligible risk of loss of precious water and also resultant reduction of fury of flood otherwise endangering neighboring and downstream areas of submergence, and also for safety of hydraulic structure in case meteorological conditions further worsen. The regulated release of water on account of predepletion can further be planned through power house for incremental generation. Extended time available for depletion would not only result in reduction of intensity in downstream release but will allow more water to be released through power house resulting in incremental generation. The depletion of reservoir would depend upon the degree of accuracy of assessment and time lag assessed. Both these factors would be governed by the detail study conducted on characteristics of river basin, including river channel, and how authenticated rainfall and runoff equation is formulated. This exercise is relevant not only for reservoir but also for natural lakes as well. In recent flooding of J&K such an approach would have reduced the adversity to some extent 8.0 CONCLUSION: Water is much needed commodity and it shall be conserved. But surplus of water in terms of floods can disrupt the life killing persons and damaging the properties, submerging the area, causing deceases due to stagnant water. Draught and flood both are curse and reservoirs are answer to both these problems. Storage of water with hedging will help in fighting draught and

HYDRO 2014 International

flood control possible to an extent by pre depleting reservoir based on Real time Flow data, supported by strong Hydro meteorological network, uninterrupted communication and befitting software, and river basin characteristics are updated along with reservoir parameters, to get realistic assessment for incremental power generation and flood control. Such studies and its implementation had made the reservoir projects boom to the society

Effect of Conservation Works on Soil Erosion-A Case Study of Punegaon Reservoir Catchment Area M.B. Nakil1 M.V. Khire2 PhD student, CSRE, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India 2 Associate Professor, CSRE, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India Email: [email protected]

1

ABSTRACT: Better analysis of the erosion causing factors, knowledge of terrain uses are necessary for implementing soil and water conservation practices. The researchers have carried out experiments to quantify the effect of conservation practices in terms of value P used as a parameter in Revised Universal Soil Loss Equation (RUSLE). Conventional farm management practices implemented as per land-uses and the terrain slopes. To these land uses appropriate values of conservation practice factor P ranging between one and zero are assigned. Knowing the area occupied by various land use classes weighted mean (WM) value of P is calculated for micro watershed. The Government departments do implement erosion control works on Government land. These works make add on effect and reduce conventional values of P. The present paper deals with quantification of add-on effect of major soil and water conservation works carried on Government land. The effectiveness of works executed can be represented in terms of ratio of actual cost incurred to the estimated cost of conservation works. This ratio ranges 0 to 1 as per physical progress of works. Thus it has values 0 for not doing any work and 1 for all works completed. The modification factor defined as (1-ratio) is applied to weighted mean (WM) value of P to get modified value Pm. This value is used in RUSLE (A=RKLSCP) model, for predicting soil loss. The use of this methodology in soil loss prediction of Punegaon Reservoir catchment area shows good result. Keywords: Soil erosion, catchment area, RUSLE, Management practice factor, Soil Conservation 1. INTRODUCTION: The water induced soil erosion involves detachment, transportation and deposition of soil particles. The overall erosion process depends on six basic parameters viz. rainfall energy, properties of soil, land topography (slope steepness and slope length), land-use and cover, and support practices. These parameters decide quantity and extent of soil erosion. The support practices reduce the process of detachment of soil particles. The effectiveness of these practices is represented by

MANIT Bhopal

Page 103

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

the term P in the soil erosion prediction model. The farm practices are implemented by individual stakeholders. At the same time Government agencies do implement soil and water conservation works on government as well as on private lands. These works do reduce the value of term P and reduce overall soil erosion. This indirect effect has not been considered by any of the earlier model. The completed works in the microwatershed are considered for evaluation. This compound effect is evaluated in the present case study, in the form of modified value of support practices P. This modified value is used in soil loss prediction model for estimating the erosion quantity. 2. DETAILS OF STUDY AREA The Punegaon is medium project dam is situated in Western Ghat near Dindori town of Nashik district of Maharashtra State, on river Unanada, a tributary of Godavari river. It was built in year 1995 and functioning for 19 years. The dam is having catchment area of 63.84 km2. The area is having both hilly and gentle slope terrain. The elevation ranges from 700m to maximum 1088m. The slope varies from 0% to 80%. Most of the rock is Deccan trap basalt. The climate of the area is tropically humid with three seasons of four months duration namely rainy, winter, summer. The annual rainfall variation ranges from 450 mm to 2500 mm. The soil depth of the area varies from few centimeters to over 50 cm. The soils mainly are clay, clay-loam, sandy clay loam, gravelly loam, sandy loam. The Land use / Land cover classes prevailed in this area are namely forest plantations, deciduous forest, waste land, scrub land, built up land, paddy fields, fallow lands and wet lands and water. Agriculture is prominent. 3. SOIL LOSS ESTIMATION MODEL

The support practices are implemented by the farmers to reduce the soil erosion. There are various types of practices which are based on soil type, terrain slope and sustainability. Mechanical types like contouring, strip cropping and terracing are predominant in the study area. The reduction in soil loss from unity to fraction because of farm management practices is represented by the term ―P‖. The term is quantitative indicator of effect of management practices. The experiments on the effects of types of management practices on soil erosion have been carried out and the values of ―P‖ are estimated through the models. The values of “P” range from unity (no conservation works) to zero (no erosion). In practice it is not possible to get no erosion condition due to sustainability of the conservation practices. It is thus not possible to get 100% reduction in soil loss. 4. MATERIALS The Punegaon reservoir catchment is marked topographic sheets from Survey of India. The Indian Satellite IRS LISS III image of May 2011 is used for LU/LC supervised classification. The image analysis is carried out using ERDAS software. The ground survey data has been used as training sets. The classes identified are deciduous forest, forest plantations, waste land, scrub land, built up land, paddy fields, fallow fields, wet lands and water bodies. The support practices are crop and terrain specific. Government agencies have carried out the conservation measures on Public and private lands. The data of such conservation works are collected from Government agencies. This data is in terms of amount and is with regard to total estimate of conservation works and works actually implemented. The data of conservation scheme is village wise so the computation of factor ―P‖.

The researchers through simulation have carried out the experiments to estimate soil loss on small plots. Thus various estimation models got evolved. The model RUSLE (Renard et al., 1997) is revised Universal Soil Loss Equation model which was initially established by Wischmeier and Smith (1978) through USLE. The hybrid of USLE and RUSLE model is used in present case for estimating annual soil loss from study area. The revision of the model is with regard to revised methods of evaluation of factors. The RUSLE (Renard et al., 1997) is expressed same as USLE as shown in equation (1),

5. METHODS

A  ( R  K  L  S  C  P)

(2)

(1)

where Pt= average annual rainfall, There are five rain-gauge stations near to catchment of reservoir; however the Thiessen polygon shows only one station influences entire area. The equation is applied to average annual rainfall of 35 years. The average R value is considered for analysis. The vector layer of catchment in GIS environment is rasterised for the average R value.

where “A” is estimated annual soil loss (t/ha/yr). The factors which affect the erosion process are considered in this equation. These factors are namely ―R‖ a rainfall erosivity factor, ―K‖ the soil erodibility factor, ―L‖ the slope length factor, ―S‖ the slope steepness factor, ―C‖ the land cover management factor and ―P‖ the support practice factor. Like USLE and RUSLE many models are in practice. These models have modified approaches in evaluating the affecting factors.

HYDRO 2014 International

The erosion causing parameters namely R, K, L, S, C, P are evaluated. The methods are illustrated in following paragraphs. R symbol is used for rainfall erosivity parameter. It is having unit as (MJ mm ha-1 h-1). In the present case the R value is derived as per equation (2) developed by Nakil (2014).

R  (906.77  exp 0.0009  Pt 

The soil erodibility parameter ―K‖ is expressed in units as (t ha-1 MJ-1 mm-1 ha h). The K value is evaluated by following equation (3) as given by Wischmeier & Smith (1978).

MANIT Bhopal

Page 104

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

K  1.313 * [((2.1  10 4  M 1.14  (12  a))  ((3.25  (b  2))  ((2.5  (c  3))]  1000

LU/LC. These values and prevailing farm support practices are considered while assigning the “P” values to the LU/LC fields in the present case. The weighted mean value of parameter “P” is calculated for each micro watershed. It is denoted as “Pw”

(3) where M=[(% silt + very fine sand) x (100-% clay)], a=% organic matter, b= soil structure code number and c= permeability class number. The relevant properties of soil are used in the equation (3) to get value of “K” for each class. The soil class‟s areas are vectorized and then rasterised around “K” values. The slope length parameter “L” is evaluated by equation (4) and the slope steepness parameter “S” is evaluated by equation (5). These equations are given by Wischmeier & Smith (1978).

L  (  22.13) m

6.2 Deriving modification factor The Government funding is cost-estimated to execute all soil and water conservation works in a micro-watershed. The details of such overall cost estimate and the actual amount spent on works are made available. The % effectiveness of the works executed is calculated in terms of the ratio of expenditure done to the estimated cost of overall works. This ratio of works “Rw‖ and the modification factor “Mf” for these works are calculated for each micro-water-shed using equations (6) & (7) respectively.

Rw  (Cw  Ew)

(4)

S  (0.0065 s  0.045  s  0.065)

(6)

(5)

Where Rw=Ratio of works, Cw= Cost of executed works, Ew= Estimated cost of works

2

where , m=exponent, s=% slope, λ = slope length= 23.5m adopted pixel size in GIS The value of ―m‖ adopted are 0.5 for slope >= 5%, 0.4 for slope= 5% to 3.5%, 0.3 for slope= 3.5% to 1%, 0.2 for slope= < 1%. These values have been suggested by Wischmeier & Smith (1978). The values of factor “L” thus are derived using equation (4) out and are assigned in GIS environment to respective slope classes. The contours are digitized using topographic sheets and are converted to percentage raster map as percentage value of “s”. This percentage slope raster for “s” is used in equation (5) to get rater map for “S” parameter. The researchers have assigned the values of cover management parameter “C” as per land use and land cover (LU/LC), after due experimentation. The values range in between 0 to 1. These values are assigned to respective matching LU/LC units. 6. ANALYSIS OF CONSERVATION PARAMETER Large numbers of micro conservation works are carried out using Government funding in a catchment area. Each funded work is executed fully in a season, once it is commenced. Thus if funding is utilized say 30% then it means 30% number of works are fully completed. It does not mean that all works are at 30% progress level. These conservation works make add-on effect and reduce soil loss from catchment. As such the add-on effects result in modification of Conservation Practice parameter “Pm”. This modification of the parameter “Pm” is made as per following procedure as given by Nakil (2014).

Mf  (1  Rw) (7) Where Mf= Modification factor, Rw =Ratio of works Ideally when all proposed works in a catchment are executed (that is when expenditure on works is equal to estimated cost) it can be assumed that soil conservation is achieved fully for that catchment (the ratio of works “Rw” is one and modification factor “Mf” is zero). However such situation occurs occasionally. When no works are carried out, the ratio of works is zero and modification factor becomes (1-0=1). Thus the value of modification factor ranges between one to zero. 6.3 Deriving modified parameter The weighted mean value of conservation practice parameter “Pw” derived in 6.1, is multiplied by the modification factor ―Mf‖, as per equation (8), to account for compounded effect of Government conservation works.

Pm  ( Pw  Mf ) (8) where Pm= modified value of P, Pw = weighted mean value of P, Mf= Modification factor The modified values of conservation practice parameter are derived using equation (8) for each micro-water-shed. The derived values for each watershed as shown in Table 1 are vectorized and rasterised in GIS environment.

6.1 Deriving weighted value “Pw”

6. RESULTS AND DISCUSSION

In a micro water-shed the farmers adopt different practices as per land form and as per LU/LC. The researchers after due experimentation have assigned the “P” values according to

The Hybrid model of RUSLE and USLE are used in the present case to estimate the soil erosion of a reservoir catchment. The parameters namely R, K, L, S, C and P derived are integrated in the model. The model is run in GIS for conservation practice

HYDRO 2014 International

MANIT Bhopal

Page 105

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

parameter ―P‖ and for modified conservation practice parameter “Pm”. The map of erosion rates “A” derived for study area is shown in Figure1 while the raster maps for the parameters R, K, L, S, C, P are shown in Figure 2. The integration of parameters in both the cases resulted in following erosion rates.

Nakil M.B. (2014), Analysis of parameters causing water induced soil erosion, annual progress seminar Indian Institute of Technology Bombay,

Sedimentation rate with conventional conservation practice factor “P‖=0.072 Million tons per year Sedimentation rate with modified conservation practice factor ―Pm‖ =0.065 Million tons per year The observed average rate of sedimentation for this catchment =0.048 Million tons per year. The results reveal that the sedimentation rate estimated using conventional ―P‖ is 50% higher than the observed value. While the estimated sedimentation rate using modified ―Pm‖ is 35% higher than the observed value. The evolved method has helped to quantify the use of the soil conservation works executed through Government funding. It is seen here that the prediction of soil loss associated with modified conservation parameter “Pm” is reduced by 15%. The revised value of predicted soil loss is nearer to the observed value. While the soil loss estimation by using unrevised value of “P” results in over-prediction. The higher prediction even after using modified value “Pm” attracts refinement in other parameters of model.

Figure 1. Erosion rates of Punegaon Catchment

7. CONCLUSIONS The soil & water conservation works carried on Government / common lands in the watershed modify the conservation practice parameter used in soil loss equation model. These public works reduce the conservation parameter “P”. The methodology is evolved here to derive the modified parameter “Pm”. Normally it is examined that the soil loss estimated for catchment area using soil loss equation model is on higher side. The use of modified conservation practice parameter “Pm”, in-place of conventional value “P” helps to overcome this over-estimation. This approach can make the soil loss equation more accurate and thus acceptable particularly for large catchment area. Here the refinement and accuracy in quantification of soil loss estimation in view of public conservation works helps to correctly assess the reservoir sedimentation. The methodology can be used for any soil loss prediction model. REFERENCES: i. Feb. 2014: 42 ii. Renard, K.G., Foster G.R., Weesies G.A., Mc Cool D.K., and Yoder D.C. (1997) Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agriculture Handbook No. 703, U.S. Department of Agriculture, Agriculture Research Service, Washington, District of Columbia, USA. iii. Wischmeier, W.H. and Smith, D.D., (1978), Predicting Rainfall erosion losses- A guide to conservation planning, Agricultural Handbook number 537, USDA, Science and Education Administration, washington, District Columbia, USA

HYDRO 2014 International

MANIT Bhopal

Figure 2. Erosion causing parameters R, K, L, S, C, P of Punegaon reservoir catchment

Table 1. calculation of values of „Pm” on the basis of LU/LC

Page 106

International Journal of Engineering Research Issue Special3

S r. N o .

1

1

N a m e of V ill a g e

2 K ar a nj k h e d

Sl op e

LU /LC

%

3

up to 25

4

Ve get atio n/cr op fall ow Wa ste/ scr ub for est

2

3

4

Pi n g al w a di

B il w a di C h a us al e

up to 80

> 5

U pto 5

Ve get atio n/cr op fall ow Wa ste/ scr ub for est

for est Ve get atio n/cr op Wa ste/ scr ub for est fall ow buil t up

Appr ox. Area %

5

P

6

are a x P

7

15

0. 6 0. 9

15 13. 5

45

0. 7 5

33. 75

15

0. 7

10. 5

25

25 15

25 35

100

37

30 15 15 3

0. 6 0. 9

15 13. 5

0. 7 5 0. 7

18. 75 24. 5

0. 7

0. 6 0. 7 5 0. 7 0. 9 0. 5

Su m of (ar eas P)

W ei gh te d P = (8 /la nd ar ea )

T o t a l C o s t o f w o r k s

9

1 0

72. 75

0. 73

71. 75

8

C os t of w or ks co m pl et ed

R ati o of w or ks ca rri ed ou t (1 1/ 10 )

22. 5 10. 5 13. 5 1.5

HYDRO 2014 International

E f f e c t i v e n e s s I n d e x ( 1 r a t i o ) 1 3

Sediment Management in Reservoir of Hydroelectric Power Projects - Numerical Simulation Studies for Punatsangchhu – I, Bhutan Neena Isaac1 T.I. Eldho2 S.B. Tayade3 Chief Research Officer, Central Water and Power Research Station, Khadakwasla, Pune-411024, India Research Scholar, Department of Civil Engineering, IIT Bombay,Mumba-400076, India Email: [email protected], [email protected] 2 Professor, Department of Civil Engineering, IIT Bombay, India Email: [email protected] 3 Assistant Research Officer, Central Water and Power Research Station, Khadakwasla, Pune-411024, India Email: [email protected] 1

Fi na l P m

11

12

1 6 1 . 7 4

64 .8 5

0. 40

0 . 6 0

0. 44

0. 72

2 7 8 . 8 6

30 .7 6

0. 11

0 . 8 9

0. 64

70

0. 70

1

0

0. 00

1 . 0 0

0. 70

70. 2

0. 70

2 7 8 . 8 6

30 .7 6

0. 11

0 . 8 9

0. 62

70

22. 2

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

14

ABSTRACT: Run-of-the-river hydroelectric power projects in the Himalayan Region are developed on the principle of sustaining reservoir life by sediment management. Sediment management is generally achieved by sluicing or drawdown flushing through low level outlets during peak flows. The sedimentation in reservoirs depends on various factors such as reservoir geometry, flow and sediment characteristics and reservoir operation schedule. Hence, design and operation of such projects is highly site specific and simulation using numerical and physical models is essential for optimizing the design during planning stage. One dimensional numerical model is useful for predicting long term sediment deposition pattern in elongated reservoirs. In this paper, reservoir sedimentation studies carried out using numerical model simulations the run-of-the-river Punatsangchhu- I Hydro Electric Project (1200 MW) located on Punatsangchhu river in Wangdue District, Bhutan is presented. Simulations using one dimensional model HEC-RAS 4.1 were carried out to predict the sedimentation profiles along the reservoir covering a reach of about 18.5 km upstream of dam. Sediment rating curve was developed from available suspended sediment data. Simulations were carried out to predict the sedimentation profile after various duration of reservoir operation. It was observed that sedimentation in the reach from about 10.5km to 12.5km upstream of dam axis is very high. Simulations were continued for reservoir operating at MDDL till the sediment deposition at dam reached the spillway crest level. Hydraulic flushing is proposed to restore the reservoir capacity. Keywords: Run-of-the-river, sediment management, numerical model, reservoir sedimentation, Punatsangchhu- I H.E. Project 1. INTRODUCTION: Hydropower projects in Himalayan region are nowadays developed as run-of-the river schemes. The rivers in this region carry huge quantity of sediment load during monsoon season and the reservoirs gets silted up within a few years of operation. The life of such projects can be sustained by proper sediment management. Sediment management is generally achieved by sluicing or drawdown flushing through low level outlets during peak flows. The choice of the most efficient method depends on various factors such as reservoir geometry, flow and sediment

MANIT Bhopal

Page 107

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

characteristics and reservoir operation schedule. Run-of-the-river hydropower projects are generally developed at head reaches of perennial rivers by diverting available water to utilize the high elevation difference for power generation. Since sediment concentration is generally very high during peak flow season, these reservoirs are operated at MDDL by sluicing during monsoon. Since sedimentation problems of such projects are highly site specific, design of various components and operation schedule required to be optimized by hydraulic model studies Sediment deposition pattern in reservoirs can be estimated using mathematical models. Many such models have been developed and are being applied to simulate sediment deposition in reservoirs. One dimensional (1D) numerical model can be applied to predict long term deposition in reservoirs. A detailed description of 1D modelling and review of some of the available sediment models were presented by Morris and Fan (1997). Detailed review of the reservoir sedimentation and flushing processes including case studies, numerical, and physical models are reported by Batuca and Jordan (2000). Guidelines for predicting long term reservoir sedimentation and description with representative case studies of 1D and 2D sediment transport models were presented by Basson (2007). Mike 11(1D), RESSASS (1D), GSTARS (Quasi 2D), Mike 21(2D) (Quasi 3D) and a 3D model applied to the sedimentation studies of Three Gorges Project, China were described (Basson, 2007). Jungkyu Ahn and C T Yang (2010) studied the reservoir sedimentation and flushing processes of Xiaolangdi Reservoir on the Yellow River in China using GSTARS3 model. Nils Reidar B. Olsen and Stefan Haun (2010) reported application of a 3D numerical model with an adaptive grid for flushing of the Kali Gandaki reservoir in Nepal. Seyed Hossein Ghoreishi, et.al (2010) simulated the process of sediment flushing by a three dimensional numerical model based on Reynolds Averaged Navier-Stokes (RANS) equations. Application of a 2D numerical model (CCHE-2D) to simulate the sedimentation along a 150 km reach of the Aswan High Dam Reservoir, Egypt, was presented by Ahmed and Ahmed (2013). Isaac et al. (2013) reported application of 1D numerical model for predicting the reservoir sedimentation and geometrically similar scale physical model for hydraulic flushing of sediment from reservoir of Chamera-II reservoir, India. In this paper, 1D numerical model based on HEC-RAS used to simulate the reservoir sedimentation of Punatsangchhu –I H. E. project on Punatsangchhu river, Bhutan is presented. The project has been proposed as a run-of-the-river project with the provision for annual flushing of reservoir through low level sluice spillways to remove deposited sediment. The project site is in the Himalayan ranges where the sediment load in rivers is generally high during monsoon season. 1 D numerical model has been used to predict the long term sedimentation profiles. 2. STUDY AREA DESCRIPTION: The Punatsangchhu –I project is located on Punatsangchhu river, between 8 km and 16 km downstream of Wangdue Bridge, Bhutan. The dam site is about 80 km from Thimphu. The rivers Phochhu and Mochhu rises from the snow covered peaks of the

HYDRO 2014 International

Himalayan ranges in the North-West Bhutan at an elevation of about 7000m and join at Punakha to form the river Punatsangchhu. The Punatsangchhu River has a total length of about 320 km from its source in Bhutan to its confluence point with Brahmaputra in Assam. Its course in Bhutan has a length of about 250 km. The catchment area of Punatsangchhu river upto dam site extends from latitude 27015‟N to 28030‟N and longitude 89015‟ E to 90030‟ E. The total Catchment area upto the project site is 6390 km2 out of which 3115 km2 is snowfed area and the remaining 3275 km2 is rainfed area.

Figure 1. The location plan for Punatsangchhu-I H. E. Project is presented in. The project complex consists of a 136 m high (from deepest foundation level) concrete gravity dam, 7 numbers of sluice spillways (8 m width and 14.65 m height with crest at El.1166 m), 4 intakes with crest at El. 1182 m, 300 m long desilting basins and 9 km long and 10m diameter circular Head Race Tunnel (HRT). The sluice spillways are designed for the Probable Maximum Flood (PMF) of 11500 m3/s and 4300 m3/s GLOF. The reservoir is to be operated between Full Reservoir Level (FRL) of El.1202 m and Minimum Draw Down Level (MDDL) of El. 1195 m. The gross storage capacity of the reservoir is 25 Mm3 and live storage is 16 Mm3 with 3. NUMERICAL MODEL: Sediment transport and deposition in reservoirs are three dimensional in nature. The physical processes are very complex and could be simulated using three dimensional (3D), two dimensional (2D) or one dimensional (1D) numerical model. A number of such commercial or free models are available. The selection of the model depends on the objectives of the study, availability of data and computational resources. 3D numerical models are essential to reproduce complex flow patterns and flow near hydraulic structures. However, simplification with 1D approach is well suited for narrow and gorge type reservoirs where longitudinal processes are prevailing and if long periods need to be simulated. Based on the above criteria, the one dimensional model, HEC-RAS 4.1 (USACE, 2010) developed by the U.S. Army Corps of Engineers at the Army‟s Hydrologic Engineering Centre was selected in the present study to simulate

MANIT Bhopal

Page 108

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

the sediment deposition in the reservoir of Punatsangchhu –I H.E. project. The sediment transport module of HEC-RAS simulates streambed profile changes resulting from varying river flow and tail water conditions. The model is based on 1D gradually varied flow hydraulics and sediment transport theory. Water surface profiles and other hydraulic parameters such as water velocity, hydraulic depth, hydraulic roughness, energy slope, and width at each cross section are computed from one cross section to the next by standard step method according to the energy equation. If water surface profiles are rapidly varied, momentum equation is applied. HEC-RAS uses the quasi-unsteady flow approach for sediment transport simulation. The continuous flow hydrograph is approximated with a series of discrete steady flows of specific durations. Hydrodynamic computations are performed for each of these steady flows and transport parameters are generated at each cross section. Flow durations are subdivided into computational time steps, since bathymetry updates are required more frequently than the flow increment durations. The geometry file is updated and new steady flow hydrodynamics are computed at the beginning of each computational time step (Gibson et. al., 2006). The sediment continuity (Exner) equation is then solved over the control volume associated with each cross section, computing from upstream to downstream. At the end of each computational time step, the aggregation or degradation is translated into a uniform bed change over the entire wetted perimeter of the cross section. The cross sectional station-elevation information is updated and new hydrodynamic computations performed before the transport capacity is computed for the next sediment routing iteration.

Figure 2. River system schematic For hydraulic computations, the roughness coefficient was simulated by Manning‟s „n‟. In the present study, steady flow computations were carried out for calibrating the model by adjusting the „n‟ value. Water levels observed during flood on 26th May 2009 and 3rd July 2010 were used for calibration of the model. Water levels along the river reach was matched with the model results. The results are presented in Fig. 3. Manning‟s „n‟ was assumed as 0.048 for the channel portion.

The main input data required for HEC-RAS include cross sections of river reach, inflow hydrograph, grain size distribution curve of bed material, sediment Vs discharge relation, rule curve for reservoir operation and sediment transport equation (USACE, 2010). Figure 3. Observed and computed water surface profiles

4. MODEL SETUP: The 1D numerical model of river Punatsangchhu covering a reach of about 18.5 km upstream of dam and 1.5 km downstream was developed using HEC-RAS. The river schematic was developed as per the river plan. The river geometry was reproduced in the model using the river schematic and the cross section data. Cross sections data was available at 35 m interval near the dam axis and at 500m interval in the remaining reaches. Fig. 2 gives the river schematic with the locations of cross sections.

HYDRO 2014 International

4.1 Upstream boundary: The inflow discharge hydrograph and total sediment load data were specified as upstream boundary condition for the simulations. Daily observed discharge data was available at the Wangdue rapid gauging site for the period from July 1992 to July 2009. The above daily discharge hydrograph after correcting errors and filling the gaps was used as the upstream boundary in the simulation runs. The inflow hydrograph was repeated for longer duration simulations. The inflow hydrograph is presented in Fig. 4.

MANIT Bhopal

Page 109

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 material gradation curves (Fig.6) at five locations upstream of dam axis were available and hence used in the simulations.

Figure 4. Inflow hydrograph

Figure 6. Bed material gradation curve

Suspended sediment concentration along with corresponding discharge observations were made at the Wangdue rapid gauging site for the period from July 1992 to July 2009. Using the above data sediment rating curve was developed and the same is presented as Fig.5. The sediment rating curve was verified with the sediment data available from gauge site at 1.5 km upstream of dam axis established by M/s WAPCOS. Sediment data from January 2010 to June 2010 was available and used for verification.

5. RESULTS AND DISCUSSION Simulations were carried out to predict the sedimentation profile after different durations of reservoir operation. Initially, the sediment rating curve developed from observed data at Wangdue rapid gauging site was used as the upstream boundary for sedimentation runs. Since no measured data was available, the bed load was assumed as 20% of the suspended load and the total load was specified at the upstream boundary. The gauging site was located at the pool area just upstream of the rapid such that most of the incoming sediment settles in the reaches immediately upstream. Hence the sediment concentration measured at the gauging site was observed to be less. In order to account for the unmeasured sediment load, the observed values were increased by 4 times and the rating curve was modified. Simulations were conducted for reservoir operating at FRL and MDDL. It was observed from the sedimentation profile obtained after 5 years of reservoir operation that deposition takes place in the reaches between 4km to 6km and 9km to 13 km from dam axis. The river slope is mild and the cross sections are comparatively wider in the upstream reaches. Hence sedimentation in the above reach was observed to be high. The area near the dam and intakes remains clear of sediment deposition.

Figure 5. Sediment rating curve 4.2 Downstream boundary: The reservoir operation level at dam was specified as downstream boundary in simulations. Simulations were carried out by maintaining the reservoir water level at the FRL of El. 1202 m and at the MDDL of El.1195 m. 4.3 Bed material gradation curve: In HEC-RAS, the sediment continuity equation is solved separately for each grain size and material is added or removed to the active layer. Hence it is required to specify the initial grain size distribution of the bed material. In the present study, bed

HYDRO 2014 International

To get the pattern of bed profile near dam and intake area, simulation runs were carried out by specifying equilibrium sediment load condition at the upstream boundary. The bed profiles obtained by the simulation of daily hydrograph for a period from January 1992 to July 2025 (about 33 years), and reservoir operating at MDDL is presented in Fig. 7. It was observed that the sedimentation level at the dam axis reached about the spillway crest level of El.1166 m. The delta deposition in the pool area between 1.5 km and 5.5km was progressing. The cross section of river in the reach from about 10.5km to 12.5km upstream of dam axis is very wide compared to the sections just upstream and downstream. Hence sedimentation in the above reach was observed to be very high.

MANIT Bhopal

Page 110

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 load at upstream boundary indicated that sediment deposition will reach the dam and spillway crest level after about 33 years of reservoir operation without annual flushing. Simulations with annual flushing indicated that sediment deposition will move from upstream towards the dam during drawdown flushing. It was observed from the results of simulation that due to the flatter bed slope and wider river sections, sedimentation is high in the upstream reaches of reservoir. ACKNOWLEDGEMENTS:

Figure 7. Bed Profiles after different years with reservoir at MDDL Sediment management in the reservoir of Punatsangchhu – I project is proposed by drawdown flushing during monsoon when the sediment concentration exceeds the design value of desilting basins. Hence in order to obtain the sedimentation profile with annual flushing, simulations were carried out by lowering the water level at dam axis during annual peak flows. The bed profiles obtained after 5 years with and without annual flushing are presented in Fig. 8. It was observed that during flushing, the sediment deposition from the area around 5 km is moving downstream towards the dam axis.

Figure 8. Bed Profiles after 5 years with and without drawdown flushing 6. CONCLUSIONS: The Punatsangchhu – I H. E. project is planned as a run-of-the river scheme. Sediment management in the reservoir is proposed by annual drawdown flushing during peak flow and sluicing during monsoon by operating the reservoir at MDDL. In this study, one dimensional numerical model was used to obtain the sedimentation profile of reservoir under different operating conditions. Simulations with the measured sediment inflow rate indicated very low deposition when the reservoir was operated at FRL and MDDL. No sediment deposition was observed near the dam and intake area. Simulations with equilibrium sediment

HYDRO 2014 International

Kind permission given by Shri. S. Govindan, Director CWPRS, Pune for publishing the paper is acknowledged with thanks. Support and input given by Dr.(Mrs.) V. V. Bhosekar, Joint director, CWPRS, officials of WAPCOS India and PHPA, Bhutan are thankfully acknowledged. Authors also express thanks to officers and staff members of HAPT division, CWPRS, Shri. P. S. Kunjeer, and Shri. S.A. Kamble, Research Officers for their co-operation and support in conducting the studies. REFERENCES: i. Ahmed Moustafa, Ahmed Moussa (2013). Predicting the deposition in the Aswan High Dam Reservoir using a 2-D model. Ain Shams Engineering Journal 4, 143–153. ii. Basson, G. (2007). Mathematical Modelling of Sediment Transport and deposition in Reservoirs, Guidelines and Case Studies. ICOLD Bulletin No.140. International Commission on Large dams, 61,avnue Kleber, 75116, Paris. iii. Basson, G. (2009). Sedimentation and sustainable Use of Reservoirs and river Systems. ICOLD Bulletin No.147. International Commission on Large dams, 61,avnue Kleber, 75116, Paris. iv. Batuca, D.G., and Jordan, J.M. (2000). Silting and desilting of reservoirs, A.A. Balkema, Rotterdam. v. Gibson, S., Brunner, G., Piper, S., and Jensen, M. (2006). Sediment Transport Computations with HEC- RAS, Proceedings of the Eighth Federal Interagency Sedimentation Conference (8thFISC), April2-6, 2006, Reno, NV, USA. vi. Isaac, N., Eldho, T. I., Gupta, I. D. (2013). Numerical and physical model studies for hydraulic flushing of sediment from Chamera-II reservoir, Himachal Pradesh, India, ISH Journal of Hydraulic Engineering, DOI: 10.1080/09715010.2013.821788. vii. Jungkyu Ahn, Chih Ted Yang, (2010). Simulation of Xiaolangdi Reservoir Sedimentation and Flushing Processes, 2nd Joint Federal Interagency Conference, Las Vegas, NV, June 27 - July 1. viii. Morris, G.L., and Fan, Jiahua (1997). Reservoir sedimentation hand book. McGraw-Hill Book, New York. ix. Nils Reidar B. Olsen & Stefan Haun (2010). Free surface algorithms for 3D numerical modelling of reservoir flushing. River Flow 2010 - pp 11051110. x. Seyed Hossein Ghoreishi, Mohammad Reza Majdzadeh Tabatabai (2010). Model study reservoir flushing. Journal of Water Sciences Research, ISSN: 2008-5338 Vol.2, No.1, Fall 2010, 1-8, JWSR. xi. Sonam Choden (2009).Sediment Transport Studies in Punatsangchu River, Bhutan. Water Resources Engineering, Department of Building and Environmental Technology, Lund University, P.O.Box 118, SE-221 00 Lund xii. USACE. (1993). Scour and deposition in river and reservoirs: HEC 6 – User‘s manual. US Army Corps of Eng., Hydrol. Eng. Center, 690 Second Street, Davis, CA, 95616–4687. xiii. USACE. (2010). HEC-RAS River analysis system – Hydraulic reference manual and user‘s manual. US Army Corps of Eng., Hydrol. Eng. Center, 690 second street, Davis, CA, 95616. xiv. USBR. (2006). Erosion and sedimentation manual. US Department of the Interior Bureau of Reclamation.

MANIT Bhopal

Page 111

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Sedimentation Assessment in Nath Sagar Reservoir (Jayakwadi Project) of Maharashtra by Remote Sensing Technique – A Case Study Prakash Bhamare1 Manoj Bendre2 Ravindra 3 Shrigiriwar Mahendra Nakil4 Sudhir Kalvit5* Maharashtra Engineering Research Institute Dindori Road, Nashik – 422004, Maharashtra, India Phone 0253-2534676 E mail – [email protected] *Corresponding Author ABSTRACT: Jayakwadi irrigation project is a major project in Maharashtra State constructed on the river Godavari in the year 1975-76 with a gross and live storage potential of 2909 Mm3 and 2171 Mm3 respectively. The reservoir has been named as Nath Sagar reservoir after well known Marathi Saint Eknath of the 16th Century. The project has been instrumental in the economic development of Marathwada region of the State. However, since last few years, due to vagaries of monsoon and inadequate run off from the catchment, the reservoir has been facing shortage of water. The reservoir was filled up to F.R.L. hardly three to four times in last decade. More over sedimentation in this reservoir has been another issue before the reservoir management authority. Inadequacy of water storage and the reduction in storage potential of the reservoir on account of sedimentation have forced the reservoir authority to conduct sedimentation assessment survey of this reservoir for assessing the net storage available in the live storage zone. The sedimentation assessment survey was entrusted to Maharashtra Engineering Research Institute (M.E.R.I.) Nashik by Jayakwadi reservoir authorities. In April 2014, M.E.R.I., conducted the survey by satellite remote sensing technique using IRS LISS III images, and the present live storage capacity between Full Reservoir Level (FRL) and Minimum Draw Down Level (M.D.D.L.) had been estimated. A revised Elevation –Area – Capacity table at 0.10 meter interval had been prepared for the live storage zone which can be very useful for the reservoir management authority while operating the reservoir. (Key words – Dead Storage, Live storage, D.G.P.S., M.D.D.L, bathymetry.) 1.0 INTRODUCTION Apart from the hydrological factors, deforestation, rapid urbanization, developmental activities in the catchment area such as construction of roads, railway lines, land leveling and terracing excessive quarries and mining etc are some other important factors responsible for rapid erosion of the land in the catchments of reservoirs. The soil erosion in the catchment accelerates the process of sedimentation in reservoirs. The sedimentation results in reduction of storage capacity of reservoirs. Many lakes and reservoirs, which are important fresh water resources, are under the threat of sedimentation today. The reduction in water storage potential of the multipurpose reservoirs affects the entire irrigation and domestic water planning. Therefore, sedimentation is a matter of concern for the reservoirs in the context of their utility and useful life. The

HYDRO 2014 International

siltation in reservoirs does not have uniform pattern everywhere. It is obvious because climatic and topographical conditions and the land use pattern in the catchment area are different in different regions of the state. Periodic reservoir capacity assessment surveys provide useful information about storage availability at different levels in different periods which is important scheduling the water use effectively. Remote sensing based reservoir sedimentation surveys are essentially based on mapping of water-spread areas at the time of satellite over pass. It uses the fact that water-spread area of the reservoir reduces with the sedimentation at different levels. The water-spread area and the elevation information are used to calculate the volume of water stored between different levels. These capacity values are then compared with the previously calculated capacity values to find out change in capacity between different levels. The Maharashtra Engineering Research Institute which is the Research Wing of the State‟s Water Resources Department has done substantial work in the field of reservoir sedimentation assessment. First sedimentation assessment survey of Nath Sagar reservoir was conducted by Maharashtra Engineering Research Institute, Nashik by satellite remote sensing technique using digital images of IRS 1B satellite with LISS II sensor (36 m spatial resolution) for the period between years 1994 - 1997. The next survey of Nath Sagar reservoir was conducted using most of the digital images of RESOURCESAT 1 satellite with LISS III sensor (24 m spatial resolution) for the period between years 2011-2013. Temporal sedimentation assessment surveys are useful to keep the content table of reservoir updated which is a pre requisite for realistic planning of reservoir storage. 2.0 OBJECTIVES OF THE SURVEY The sedimentation assessment study was conducted with the following objectives  To estimate the present live storage capacity of reservoir  To update Elevation-Capacity curve for the live storage zone of reservoir.  To estimate storage capacity loss in reservoir since it‟s first impounding.  To update the content table of the reservoir for live storage zone. 3.0 STUDY AREA The Nath Sagar reservoir lies between Latitude 19 0: 19‟: 13” and 190: 41‟: 46” N and Longitude 740: 49‟: 23” and 750: 24‟: 22” E. The reservoir is constructed on river Godavari, near village Jayakwadi in Paithan Taluka of Aurangabad district. The project comprises earthen dam of nearly 10.5 Km in length. Total catchment area of the reservoir is 21750 sq. km. The designed gross storage capacity of the reservoir at FRL 463.906 m is 2909 Mm3 and live storage capacity between FRL & MDDL is 2170.92 Mm3. The MDDL of the reservoir is at R.L. 455.524 m. Designed dead storage capacity is 738.08 Mm3. The reservoir was first impounded in the year 1975-76.

MANIT Bhopal

Page 112

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

4.0 DATA USED FOR THE PRESENT STUDY

RESOUR CESAT 1

LISS III

16 Dec 2007

462.8 97

(A) Field Data

RESOUR CESAT 1

LISS III

22 Nov 2007

463.3 27

RESOUR CESAT 1

LISS III

23 Oct 008

463.7 9

RESOUR CESAT 1

LISS III

5 Oct 2007

463.9 06

Following field data required for this survey was obtained from Jayakwadi Project authority. i) Reservoir Levels for given dates of the satellite pass ii) Reservoir F.R.L. and M.D.D.L. iii) First year of reservoir impounding. (B) Satellite Data NRSC website was browsed and a list of cloud free dates of RESOURCESAT 1 and RESOURCESAT 2 satellite pass over Nath Sagar reservoir was prepared for the period between Year 2011 and 2013. The selection of the satellite images was done after studying the draw down pattern of the lake levels and selected satellite data was procured from the NRSC Hyderabad. In all, total 19 images of RESOURCESAT 1 and RESOURCESAT 2 satellites together, with LISS III sensor having a spatial resolution of 24 m are used for this survey. These satellite images were of different water levels between R.L.455.066 m to F.R.L 463.906 m. Out of these, 14 images were of period between Oct 2011 to Jan 2014 and 5 images were of the year 2007-08. Since Nath Sagar reservoir did not have full storage in last 6-7 years, the images of old period (year 2007-08) had to be used to cover the study up to F.R.L. avoiding extrapolation of result. Thus the present study covered 100% of live storage zone. Table 1 gives the dates of satellite passes with respective water levels.

5.0 METHODOLOGY The satellite images procured from National Remote Sensing Centre were already rectified (geo-referenced). Hence preprocessing of images was not necessary. The images were analysed digitally using standard image analysis software. Classification technique was adopted for the analysis and the water spread areas of the reservoir in all the images were measured. The following flow chart describes the methodology in brief.

Table- 1.Details of Satellite pass, sensor, path and row and water levels Satellite

Sensor

Date of Pass

RESOUR CESAT 1

LISS III

06 May 2013

RESOUR CESAT 1

LISS III

12 Apr 2013

455.3 59

RESOUR CESAT 2

LISS III

19 Nov 2012

455.9 45

RESOUR CESAT 2

LISS III

11 Feb 2013

456.1 24

RESOUR CESAT 1

LISS III

30 Jan 2013

456.2 80

RESOUR CESAT 1

LISS III

13 Dec 2012

456.7 60

RESOUR CESAT 2

LISS III

05 Apr 2012

457.7 91

RESOUR CESAT 2

LISS III

24 Mar 2012

Elev ation m 455.0 66

457.9 44

RESOUR CESAT 1

LISS III

1 Jan 2014

458.6 94

RESOUR CESAT 1

LISS III

21 Oct 2013

459.2 67

RESOUR CESAT 2

LISS III

24 Jan 2012

459.4 59

RESOUR CESAT 1

LISS III

19 Dec 2011

460.2 88

RESOUR CESAT 2

LISS III

13 Nov 2011

461.1 62

RESOUR CESAT 1

LISS III

8 Oct 2011

461.7 00

RESOUR CESAT 1

LISS III

HYDRO 2014 International

9 Jan 2008

Table 2 shows the Water Spread Areas (WSA) of Nath Sagar reservoir in all the images corresponding to their water levels Table 2. Water spread areas estimated from satellite data

462.3 95

MANIT Bhopal

Date of pass 06 May 2013 12 Apr 2013 19 Nov 2012 11 Feb 2013 30 Jan 2013 13 Dec 2012 05 Apr 2012 24 Mar 12012 Jan 2014 21 Oct 2013 24 Jan 2012 19 Dec 2011 13 Nov 82011 Oct 2011 9 Jan 2008 16 Dec 2007

Elevation in m.

Area in Mm2

455.066

112.54

455.359

124.47

455.945

137.84

456.124

141.78

456.280

147.72

456.760

160.81

457.791

182.38

457.944

185.46

458.694

202.97

459.267

223.05

459.459

231.34

460.288

248.71

461.162

271.05

461.700

290.2

462.395

311.89

462.897

325.31

Page 113

International Journal of Engineering Research Issue Special3 22 Nov 2007 23 Oct 008 5 Oct 2007

463.327

343.27

463.79

367.52

463.906

371.69

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Figure -2 Satellite images of Nath Sagar Reservoir of different dates in order of reducing water levelimage Jayakwadi Project (Nathsagar reservoir) Satellite Remote Sensing based Elevation Vs revised area curve for live storage zone 400

Interpolation of Water spread area (WSA) at Regular Interval

y = 0.1776x3 - 1.6011x2 + 29.146x + 111.87 R2 = 0.9991

300

Revised Area in Mm2 ---->

For the present survey cloud free satellite images of different water levels for the reservoir portion between R.L. 455.066 m and F.R.L. 463.906 m were available. Water levels on the date of satellite pass for selected satellite data were not at regular interval. To get WSA values at regular elevation interval, a curve was plotted between Elevation and the Revised Area and a best fit polynomial equation of third order was derived for the graph. The best fit equation is as follows. y = 0.1776x3 - 1.6011x2 + 29.146x + 111.87 R2 = 0.9991 (R = Coefficient of co-relation) where x = Elevation difference in meters (measured above R.L. 455.00 m) y = Water spread area in Mm2 Using this equation, the Water Spread Areas at regular interval of elevation between R. L. 455.00 m and F.R.L. 463.906 m have been worked out.

Third order polynomia equation for best fit curve for the graph is as below

350

where R = Coefficient of co-relation x = elevation measured above R.L. 455.00 m y = revised water spread areas

250

200

150

100

50

0 0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

Elevation measured above R.L. 455.00 m considering R.L. 455.00 m as datum ----->

Figure -3 Graph between R.L. and respective water spread areas (estimated from satell

Calculation of Reservoir Capacity Computation of reservoir capacity at different elevations has been done using following prismoidal formula. V = h/3*(A1 + A2 + SQRT (A1 * A2)). Where V- Reservoir capacity between two successive elevations h1 and h2 h- Elevation difference (h2 – h1) A1 and A2 are areas of reservoir water spread at elevation h 1 and h2.

Figure -4Graph showing comparison of Live storage capacity as per different surveys

Sr. No

1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

HYDRO 2014 International

MANIT Bhopal

Elevation

Original Capacity

in meters 2 455.524 455.600 456.000 456.499 456.999 457.499 457.999 458.499 458.999 459.499 459.998 460.499 460.998 461.498 461.997 462.498 463.000 463.500

3 0 11.171 75.559 157.896 240.234 333.743 434.51 535.377 648.323 771.203 894.082 1028.98 1178.431 1327.038 1486.624 1659.565 1833.561 2013.35

Revised live storage capacity as per 1994-96 survey 4 0 10.288 66.305 140.744 220.658 306.180 397.586 495.150 599.149 709.857 827.308 952.504 1084.722 1225.008 1373.075 1530.115 1696.166 1870.487

Revised Live storage Capacity as per 2012-13 survey 5 0 9.711 63.379 136.247 215.655 301.294 393.064 490.930 594.924 705.144 821.516 944.992 1074.884 1212.315 1357.167 1510.833 1673.649 1845.251

Page 114

International Journal of Engineering Research Issue Special3

19

463.906

2170.935

2018.782

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

1991.987

6.0 RESULT AND DISCUSSION The result of the present survey and its comparison with original survey and year 1994-96 survey are given in following table. Table – 3 showing the comparison of original live storage capacity with that of year 1996 survey and year 2012-13 survey

Table 3 Description

As per original survey year 197576

Live storage capacity

As per survey of year 1994-96

2170.935

2018.782

-------

0.35

As per survey of year 2012-13

1991.987 in Mm3 Average Annual loss

0.23

%

their prediction have become an important process in the part of a hydrologist. In recent years various models like autoregressive (AR) integrated with Moving Average (MA), Neural Network (NN), Fuzzy Theory, Genetic Algorithm (GA), Support Vector machines (SVM) and Wavelet Transformation have been used in analysis and prediction of the hydrological data. These models were not only used individually for analysis of the data, but also a combination of these models were used for better representation of the data and subsequent predictions. The models can be developed using the standard software packages as available and with R/Matlab. In this paper the ARIMA, ANN and the Wavelet combined with ANN was analyzed for better performance and validated with the data as available for the catchment. The model efficiency was also reported in various parameters like root mean square error (RMSE), coefficient of correlation (R) and other model specific parameters. Keywords: Wavelet, Neural Network, Autoregressive Model

Revised content table for the live storage zone has been prepared at 0.1 m contour interval which can be of great use for the reservoir management authority during reservoir operation. Revised Live storage capacity of Nath Sagar reservoir between M.D.D.L. 455.524 m and FRL 463.906 m is estimated to be 1991.987 Mm3 for the year 2012-13 as against Original Live storage capacity of 2170.935 Mm3 between these levels, with a loss of 178.948 Mm3 (8.24 %). The average annual percent loss in live storage for the period of 36 years between 1976 and 2012 works out to 0.23% which is not severe. Sedimentation in the dead storage zone i.e. below M.D.D.L. 455.52 m could not be estimated by remote sensing method. For this, a hydrographic survey is necessary.

7.0 REFERENCES i. Technical Report on revised storage capacity assessment of Jayakwadi reservoir by satellite remote sensing technique. (year 2014), Maharashtra Engineering Research Institute, Nashik – 4

ii. Figure -3 Graph between R.L. and respective water spread areas (estimated from satellite images

Hydrological Data Modelling Using Wavelet, Neural Network And Ar Models G.Khadanga1, B.Krishna2 Scientist, National Informatics Centre, CGO Complex, New Delhi 2 Scientist, NIH, Kakinada, Deltaic Regional Center, Kakinada Email: [email protected] 1

ABSTRACT: Hydrological data like rainfall, runoff, evapotranspiration, water table, reservoir water level etc. and

HYDRO 2014 International

1. INTRODUCTION: A time series is a sequence of observations that are arranged according to the time of their outcome. In time series the physical quantity and the sequence and the order of the data collection is very important. Meteorology records like hourly wind speeds, daily maximum and minimum temperatures, daily monthly and annual rainfall, discharge data of a river or dam are few examples of the time series data. Various statistical approaches like regression, auto regression, auto regressive integrated moving average time series modeling, stochastic approaches, machine learning, data mining, ANN, fuzzy set, neuro fuzzy, support vector machine, fourier transform, wavelet combines with ANN have been used to model the time series data. The analysis of the nonlinear behavior and raise the forecast precision and lengthen the forecasted time are a challenging task in time series modeling. In this paper the ARIMA model is explored with the sample data using R as modeling tool. Then the other modeling tools like ANN, Wavelet and combined with ANN are used for the rainfall data analysis and the models output was interpreted. 1.1 Arima Model in R: The acronym ARIMA(p,d,q) stands for "Auto-Regressive Integrated Moving Average." Lags of the differenced series appearing in the forecasting equation are called "autoregressive" terms, lags of the forecast errors are called "moving average" terms, and a time series which needs to be differenced to be made stationary is said to be an "integrated" version of a stationary series. In ARIMA the p is the number of autoregressive terms, d is the degree of first differencing, and q is the order of the moving average part. The auto.arima() function of R (open source software package for statistical modeling) uses the Hyndman and Khandakar algorithm which combines the unit root tests, minimization of the AICs (Akaike‟s Information Criterion) and MLE to obtain the ARIMA model. The daily rainfall data of the test location is

MANIT Bhopal

Page 115

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

0.8 0.4

Daily Rainfall in MM

ACF

0.0

collected for 10 years and the plot of the data and the 1st difference is shown in fig1. The decomposition of the daily time series data is shown in fig2. The auto correlation and the partial auto correlation is shown in fig3. As the ACF is dropping to zero the time series is stationary. The PACF is exponentially decaying and sinusoidal and there is a significant spike at lag 2 in ACF, but none beyond lag 2.

0

10

20

30

Days

0.3 0.2 0.1 0.0

Daily Rainfall in MM

PACF

0

10

20

30

Days

Figure 4. The ACF and PACF of the Rainfall Data Forecasts from ARIMA(2,0,3) with non-zero mean

20

50 100

10 0 -10

6

0

50 100

remainder

200

2

4

trend

8

0

5

seasonal

10

0

data

200

30

Figure 1. Daily Rainfall Data of the series with 1st difference

1990

1995

2000

2005

2010

7600

7620

7640

7660

7680

7700

Figure 5. The forecast of the rainfall data using auto.arima R function

time

Figure 2. Decomposition of the Daily Rainfall Data The summary of the fit is shown below: Coef ar1 ar2 ma1 ma2 ma3 intercept -0.0727 0.6202 0.4345 -0.5172 -0.1872 3.1970 s.e. 0.1570 0.0920 0.1579 0.1183 0.0378 0.1823 sigma^2 estimated as 97.96: log likelihood=-28465.15, AIC=56944.29 AICc=56944.31 BIC=56992.91 Figure 3. The summary of the fitness of the ARIMA Model

2. Artificial Neural Network (ANN) The Artificial neural network (ANN) offers a quick and flexible means of modeling hydrologic data analysis and prediction. ANN tolerate imprecise or incomplete data, approximate results and are less vulnerable to outliers. The ANNs can be described either as a mathematical and computational model for non-linear relationship, data classification, clustering and regression or as simulations of the behavior of collections of the biological neutrons. The feed-forward multilayer perceptron (MLP) is the most commonly used ANN in hydrological applications. The first step in back propagation learning is the initialization of the network. The structure of the network is first defined. In the network, activation functions are chosen and the network parameters, weights and biases, are initialized. The parameters associated with the training algorithm like error goal, maximum number of epochs (iterations), etc, are defined. Then the training algorithm is called. After the neural network has been determined, the result is first tested by simulating the output of the neural network with the measured input data. This is compared with the measured outputs. Final validation is carried out with

HYDRO 2014 International

MANIT Bhopal

Page 116

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

independent data. The input values were normalized before use in the ANN. The result of the training using the feed forward network is shown in the table. The first two year daily data was taken for training and one year data was taken for validation. The regression coefficient is found out to be 0.397 (Fig6) after lots of trial with different models. The model with 4 input layer, one hidden layer found with highest regression coefficient. The result is not very encouraging however moderate predictions can be taken up with this model. As the result is not very encouraging an attempt is made for analyzing the data using both wavelet and neural network.

Figure 7. Original data set is broken into wavelets

3. Wavelet The wavelet analysis has been used as alternative to Fourier transform. The fourier transform mainly concentrate on the frequency domain where as the wavelet analysis can provide the exact locality of any changes in the dynamic patterns of the sequences. Wavelet analysis is the breaking of a signal into shifted and scaled version of the original data. Sometimes it is also called as multi resolution analysis. The original signal is passed through loss pass and high pass filters and emerges as two signals as Approximations (A) and Details (D). The approximations as low–scale and high frequency components of the signal. The details are the highscale and low frequency components. The Daibechies and Morlet wavelet transforms are more frequently used for hydrological time series data.

Figure 8. Forecast & correlation coeff.

The Decomposed details (D) and approximations (A) are taken as inputs into a neural network and then resultant wavelets were combined to form the original data. Optimal structure of the neural network (input layers, number of hidden, optimal parameters of the neural network for train, transfer functions) nodes was used to get the best performance. The output node is taken as one step ahead of the original time step. 3. RESULTS AND ANALYSIS

Figure 9. Training data and model output data

The daily rainfall data for the first two year is taken as the calibration data and one year data is taken as validation data. The original time series data is decomposed in to details and approximate components using the wavelet transform algorithms (DB5, D1, D2, D3, A4). The original timeseries and decomposed parts are shown in fig 7. Figure 6. Regression coeff in ANN

Table 1. Staistics of WNN and ANN for Calibration and Validation period Model x(t)=f(x[t-1],x[t2],x[t-3],x[x-4]) WNN ANN

HYDRO 2014 International

MANIT Bhopal

Validation Calibration RMSE R 20.05 0.697 13.79 0.419

RMSE 13.79 8.918

R 0.734 0.197

Page 117

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

4. CONCLUSIONS The performance of the model were experimented with various combinations and the best performance is found with regression coefficient 0.985 (Fig7). This is much better than the ANN case. The observed and model output is shown in the fig 9. The table 1 shows the various statistical parameters for the ANN and the WNN case. The coefficient of correlation is better in WNN. From the figure it is observed that the peak rainfall data is predicted with minimum errors. The forecasted values are well fitted to the 45 degree line. It was concluded that the best predication of the data is possible with WNN model. REFERENCES: i. Box, G. E. P. and G. M. Jenkins,(1976), ―Time Series analysis, forecasting and control‖ Holden day, Oakland, California. ii. Goel N.K., Stochastic Modeling of Hydrological Process, Training Course on Integrated Catchment Modelling, NIH, Roorkee, Nov. 2013. iii. Haykin, S. (1994), Neural Network: a comprehensive foundation. MacMillan, New York. iv. J.S. Yang, S.P. Yu, and G.-M. Liu.,‖ Multi-step-ahead predictor design for effective long-term forecast of hydrological signals using a novel wavelet neural network hybrid model‖, Hydrol. Earth Syst. Sci., 17, 4981–4993, 2013. v. Kottegoda N.T.(1979), Stochastic Water Resources Technology, John Wiley and Sons New York. vi. Nayak, P.C., Sudheer, K.P., and Ramasastri, K.S.( 2004a)‖ A Neurofuzzy Computing Techniques for modeling Hydrological Time Series:, Journal of Hydrology, 291(102):52-66. vii. Shumway, H. Robert., Stoffer S. David., ―Time Series Analysis and its Applications with R Examples‖, Third Edition, Springer, 2011. viii. Yevjevich, V., Stochastic Processes in Hydrology, May 1971.

Improved Neuro-Wavelet Model for Reservoir Inflow Forecast B.Krishna, Y.R.Satyaji Rao and R.Venkata Ramana Scientists, Deltaic Regional Center, National Institute of Hydrology, Siddartha Nagar, Kakinada-3, Andhra Pradesh. Email: [email protected] ABSTRACT : There is a need for forecasts of reservoir inflow events in order to: a basin wide consistency in management operations based on a thorough knowledge of variation in inflows, an improved capability for predicting and monitoring flood events. Using hybrid model or combining several models has become a common practice to improve the forecasting accuracy. The combination of forecasts from more than one model often leads to improved forecasting performance. An attempt has been made to find an improved method for accurate prediction of inflow by combining the wavelet technique with Artificial Neural Networks (WNN). Wavelet analysis effectively decomposes the main signal and diagnoses its main frequency component and abstract local information. The observed time series is decomposed into sub-series using discrete wavelet transform and then appropriate sub-series is used as an independent variable for the Neural Network

HYDRO 2014 International

model. Several hybrid models have been developed to forecast the inflow into Malaprabha reservoir in one day advance. The calibration and validation performance of the developed models is evaluated with appropriate global statistics. The results were compared with the standard models with undecomposed data. The application of wavelet based neural network models were found to be more effective as its prediction efficiency is more and its peak value is closer to observed value. Keywords: Inflow, Neural Networks, Training, Wavelet Decomposition 1. INTRODUCTION: Inflow is an important data for an optimal reservoir operation. The importance of an accurate flow forecast, especially in floodprone areas, has increased significantly over the last few years as extreme events have become more frequent and more severe due to climate change and anthropogenic factors. Data based forecasting methods are becoming increasingly popular in flood forecasting applications due to their rapid development times, minimum information requirements, and ease of real-time implementation. Using hybrid model or combining several models has become a common practice to improve the forecasting accuracy. The combination of forecasts from more than one model often leads to improved forecasting performance. An attempt has been made to find an alternative method for accurate prediction of inflow by combining the wavelet technique with Artificial Neural Networks (WNN). Artificial Neural Network (ANN) is widely applied in hydrology and water resource studies as a forecasting tool. In ANN, feed forward backpropagation (BP) network models are common to engineers. It has proved that BP network model with three-layer is satisfied for the forecasting and simulating in any engineering problem. Three-layered feed forward neural networks (FFNNs), which have been usually used in forecasting hydrologic time series, provide a general framework for representing nonlinear functional mapping between a set of input and output variables. Although ANN had been used extensively as useful tools for prediction of hydrological variables, it has also many drawbacks to deal with non-stationary data (Cannas et al., 2006). Wavelet analysis is a useful tool for non-stationary processes such as hydrological time series (Rajaee et al., 2011). Wavelet transform, which is a pre-processing decomposed technique, showed successful performance in hydrological applications. Several studies have been published that developed hybrid wavelet–ANN models. Wang and Lee (1998) developed a hybrid wavelet–ANN model to forecast rainfall–runoff in China. Rajaee et al., (2011) applied wavelet combined with neuro-fuzzy and ANN for sediment load prediction, Cannas et al. (2005) developed a hybrid model for rainfall–runoff forecasting. Okkan (2012) developed different models as Wavelet Neural Network (WNN) in combination with Discrete Wavelet Transform (DWT) and Levenberg-Marquardt based Feed Forward Neural

MANIT Bhopal

Page 118

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Networks (FFNN) and Wavelet Multiple linear Regression (WREG) for monthly reservoir inflow forecasting. 2. WAVELET ANALYSIS The wavelet transform is the tool of choice when signals are characterized by localized high frequency events or when signals are characterized by a large numbers of scale-variable processes. Because of its localization properties in both time and scale, the wavelet transform allows for tracking the time evolution processes at different scales in the signal. The continuous wavelet transform of a time series f (t) is defined as

1 f (a, b)  a



t b f ( t )  ( )dt  a  (1)

Where  (t ) is the basic wavelet with effective length (t) that is usually much shorter than the target time series f (t). The variables are a and b, where a is the scale or dilation factor that determines the characteristic frequency so that its variation gives rise to a `spectrum'; and b is the translation in time so that its variation represents the `sliding' of the wavelet over f(t). The wavelet spectrum is thus customarily displayed in timefrequency domain. For low scales i.e. when |a| > 1, the wavelet is stretched and contains mostly low frequencies. For small scales, thus a more detailed view of the signal (known also as a “higher resolution”) whereas for larger scales a more general view of the signal structure can be expected. However, in practical the hydrologic time series does not have a continuous – time signal process but rather a discrete – time signal. The Discrete Wavelet Transform (DWT) is to calculate the wavelet coefficients on discrete dyadic scales and positions in time. Discrete wavelet functions have the form by choosing and in equation (1). The Eq. (1) has takes the form

g

m, n

(t ) 

a

t  n b0 a0

m

1 m

g(

0

a

m

)

o

(2) where m and n are integers that control the wavelet dilation and translation respectively; is a specified fined dilation step greater than 1; and

is the location parameter and must be

greater than zero. The appropriate choices for

and

depend

on the wavelet function. A common choice for them is

=2,

=1. The original signal X(n) passes through two complementary filters (low pass and high pass filters) and emerges as two signals as Approximations (A) and Details (D). The

HYDRO 2014 International

approximations are part of low pass filter, high-scale and low frequency components of the signal. The details are part of high pass filter, low-scale, and high frequency components. Normally, the low frequency content of the signal (approximation, A) is the most important part. It demonstrates the signal identity. The high-frequency component (detail, D) is nuance. The decomposition process can be iterated, with successive approximations being decomposed in turn, so that one signal is broken down into many lower resolution components (Figure 1). Thus, DWT allows one to study different investigating behaviours in different time scales independently (Rajaee et al., 2011). Decomposition level is generally based on signal characteristics and experiences to selection. Mohammad, (2012) used int[lgn] as resolution level number, where n is the length of daily stream flow sequences and lg denotes the logarithm to base 10. The P may be selected from the range of 2 and int[lgn], that is, 2 ≤ P ≤ int[lgn]. Based on this concept, three decomposition levels were used in this study. In this study, wavelet function derived from the family of Daubechies wavelets with order 5 (db5) used for the selection of best architectures of ANN. Based on the physical knowledge of the problem and statistical analysis, different combinations of antecedent values of the inflow, rainfall and stream flow time series were considered as input nodes. The output node is the inflow data to be predicted in one step ahead. The time series data of all variables was standardized for zero mean and unit variation, and then normalized into 0 to 1. The activation function used for the hidden and output layer was logarithmic sigmoidal and pure linear function respectively. For deciding the optimal hidden neurons, a trial and error procedure started with two hidden neurons initially, and the number of hidden neurons was increased up to 10 with a step size of 1 in each trial. 2.1 Method of combining wavelet analysis with ANN The decomposed details (D) and approximation (A) were taken as inputs to neural network structure as shown in Figure 2. To obtain the optimal weights (parameters) of the neural network structure, Levenberg–Marquardt (LM) back-propagation algorithm has been used to train the network. A standard MLP with a logarithmic sigmoidal transfer function for the hidden layer and linear transfer function for the output layer were used in the analysis. The number of hidden nodes was determined by trial and error procedure. The output node will be the original value at one step ahead. 3. STUDY AREA AND DATA In the present study, the daily data of rainfall, stream flow at Khanapur gauging station and reservoir inflow for 11 years (from 1986 to 1996) were used to forecast the inflow in Malaprabha reservoir. The model was calibrated using 7 years of data from 1986 to 1992 and validated by using the remaining 4 years of data from 1993 to 1996.The input vectors to models are selected based on the procedure described by Sudheer et al. (2002). The following data sets identified as input neurons to ANN and WNN model were examined (i) daily inflow (at t 0 and

MANIT Bhopal

Page 119

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

t0-1), daily rainfall (at t0) and daily stream flow at Khanapur gauging station (at t0) [4 input nodes] (ii) daily inflow (at t 0 and t0-1), daily rainfall (at t0) and daily stream flow at Khanapur gauging station (at t0 and t0-1) [5 input nodes] (iii) daily inflow (at t0, t0-1and t0-2), daily rainfall (at t0) and daily stream flow at Khanapur gauging station (at t0 and t0-1) [6 input nodes]. 4. MODEL EVALUATION To find out the optimal model developed in estimating reservoir inflow, different statistical indices are introduced. The indices employed are the coefficient of correlation (R), root-meansquare error (RMSE) between the observed and forecasted values and the coefficient of efficiency (Nash-Sutcliffe) (COE). 5. RESULTS AND DISCUSSION For the above application, the data is divided into training and testing data sets. In this application, the first 7 year daily data (from 1986 to 1992) are used for training and the remaining 4 year (from 1993 to 1996) are used for testing. The standardized observed data was taken as input to ANN. ANN was trained using backpropagation (BP) with LM and Radial basis (RB) neural network algorithms. The optimal number of hidden neurons were determined by trial and error procedure. Table 1 shows the performance of ANN models for different datasets of inputs in calibration and validation periods. The decomposed data of different datasets of inputs was taken as input to ANN which makes the WNN. The number of hidden nodes were determined by trial and error procedure and the performance of these were shown in Table 1. From this table, the best performed architectures of WNN (20-3-1) was selected.

inflow values of Malaprabha reservoir. Daily rainfall, antecedent inflow values and stream flow data at upstream gauging station used in this study. The observed time series are decomposed into sub-series using discrete wavelet transform and then appropriate sub-series is used as inputs to the neural network and regression models for forecasting the reservoir inflow. Model parameters are calibrated using 7 years of data and rest of the data is used for model validation. The results were compared with the standard ANN. From this analysis, it was found that efficiency index is more than 97% for Wavelet based NN and regression models whereas it is 88% and 86% for ANN and regression models respectively. It may be noted that hydrological data used in the WNN model has been decomposed in details and approximation, which may lead to better capturing the rainfall and runoff processes. Table 1. The performance statistics for the calibration and validation period Mo del

D at a se t

AN NBP

i (4 )

WN NBP

i (1 ii 6) (5 ) ii (2 iii 0) (6 ) Iii (2 4)

No. of Hid den neu ron s 3

RMSE (cumecs ) 19.78

Validation

R

COE( %)

0.962

92.60

6 4

18.21 9.21

0.968 0.992

93.73 98.39

3 3

18.86 9.81

0.966 0.991

93.27 98.18

3

8.57

0.993

98.61

R M S E 29 .2 6(c u 18 m .1 29 ec 3.4 9s) 18 .0 29 7.3 2 20 .5 4

CO E( %) 88.2 2

R 0.953 0.955 0.978

95.4 88.0 84

0.951 0.978

95.5 88.1 17

0.974

94.2 Table 2. Statistical moments of the observed and modeled 0 inflow during validation period

Parameter

An analysis to assess the potential of each of the model to preserve the statistical properties of the observed inflow series was carried out for each year of validation period and shown in Table 2. From Table 2, it was revealed that inflow series computed by WNN model with dataset (iii) reproduces the first three statistical moments (i.e. mean, standard deviation and skewness) better than that computed by the other models. The maximum value in the testing period is fairly well estimated by the WMLR method. Table 2 shows that the percentage error in annual peak flow estimates for the validation period for all models and found that the WNN model improves the annual peak flow estimation and the error was limited to 13.4%. It was also observed that the peak flow estimation by wavelet based models is much better (% error is less than 21) than ANN. The error plots for these models in validation period are shown in figure 3. From Figure 3, it is obviously seen that the peaks could be estimated closely by the WNN model. From this analysis, it was worth to mention that the performance of wavelet based WNN models was much better than ANN models in forecasting the reservoir inflow in one-day advance.

The main purpose of the study presented is to examine the applicability and generalization capability of the wavelet based neural networks with back propagation for forecasting the

MANIT Bhopal

Year

Observed

WNN

ANN

1993

35.98

34.82

30.64

1994

60.84

56.54

46.32

1995

24.42

22.84

21.37

1996

24.23

22.96

21.79

1993

78.45

75.40

64.96

1994

126.63

121.17

99.94

1995

58.05

50.49

50.37

1996

51.71

47.65

44.41

1993

4.22

4.73

4.58

1994

3.83

4.28

4.05

1995

5.22

4.81

5.18

1996

3.48

3.35

3.67

1993

669.58

-11.5

1.1

1994

1016.00

-1.1

25.3

1995

567.53

21.5

18.3

Mean

Standard Deviation

skewness

6. SUMMARY

HYDRO 2014 International

Calibration

% Error in Peak

Page 120

International Journal of Engineering Research Issue Special3 1996

382.90

18.7

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

10.4

Figure 3. Distribution of error plots along the magnitude of flow during validation period

REFERENCES: i. Cannas, B., Fanni, A., Sias, G., Tronei, S., Zedda, M.K., 2005. River flow forecasting using neural networks and wavelet analysis. In: EGU 2005, European Geosciences Union, Vienna, Austria, 24–29 April, 2005. ii. Cannas, B., Fanni, A., See, L. & Sias, G. (2006). Data preprocessing for river flow forecasting using neural networks: Wavelet transforms and data partitioning. Physics and Chemistry of the Earth, PartsA/B/C, 31(18): 11641171. iii. Mohammad Nakhaei and Amir Saberi Nasr, (2012). ―A combined Wavelet- Artificial Neural Network model and its application to the prediction of groundwater level fluctuations‖ JGeope 2 (2), 2012, P. 77-91 iv. Okkan, U. (2012) ―Wavelet neural network model for reservoir inflow prediction‖, Scientia Iranica, 19(6), pp.1445-1455. v. Rajaee, T., Nourani, V., Mohammad, Z.K. and Kisi, O. (2011). ―River suspended sediment load prediction: application of ANN and wavelet conjunction model‖, Journal of Hydrologic Engineering, 16(8): 613-627. vi. Sudheer, K.P., Gosain, A.K., Rangan, D.M, Saheb SM. 2002. Modeling evaporation using an artificial neural network algorithm. Hydrological Processes 16: 3189–3202.

Figure 1. Diagram of multiresolution analysis of signal

Figure 2. Wavelet based multilayer perceptron (MLP) neural network

Application of Particle Swarm Optimization in Multiobjective Irrigation Planning D V Morankar1 , K Srinivasa Raju2 , A Vasan3, L Ashoka Vardhan4 1 Faculty of Civl Engineering, College of Military Engineering, CME(PO) Pune 411031 2,3,4 Centre of Excellence in Water Resources Management, Department of Civil Engineering Birla Institute of Technology and Sciences, Pilani Hyderabad Campus, Hyderabad-500078 Email: [email protected] ABSTRACT: Particle Swarm Optimization (PSO) is applied to the case study of Khadakwasla Complex reservoir system,

HYDRO 2014 International

MANIT Bhopal

Page 121

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Maharashtra, India in multiobjective irrigation planning environment. Three objectives, namely, Annual Net Benefits (ANB), Annual Crop Production (APD) and Annual Labour Employment (ALE) are considered in maximization perspective for 90% dependable inflow level scenario with groundwater. Uncertainty in objectives is tackled through nonlinear membership functions which are also used as the basis to formulate the problem in multiobjective environment. It is observed from result analysis that ANB, APD, ALE in multiobjective environment respectively are `1458.12 Million, 1.30 Million tons, 4.74 Million man-days with degree of satisfaction 0.26.Various combinations of PSO parameters such as randomness amplitude of roaming particles (  ), speed of convergence (  ), randomness control parameter (  ), inertia (θ), penalty value and population of particles were tried and the optimal set of  ,  , θ respectively are arrived at 0.10, 1.17, 0.28. Sensitivity analysis is performed to study the influence of population size, number of iterations, penalty value on ANB, APD, ALE, degree of satisfaction, α, ω, θ and CPU Run Time (CPURT). It is observed that CPURT increases with increase in population, number of iterations, while it is almost constant with increase in penalty. ANB shows no appreciable change with increase in population, with increase in number of iterations however, it decreases with increase in penalty. Keywords: PSO, Optimization, reservoir system, irrigation planning, membership function. 1. INTRODUCTION Irrigation planning is becoming complex due to increase in irrigation, municipal and industrial demands and dwindling supplies. The problem becomes aggravated in multiobjective situations where more than one objective is to be satisfied simultaneously. An optimization approach is thus essential to achieve efficient cropping pattern, reservoir operating policies in the multiobjective framework. On the other hand, Particle Swarm Optimization (PSO) is gaining familiarity in multiobjective environment due to its flexibility and handling practical problems (Morankar, 2014). Numerous authors studied irrigation planning in multiobjective environment. Some of the studies are as follows: Raju and Nagesh Kumar (2000) analyzed the irrigation planning problem in multiobjective framework with net benefits, agricultural production and labour employment as objectives for the case study of Sri Ram Sagar Project, India. Objectives were considered as fuzzy in nature. Sahoo et al. (2006) developed linear programming and fuzzy optimization models for planning and management of available land-water-crop system of Mahanadi-Kathajodi delta in eastern India. The models were used to optimize the economic return, production and labour utilization, and to arrive at the related cropping pattern. Consoli et al. (2008) proposed minimization of reservoir release deficit to meet the irrigation demands and the maximization of net benefits from Pozzillo reservoir, Eastern Sicily. They used nonlinear programming, constraint method and interactive

HYDRO 2014 International

analytical step method to find the best compromise solution. It was concluded that the interactive approach allows improving the performance of the reservoir. Deep et al. (2009) developed fuzzy interactive method for efficient management of multipurpose multireservoir problems and applied to a realistic multipurpose multireservoir. Two objectives, namely, irrigation and hydropower generation were considered in fuzzy environment. These objectives were combined into a single objective using the product operator and nonlinear optimization was adopted using Genetic Algorithm. It was concluded that the interactive approach was found to be satisfactory. Yang and Yang (2010) applied an interactive fuzzy satisfying method to solve multiobjective optimization problem for the case study of Yellow River Delta, China. Mirajkar and Patel (2013) applied multiobjective fuzzy linear programming approach to a case study of Ukai irrigation project Gujarat, India. Four objectives were considered. The model was solved for four situations of 90%, 85%, and 75% and 60% exceedance probability. It was concluded that probable inflow corresponding to 75% exceedance probability was marginally sufficient to meet the requirements of the study area. No efforts have been made till now to explore Particle Swarm Optimization in multiobjective fuzzy irrigation planning environment for a real world environment. Keeping this in view, present study adopts nonlinear membership function in PSO environment to deal with uncertainty aspects in objective functions. The main outcome from solution methodology is reservoir operating policy, cropping pattern, ANB, APD, ALE in compromise solution and degree of satisfaction. Following sections/subsections describes particle swarm optimization, mathematical modeling followed by results and discussion which includes sensitivity analysis. 1.1 Particle Swarm Membership Function

Optimization

and

Nonlinear

Particle Swarm Optimization (PSO) is a metaheuristic computational procedure (Kennedy and Eberhart, 1995; Morankar, 2014) which simulates the locomotion of swarm based organisms. PSO iteratively tries to improve a solution by moving the potential solutions called particles, through the solution space by directing them towards the present iterations optima and global optima throughout all iterations. Here, the particles keep track of their past coordinates thus keeping track of the swarms best solution (fitness) achieved so far and use this for altering direction and speed in the next iteration. The swarm‟s best position in the entire search domain is assumed to be gbest and last generations best position is pbest. In every iteration, each particle location is altered based on its current position (x), velocity (v), distance between itself and pbest, and the distance between itself and gbest which can be summarized by the following equation:

vijk 1   vijk  n1 ( pijk  xijk )  n2 ( gijk  xijk )   n3

MANIT Bhopal

(1)

Page 122

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

xijk 1  xijk  vijk 1 (2) Where, i = number of particles, j = number of decision variables, k = iteration count, g = gbest particle, p= pbest particle, θ = inertia, n q where q=1, 2, 3 is Gaussian distributed random

complex project has three storage reservoirs, Panshet, Warasgaon, and Temghar with the gross storage of 871Mm3, New Mutha Right Bank Canal (NMRBC) serving 62146 ha command area (of length 202 km), Janai-Sirsai Lift Irrigation Scheme (JSLIS) (14080 ha command area), Purandar Lift Irrigation Scheme (PLIS) (25100 ha command area) (refer Fig 2.)

variable ranged between 0 and 1,  = randomness amplitude of roaming particles,  = speed of convergence,  = randomness control parameter. The inertia factor is used for refining the swarm‟s behavior towards the magnitude of the search domain. The velocity of the particles is directly proportional to inertia; larger values of inertia increases the search domain while smaller values of inertia narrow the scope of search. The velocity of all the particles significantly reduces as the iteration count increases resulting in initial rapid search for optima in the beginning followed by a convergence towards the end. Number of iterations was specified as termination criteria. Nonlinear membership function for any objective function Z can be expressed as (Fig 1):

0    Z  Z L   Z  X      Z U  Z L  1 

for

Z  ZL

for Z L  Z  Z U for

Z  ZU

(3) Where β provides the basis for desired shape of membership function (β=1 for linear; β >1 and β 0.5 i.e. the flow variables are weighted towards the j+1 time level. An unconditional stability means that there no restriction on the size of ∆x and ∆t for stability or in general the scheme is stable for 0.55 <  ≤ 1, this scheme can be made totally implicit by taking  = 1 and explicit by taking  = 0.

array X 

i

of

nodal

variables

is of size of 2 x 4 and Y  is i

of size 2 x 1. On combining the equation for subsequent reaches the complete equation of the river is obtained as follows.  X   Y   0 …… (12)   is the vector having 2N nodal variables for that particular river and the matrix [ X] of size 2n x 2n and [Y] of size 2n x 1. [X] And [Y] being non-linear function of   so the equation (12) is solved by Newton Raphson method discussed later.



2.4 Boundary conditions: 2.6 Unsteady state formulation of Saint Venant equation:

Q

o

  Qi  ql L

…… (8)

Q

From this equation we can take o discharge at downstream as known value and depth at upstream is also known value or input value. 2.5 Steady state formulation of Saint Venant equation: For solving Saint Venant equation for unsteady state condition we require an initial steady state solution corresponding to the initial condition. The steady state equations are derived using equation (2) & (3), after neglecting time derivatives. Q( i 1)  Q( i ) x



q l ( i )  q l ( i 1) 2

…… (9)

{[  (Qi 1 ) 2 / Ai 1 ]  [  (Qi ) 2 / Ai ]} x ( Ai 1  Ai ) ( y i 1  y i )  g 2 x

g

( Ai 1  Ai ) ( S f ) i 1  ( S f ) i 2 2

g

( Ai 1  Ai ) ( S 0 ) i 1  ( S 0 ) i 2 2

HYDRO 2014 International

The finite difference form of unsteady state Saint Venant equations is derived from equations 2 & 3 using four point weighted finite difference Preismenn scheme.

( Ai j 11  Ai j 1 )  ( Ai j 1  Ai j )  2t  (Qi j 11  Qi j 1 )  (1   )(Qi j 1  Qi j ) qli  qli 1  x 2 … (13) (Qi j 11  Qi j 1 )  (Qi j 1  Qi j )  2t {[  (Qi j 11 ) 2 / Ai j 11 ]  [  (Qi j 1 ) 2 / Ai j 1 ]}  x {[  (Qi j 1 ) 2 / Ai j 1 ]  [  (Qi j ) 2 / Ai j ]}  (1   ) x ( Ai j 11  Ai j 1 ) ( y ij11  y ij 1 )  g 2 x ( Ai j 1  Ai j ) ( y ij1  y ij ) 2 x j 1 j 1 j 1 j 1 ( S ) ( Ai 1  Ai ) f i 1  ( S f ) i

 g (1   )

 g

MANIT Bhopal

2

2

Page 162

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Where T= temperature in degree Celsius, θ= 1.024 for oxygen reaeration, θ=1.047 for BOD decomposition, θ=1.08 for sediment oxygen demand (SOD) 2.8 Finite difference formulation:

j j ( Ai j 1  Ai j ) ( S f ) i 1  ( S f ) i  2 2 j 1 j 1 j 1  Ai ) ( S 0 ) i 1  ( S 0 ) i 2 2

 g (1   )

g

( Ai j 11

( Ai j 1  Ai j ) ( S 0 ) ij1  ( S 0 ) ij ( ql v x ) i  ( ql v x ) i 1  0 2 2 2

 g (1   )

Finite Difference formulation of dissolved equation for steady state and unsteady state condition is derived from equation () by four point finite difference scheme as: Steady state:

…… (14) (S f ) i 

n 2 Qi Qi Pi Ai

Where

( 4 / 3)

(10 / 3)

2.7 Newton Raphson method: The computational procedure at any time starts form assigning the trial values to the 2p unknowns at that time. The trial values may be the values known from initial conditions or from calculated values from the previous time steps in case of unsteady flow problems. Using this trial values we determine the residuals or corrections



(1) i, j



(0) i, j

 

i , j

such that

Unsteady state:

i, j

…… (15) Where

 (1)i, j 

is the better estimate for the flow depth at section

(1)

i, j (i,j) and (j=1,2…p/2) are the initial estimates for the variables (depth and discharge) , the subscript in the parentheses indicates number of iterations. The solution is obtained by finding values for the unknowns y and Q such that the residuals are forced to approach very close to aero or less than prescribed values. Following is the algorithm of Newton-Raphson method.

 J  

 F 

…… (16)

f ( )  0 the Jacobian matrix [J] and the Denoting eq (7) as column vector [F] is formed as

f ( )

…… (17)

…… (18) 3.8 Mass Balance Equation for Dissolved Oxygen: Mass balances for dissolved oxygen in natural river can be written as: C C  2C  v  DL 2  (  K d ) L  (  K a )(C s  C )  Pa  R  S B'  C D t x x

…… (19) where C = Concentration of dissolved oxygen (mg/l), v = Velocity of flow (m/day), DL= Longitudinal dispersion (square meter/day), Kd = Deoxygenating rate(per day), Ka = Aeration rate (per day), Cs = Concentration of saturated DO (mg/l), Pa = Average gross photosynthetic production of DO (mg DO/l.day), R = Respiration by plants (mg DO/l.day), SB‟ = (SB / y) (mg/l.day), SB = Sediment oxygen demand (g/square metre.day) Temperature effect on reaction kinetics:

K (T )  K (20 ) (T  20)

HYDRO 2014 International

(C i j 1  C i j11  C i j  C i j1 ) (vij11  vij 1 ) (C i j11  C i j 1 )  2t 2 x j 1 j 1 j 1 j 1 j 1 j 1 ( D  DL i 1  DL i  2 ) (C i  2  2C i 1  C i ) ((  K d ) L) ij 1  ((  K d ) L) ij11  Li  3 2 x 2 j 1 j 1 j 1 j 1 ((  K a )C s ) i 1  ((  K a )C s ) i ((  K a ) i  (  K a ) i 1 ) (C i j11  C i j 1 )   2 2 2 j 1 j 1 j 1 j 1 j 1 j 1 P  Pa i 1 Ri  Ri 1 S ' B i  S ' B i 1  ai   0 2 2 2 …… (21)

2.9 Mass Balance equation for BOD Mass balances for BOD in natural rivers can be written as: L L 2L  v  DL  (  K r ) L  C D t x x 2 …… (22)

 f    y 

J    F  

(vi 1  vi ) (Ci 1  Ci ) ( DL (i )  DL (i 1)  DL (i  2) ) (Ci 2  2Ci 1  Ci )  2 x 3 x 2 ((  K d ) L) i  ((  K d ) L) i 1 ((  K a )C s ) i 1  ((  K a )C s ) i   2 2 (  K a ( ) i  (  K a ) i 1 (Ci 1  Ci ) Pa (i )  Pa (i 1) Ri  Ri 1 S B (i ) ' S ' B (i 1)     0 2 2 2 2 2 …... (20)

where L = Concentration of BOD (mg/l) Kr = Ks + Kd Ks = effective loss rate due to settling (per day) Finite Difference formulation of BOD for steady state and unsteady state condition is derived from equation () by four point finite difference scheme as: Steady state: (vi 1  vi ) ( Li 1  Li ) ( DL (i )  DL ( i 1)  DL (i  2 ) ) ( Li  2  2 Li 1  Li )  2 x 3 x 2 ( K L)  ( K r L) i 1 (L) i  (L) i 1 C Di  C Di1  r i   0 2 2 2 …… (23)

Unsteady state:

MANIT Bhopal

Page 163

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

( Lij 1  Lij11  Lij  Lij1 ) (vij11  vij 1 ) ( Lij11  Lij 1 )  2t 2 x j 1 j 1 j 1 j 1 j 1 j 1 ( D  DL i 1  DL i  2 ) ( Li  2  2 Li 1  Li ) ( K r L) ij 1  ( K r L) ij11  Li  3 2 x 2 (L) i  (L) i 1 C Di  C Di1   0 2 2

BOD VS DISTANCE 18 16 14

BOD (mg/l)

12 10

Numerical solution

8

Analytical solution

6 4

…… (24) 3. RESULTS AND ANALYSIS:

2 0 0

20000

DO (mg/l) Discharge m3/d

7.5 500000

KP 100-80 20.59 8.987 1.842 0.764 0.514 0.0002 2 10 200 at 100 KP 2 at 100 KP 540000

KP 80-60 20.59 8.987 1.842 0.514 0.514 0.0002 2 10 5 at 60 KP

120000

DO Variation along a channel 9 8 7 6 5

DO from analytical soln

4

DO from numerical soln

3 2

9 at 60 KP 540000

1

640000

0 0

20000

40000

60000

80000

100000

120000

Distance (m)

N = 10 B = 10m L = 100 Km Δx = 10000 m Δt = 0.1 day n = 0.035

h

1 = 1.24m Depth on upstream side Channel longitudinal bottom slope So = 0.0002 For second point source

h

1 = 1.42m Depth on upstream side Channel longitudinal bottom slope So = 0.00018

HYDRO 2014 International

100000

Figure 2. Distance Vs BOD

KP 100 20 9.092 1.902 0.5 0.5 0.0002 2 10 2

60000 Distance (m)

For verification of the results hypothetical problem solved by finite difference method and determined the depth and discharge at various nodes by using the Saint Venant equation. The problem solved by numerical method and compares it by analytical solution. We have no data for unsteady state problem, so compared unsteady state solution to the steady state solution. Problem: A river receives a sewage treatment plant effluent at kilometer point (KP 100) and a tributary inflow at KP 60. The channel is trapezoidal. The deoxygenation rate for BOD is equal to 0.5 per day at 20 degree Celsius. For 20 KM downstream from the treatment plant, there is a BOD settling removal rate of 0.25 per day. Parameter T (0C) DO Sat.(mg/l) Ka (per day) Kr (per day) Kd (per day) Channel slope Side slope Bottom width BOD (mg/l)

40000

Figure 3. DO Vs Distance

Saint Venant Equation, mass transport equation for BOD and DO solved by finite difference method and compares with analytical solution. The results are found closed to analytical solution. The dispersion term is in mass transport equation whereas in analytical solution dispersion term is not included. Thus results obtained by this methodology for BOD and DO mass transport equation along a channel could be used at field with boundary conditions. 4. CONCLUSIONS: The mass transport equation for DO and BOD solved by finite difference implicit scheme and the Saint Venant Equation for depth and discharge at various nodes is also solved by finite difference scheme which used in solving of mass transport equation. On the basis of results it is observed that there is no problem of Courant condition and it gives very good results. Dispersion is not much significant for steady state problems but it has significance for unsteady state condition. Thus we prepare

MANIT Bhopal

Page 164

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

a mathematical model for estimating the different water quality measures along a river at different cross sections by using MATLAB for simulation of these measures. By estimating these measures by using this model we can control the water quality at different cross section along a river as required.

REFERENCES: i. Brown LC & Barnwell TO (1987) The enhanced stream water quality models QUAL2E and QUAL2E-UNCAS: Documentation and User Manual, Report PA/600/3-87/007, U.S. EPA, Athens, GA, USA. ii. Cox BA (2003) A review of currently available in-stream waterquality models and their applicability for simulating dissolved oxygen in lowland rivers. Science of the Total Environment 314: 335-377. iii. Downer CW and Ogden FL, 2004, GSSHA: A model for simulating diverse stream flow generating processes, J. Hydrol. Engrg., 9(3):161-174. iv. Effler SW, Brooks CM, Whitehead K., Wagner B., Doerr SM, Perkins M, Siegfried CA, Walrath L & Canale RP (1996) Impact of zebra mussel invasion on river water quality. Water Environment Research 68(2): 205-214. v. Giri, BS, Karimi IA & Ray MB (2001) Modeling and Monte Carlo simulation of TCDD transport in a river. Water Research 35(5): 1263-1279. vi. Guitjens JC, Ayars JE, Grismer ME & Willardson LS (1997) Drainage design for water quality management: overview. Journal of Irrigation and Drainage Engineering – ASCE 123(3): 148-153. vii. Horn AL, Rueda FJ, Hormann G & Fohrer N (2004) Implementing river water quality modelling issues in mesoscale watershed models for water policy demands – an overview on current concepts, deficits and future tasks. Physics and Chemistry of the Earth 29(11-12): 725- 737. viii. Lindenschmidt KE, Rauberg J & Hesser F (2005) Extending uncertainty analysis of a hydrodynamic – water quality modeling system using High Level Architecture (HLA). Water Quality Research Journal of Canada 40(1): 59-70. ix. Mujumdar PP (2002) Mathematical tools for irrigation water management – an overview. Water International 27(1): 47-57. x. Supriyasilp T, Graettinger AJ & Durrans SR (2003) Quantitatively directed sampling for main channel and hyporheic zone water-quality modelling. Advances in Water Resources 26: 1029-1037.

Assessment of groundwater quality of bah block, agra, india. Azmatullah Noor1Dr. Izharul Haq Farooqi2 Assistant Professor, Vivekananda College of Technology and Management, Mathura Bye pass, Near Khair road, Aligarh202002, U.P., India. 2 Associate Professor, ZakirHussain College of Engg. & Tech., A.M.U, Aligarh-202002, U.P., India. Email:[email protected], 1

ABSTRACT:The study was conducted in the month of May and June 2012, to evaluate the water quality in the rural areas of Agra. A total of 60 groundwater samples from 28 locations which comprises of villages of Bah block. The samples were collected from tube wells, bore wells, and hand pumps with recording the position of sampling point, by Global Positioning System (GPS) device. The samples were examined for physicochemical parameters of water such as pH, alkalinity, total hardness, electrical conductivity, turbidity, iron, fluoride, chloride, nitrate, total dissolved solid, and dissolved oxygen. The main objective of the study was to get information on the distribution of water quality on a regional scale as well as to create a background data bank of different chemical

HYDRO 2014 International

constituents and their quantities in ground water. All data were statistically analyzed by SPSS package for mean, median, mode and standard deviation. The Pearson correlation was also established between physico-chemical parameters of groundwater. The mean value for pH-8.1098, alkalinity-455.70 mg/l, total hardness-439.94 mg/l, electrical conductivity1541.68 (µS/cm), turbidity-5.86 NTU, iron-.5166 mg/l, fluoride-1.5672 mg/l, chloride-336.5269 mg/l, nitrate-5.4703 mg/l, total dissolved solid- 737.38 mg/l, and dissolved oxygen4.4342 mg/l. When Pearson correlation was established it was seen that thereare positive correlation of conductivity with dissolve oxygen, fluoride, iron, total hardness, alkalinity,total dissolved solid, and chloride. The correlation of fluoride with iron, total hardness, alkalinity, total dissolved solid, nitrate, and chloride is also positive. The result so obtained reveals that the groundwater is contaminated because of penetration of chemicals from river Yamuna which is passing along Bah block. Keywords: Groundwater, physico-chemical parameters, SPSS, Agra, Bah. 1.

INTRODUCTION

Groundwater is important for human water supply and, in Asia alone, about one billion people are directly dependent upon this resource (Foster SSD., 1995). The groundwater resources play a very significant role in meeting the ever increasing demands of the agriculture, industry and domestic sectors (Saleem R., 2007). India supports more than 16% of the world‟s population with only 4% of the world‟s fresh water resources (Singh AK., 2003). The potable nature of groundwater is mainly based on the physico-chemical characteristics of the water sample. The impact of industrial effluents is also responsible for the deterioration of the physico-chemical and bio-chemical parameters of groundwater.In a reporton "Status of groundwater quality in India part-1"by (Center Pollution Control Board, 2006-2007) it is mentioned thatin Agra there are 73 industries and 2 industrial clusters, which discharges their effluent into the river. Of these industries, only 64 industries have effluent treatment plants.Other industries which discharge their effluent directly into the river, playsvital role in groundwater contamination.The wide range of contamination sources is one of the many factors contributing to the complexity of groundwater assessment. It is important to know the geochemistry of the chemical-soil-groundwater interactions in order to assess the fate and impact of pollutant discharged on to the ground. Pollutants move through several different hydrologic zones as they migrate through the soil to the water table. The serious implications of this problem necessitate an integrated approach in explicit terms to undertake ground water pollution monitoring and abatement programs. The intensive use of natural resources and the large production of wastes in modern society often pose a threat to ground water quality and have already resulted in many incidents of ground water contamination. Pollutants are being added to the ground water system through human activities and natural processes. Solid waste from industrial units is being dumped near the factories, which is subjected to reaction with percolating rain

MANIT Bhopal

Page 165

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

water and reaches the ground water level. The percolating water picks up a large amount of dissolved constituents and reaches the aquifer system and contaminates the ground water. The problem of ground water pollution in several parts of the country has become so acute that unless urgent steps for detailed identification and abatement are taken, extensive groundwater resources may be damaged.

Table 1. List of panchayats of study area S.No. 1. 2. 3. 4. 5. 6.

1.1 Objective and scope of study:

Panchayat Derak Kenjra Dodapura Badous Veri Bitholi

Station I II III IV V VI

The main objective of present study was to carry out ground water quality monitoring of 60 groundwater samples from 28 locations which comprises villages of Bah block inAgra and to get information on the distribution of water quality on a regional scale as well as to create a background data bank of different chemical constituents and their quantities in ground water. One of the main objectives of the ground water quality monitoring was to assess the suitability of ground water for drinking purposes. The physical and chemical quality of ground water is important in deciding its suitability for drinking purposes. 2.

MATERIAL AND METHOD

2.1 Collection of sample: To study the physical and chemical quality of ground water of the area for deciding its suitability for drinking purposes. A survey of villages of Bah was conducted in the month of May and June, 2012 by collecting 60 samples of groundwater from 28 villages. The groundwater samples were collected by grab sampling after flushing hand pumps for 5 to 10 minutes. The samples were collected in 1litre plastic bottle. Groundwater samples were immediately transferred to the laboratory and were stored at 4˚C to avoid any major chemical alteration. 2.2 Study area: Agra district occupies the southwestern part of the state of Uttar Pradesh (India) and is bounded by the state of Rajasthan in the west and the state of Madhya Pradesh in the south. Bah Tehsil is the easternmost part of Agra district and belongs to both the marginal and central alluvial plain (Ganga Plain). The Bah Tehsil area is situated between 26˚45' and 27˚ 0'N latitudes and between 78˚10' and 78˚50'E longitudes at approximately 178 m above sea level. The study area has a semi-arid to arid climate with an average monthly temperature varying between 38˚C and 46˚C in the summer and between 25˚C and 32˚C in the winter. The average weather conditions allow recognizing six well marked traditional seasons, i.e. spring (March–April), summer (May–June), monsoon (July–August), sharada (September– October), hemanta (November–December) and winter (January– February). The average annual rainfall variation is between 600 and 650 mm(Misra, A. K. et al. 2007). In present study, samples of groundwater were taken from six panchayats, which are mentioned in Table 1. From each station 10 samples were collected. The coordinate position of sampling point is located by GPS device, which is further plotted by ArcGIS 10 on the map of study area as shown in Figure 1.

HYDRO 2014 International

Figure 1.Map of study area with sampling points. 2.3 Analytical methodology: The groundwater samples were analyzed for total hardness, total alkalinity, chloride using(APHA,1995) procedure, and suggested precautions were taken to avoid contamination. The electrical conductivity, pH, dissolve oxygen, total dissolve solids were determined by LDO probe (HACH) and turbidity by Digital Nephlometer. The fluoride, iron, nitrate were determined by Spectrophotometry (DR 5000- HACH). 2.4 Statistical analysis: The observed data of physico-chemical parameters were analyzed by SPSS 19.0 software to measure its central tendency, and deviation of the values from its mean i.e., standard deviation. The Pearson correlation was also established among them to identify their relation with each other. 3. RESULTS AND ANALYSIS The result obtained after the analysis of physico-chemical characteristics of groundwater sampleare tabulated below from Table 2a to Table 2f. The number of samples which are exceeding IS: 10500 (2003) for physico-chemical parameters are mentioned in Table 3. Station I The fluoride concentration is exceeding the limit as per IS: 10500 (2003) in six samples of groundwater out of ten samples. Total alkalinity and turbidity is in excess in three samples. The

MANIT Bhopal

Page 166

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

EC is exceeding the limit in all samples of the station. In this station the correlation between TH and TDS is significant which indicates the concentration of Ca2+ and Mg2+ salts in groundwater. TDS is also correlated with Fe and TA. The correlation between TA and TH is in significance which indicates the presence of carbonate and bicarbonate salts of Ca 2+ and Mg2+. Significant correlation of turbidity with pH and F, TH with Fe, and pH with D.O. has also been noticed. Station II In this station eight samples are having higher concentration of fluoride. The groundwater is brackish in taste which is indicated by the range of EC lying between 1000 to 1500 micro mhos/cm.The correlation of D.O. with turbidity and Cl is significant.

Ganga Plain. The main characteristics of soil horizons of the area are the high content of carbonate, distributed throughout the depth of the profile. In addition, the study area shows frequent alternations of mud and clay layers in the subsurface lithology and has very low hydraulic conductivity (Misra 2005). These factors together constitute a favourable condition for the maximum absorption of Na+, K+, and by the clay minerals in the soil of shallow and intermediate aquifers.Generally, Na+, K+, and are added to the soil from several anthropogenic sources both directlythrough phosphate fertilizers, and indirectly, through atmospheric pollution from industries and burning of fossil fuels (Drury et al. 1980). Table 2a.Physico-chemical characteristics of groundwater of Station I.

Station III In this station TA is correlated with pH which indicates that groundwater is alkaline in nature. The correlation of nitrate with Fe, TDS and EC is the indication of presence of nitrate salts of iron. Consequently, the correlation between EC and TDS has been noticed. Station IV The correlation of nitrate with TDS and EC has been noticed in this station and EC with TDS which is the sign of presence of nitrate salts.

Table 2b.Physico-chemical characteristics of groundwater of Station II.

Station V About 80% of the sample is contaminated with fluoride in this station. The salts of chloride are present in the groundwater of this station which is indicated by the correlation of chloride with EC and TDS. Trace of nitrate salts of iron is present in this station. Station VI There is noteworthy relation of F with Cl, D.O. and EC. Turbidity is caused mainly due to nitrate concentration in groundwater at this station. The correlation matrix also shows the relation of TDS with Cl, turbidity, pH, and EC.

Table 2c.Physico-chemical characteristics of groundwater of Station III.

The maximum and minimum values, standard deviation and central tendencies are tabulated in Table 4a to 4f. Karl Pearson correlation was established among all the parameters, it was observed that TDS and EC are having positive correlation coefficient, except in Station II. The correlations of all parameters have been given in Table 5a to 5f. The groundwater of the study area are characterized by a high concentration of Na+, K+, , and TDS in shallow and intermediate aquifers due to some factors which is postulated that salt-rich geological formations have contributed to these alluvial deposits (Kumar et al. 1993, 1995; Kumar 1998) of the

HYDRO 2014 International

Table 2d.Physico-chemical characteristics of groundwater of Station IV.

MANIT Bhopal

Page 167

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Table 2e.Physico-chemical characteristics of groundwater of Station V. Table 4a.Statistical data of physico-chemical parameters of Station I.

Table 2f.Physico-chemical characteristics of groundwater of Station VI.

Table 4b.Statistical data of physico-chemical parameters of Station II.

F-Fluoride, Fe-Iron, N- Nitrate, TH-Total hardness, TA-Total alkalinity, TDS-Total dissolve solid, Cl-Chloride, D.O.-Dissolve oxygen, EC-Electrical conductivity. Table 3.Number of samples exceeding IS: 10500 (2003) limit in all stations.

HYDRO 2014 International

MANIT Bhopal

Page 168

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Table 4c.Statistical data of physico-chemical parameters of Station III.

Table 5a. Pearson correlationmatrix for physico-chemical parameters of Station I.

Table 4d.Statistical data of physico-chemical parameters of Station IV.

Table 5b. Pearson correlationmatrixfor physico-chemical parameters of Station II.

Table 4e.Statistical data of physico-chemical parameters of Station V. Table 5c. Pearson correlationmatrix for physico-chemical parameters of Station III.

Table 4f.Statistical data of physico-chemical parameters of Station VI.

HYDRO 2014 International

MANIT Bhopal

Page 169

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Table 5e. Pearson correlationmatrix for physico-chemical parameters of Station V.

Table 5f. Pearson correlationmatrix for physico-chemical parameters of Station VI.

4. CONCLUSION

ii. Bhargava GP, Abrol IP, Kapoor BS, Goswami SC (1981) Characteristics and genesis of some sodic soils in the Indo- Gangetic alluvial plains of Haryana and Uttar Pradesh. J Indian Soc Soil Sci 29(1):61–70 iii. BIS Bureau of Indian Standards Drinking water-specification (2003) IS:10500, New Delhi iv. Central pollution control board, Report on Status of groundwater quality in India part-1. Groundwater quality series:Gwqs/ 09/2006-2007 v. Drury JS, Ensminger JT, Hammonds AS, Hollem JW, Lewis EB, Elemental and mineralogical composition of the coarse Environmental effects of Pollutants, IX Flouride. US Environmental Protection Agency, Cincinnati, 549 p vi. Foster SSD., 1995 Groundwater quality, 17th Special Report. Chapman and Hall, London vii. Kruawal, K., Sacher, F., Werner, A., Mu¨ller, J., &Knepper, T.P. (2005). Chemical water quality in Thailand and its impacts on the drinking water production in Thailand. The Science of the Total Environment, 340, 57– 70. doi: 10.1016/j.scitotenv.2004.08.008. viii. Kumar R (1998) Role of Himalayan Orogeny in the formation of salt affected soils of the Indian sub-continent. In: Proceedings of 16th World Congress of Soil Science, held at Montpellier, August 20–26, 1998. Symposium: 15 Reg No: 277 ix. Kumar R, Ghabru SK, Ahuja RL, Singh NT, Jassal HS (1993) Clay minerals in the alkali soils of Ghaggar river basin of Satluj–Yamuna divide in North-West. Clay Res 12:43–51 x. Kumar R, Ghabru SK, Ahuja RL, Singh NT, Jassal HS (1995) Elemental and mineralogical composition of the coarse fraction of the normal and alkali soils of the Satluj–Yamuna divide of North-West India. Clay Res 14:29–48 xi. Misra AK (2005) Integrated water resource management and planning for its sustainable development, using remote sensing and GIS techniques in dark areas of Agra and Mathura districts of Uttar Pradesh. Dissertation. University of Lucknow xii. Misra, A. K. and Mishra, A., (2007). Escalation of salinity levels in the quaternary aquifers of the Ganga alluvial plain, India. Environ. Earth Sci. Journal. 53(1), 47. xiii. Mor, S., Ravindra, K., &Bishnoi, N. R. (2007). Adsorption of chromium from aqueous solution by activated alumina and activated charcoal. Bio resource Technology, 98, 954–957. xiv. Ravindra, K., &Garg, V. K. (2007). Hydro-chemical survey of groundwater of Hisar city and assessment of defluoridation methods used in India. Environmental Monitoring and Assessment, 132, 33–43. doi: 10.1007/s10661-006-9500-6. xv. Robins, N. S. (2002). Groundwater quality in Scotland: Major ion chemistry of the key groundwater bodies. The Science of the Total Environment, 294, 41–56. Doi: 10.1016/S0048-9697(02)00051-7. xvi. Saleem R., (2007) Groundwater management—emerging challenges. Water Digest xvii. Singh AK., (2003) In: National symposium on emerging trends in agricultural physics, 22–24 April 2003. Indian Society of Agro physics, New Delhi. xviii. World Health Organization (WHO). (2006). Guidelines for Drinkingwater Quality. Third Edition. 1st Addendum to Vol. 1. WHO Press, 20 Avenue Appia, 1211 Geneva 27, Switzerland. (http://www.who.int/water_sanitation_health/dwq/gdwq0506.pdf).

The quality of the groundwater of the study area is critical due to , and TDS contamination from; dissolve salts in rainwater, the canal network, low precipitation and high evaporation due to arid climatic conditions. Among all the station, the groundwater of station III is slightly potable and rest of samples are having higher concentration of fluoride which can cause skeletal fluorosis to the human life of that area. The electrical conductivity of almost all samples is having higher values which indicate the level of salinity in groundwater. REFERENCES: i. APHA. (1995). Standard methods for the examination of water and wastewater (19th ed., pp. 1–467). Washington, DC: American Public Health Association.

HYDRO 2014 International

MANIT Bhopal

Page 170

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

CHANGING WATER QUALITY SCENARIOS OF TANK CASCADE SYSTEM AND ITS IMPLICATIONS J.HEMAMALINI 1, B.V.MUDGAL2 , J.D.SOPHIA 3 1

3

Research Scholar, Centre for Water Resources, Anna University, Chennai 600025. 2 Professor, Centre for Water Resources, Anna University, Chennai 600025. Principal Scientist, M S Swaminathan Research Foundation, Chennai 600113. Correspondence to: [email protected] ABSTRACT

The changing scenarios of tank cascade reveal that the livelihoods of rural community and tank ecosystem are under severe threat which needs immediate attention. A cascade constituting four non-system tanks viz. Athimanjeri, Konasamudram, Podatturpet, Pandravedu located in Pallipet Taluk of Thiruvalore district, Tamil Nadu is chosen as study area. Water samples drawn from four tanks, bore and open wells adjacent to tanks during rainy and summer seasons was tested for its physico-chemical and biological parameters. Water quality index calculated for the tanks to assess its suitability for drinking shows that the status of four tanks is eutrophic and needs proper care and interventions to improve its quality. The irrigation water quality of the four tanks, bore wells andopen wells are assessed using the irrigation water quality indices namely Sodium Absorption Ratio (SAR), Soluble Sodium Percentage (SSP), Magnesium Absorption Ratio (MAR) and Kelly‟s Ratio (KR). The results indicate that in the Pandravedu tank,the change in water quality isdue to discharge of untreated sewage and dyeing unit wastewater. The community perception on changing water quality and its impact was ascertained through qualitative research methods like focused group discussion and one to one interactions which confirms that due to water quality changes in Pandravedu tank there is reduction in paddy yield to about 40%, the water is also not suitable for livestock drinking as it causes diseases, noneof the fish species are consumed since it causes vomiting and diarrhea.

utilization compared to the groundwater system or even the major irrigation projects. (Lenin B 2006).The cascade approach should be followed in restoring tanks if the full benefits of harvesting the runoff from a micro watershed and effective groundwater recharge are to be realized. Another concept that can ensure the sustainability of tanks cascade system is to have ecological andsocio-economic harmonywhere the village society and its economy can evolve and thrive on the judicious utilization of the local resource base.The current study aims atinterlinking ecosystem and the tank cascades with the following objectives 1. To analyze and ascertain the suitability of surface and ground water quality for drinking and irrigation. 2. To conduct an in depth quality analysis of water used for multiple purposes in Pandravedu village. 3. To elicit community perceptions on the implications of changing water qualityand coping strategies. 2. MATERIALS AND METHOD Description of study area The study area is located in state of Tamil Nadu, India and is a part of Nagari watershed. It spreads out in Pallipattu block of Thiruvallur district. The study area comprises four non-system tanks in a cascade namely Athimanjeri, Konasamudram, Podatturpet and Pandravedu. There are nine villages benefiting from this tank cascade. In addition to agriculture most of the villagers largely depend on non-farm activity like weaving and dyeing. The area is generally hilly and sloppy with hard rock formations overlain by top sandy soil. Figure-1 shows the index map of the study area.

KEYWORDS:Tank Cascade, Physico-chemical parameters, water quality index, irrigation water quality indices, community perception, focused group discussion. 1. INTRODUCTION Tanks have been the main source of irrigation in many parts of India for centuries. Conserving the tank eco-systems for multiple uses such as irrigation, domestic, livestock use and groundwater recharge is a way to provide a safety net to protect the livelihood of millions in a semi-arid India (Sakthivadivel 2004). Tanks are eco-friendly and proper management ensure protection and preservation of the micro ecosystem which in turn provides services like recycling of nutrients, purification of water, recharge of groundwater and habitat provision for a wide variety of flora and fauna in addition to aesthetic values. Further, it serves as flood moderators during heavy rains and serves as water points during drought conditions. Tank irrigation was superior in distributing water, economical in terms of energy

HYDRO 2014 International

Figure-1.Index map of study area Water quality analysis Water samples collected from four tanks, nearby irrigation wells both bore and open were tested for its physico chemical and

MANIT Bhopal

Page 171

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

bacteriological parameters for two seasons namely rainy and summer. During water sample collection it was observed that the Pandravedu tank receives untreated wastewater generated from the dyeing processes along with untreated domestic wastewater from the Podatturpet households through a lined channel. The community expressed that in the recent years the water quality of the Pandravedu tank has deteriorated which in turn affect their livelihoods including environment. Therefore in addition to water quality analysis the community perception on changing water quality and its implication on economic uses of water, ecological functions for healthy environment as well as sociocultural uses was ascertained.The samples are coded as given in Table-1: Table-1.Abbreviations of Sampling Stations

= Ideal value for nth parameter in pure water i.e 0 for all parameters and 1.0 for pH Vio

Sn

= Standard permissible value for nth parameter

Water quality index =

Wnqn /

Wn(2)

Where Wn = Unit weight for nth parameter The water quality index obtained for the four tanks Athimanjeri, Konasamudram, Podatturpet and Pandravedu are 178, 1148, 151 and 261 respectively. Comparison of Drinking water quality parameters are expressed in Table-2. Table-2. Seasonal Variation of Drinking water quality parameters

3. RESULTS AND DISCUSSION Water quality index Water quality index is calculated for the four tanks using equations 1 and 2 as it is a useful tool to assess the present drinking water quality status andto compare with the BIS standards (Yogendra et al, 2007).

(1) Where qn = quality rating for nth parameter Vn

= Estimated value for nth parameter

HYDRO 2014 International

Water quality analysis of water samples confirm that at certain locations the values exceeded the permissible limits of drinking standards. The presence of E coli in tank water and in groundwater at certain places indicates that the water is polluted with waste water. The higher values of TDS ranging between 188 mg/l to 1133 mg/l prove that water is unfit for drinking. The total hardness and presence of chlorine is very high in the Pandravedu tank which made unfit for domestic use and cattle drinking. The BOD, COD and DO also exceeded the permissible values at certain locations. Irrigation water quality indices Irrigation water quality of the four tanks, bore wells and open wells are assessed using the indices namely Sodium Absorption Ratio (SAR), Soluble Sodium Percentage (SSP), Magnesium Absorption Ratio (MAR) and Kelly‟s Ratio (KR) (Raihan et al, 2008). Comparison of irrigation water quality indices with the standard values are expressed in Figure-2.

MANIT Bhopal

Page 172

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 The SAR for all the tanks and wells fall within the range of 26 for both the seasons. All the four tanks exceeded the standard permissible value of 40 for SSP whereas the bore wells and open wells are found within the limit. The MAR values for all the locations fall within the standard permissible limit of 50%. The Kelly‟s Ratio is found to be greater than 1 in all the tanks whereas the well samples are within the permissible range. (Ramesh et al, 2010). The Table-3gives the seasonal variation of irrigation water quality indices for rainy and summer seasons. Table-3. Seasonal Variation of Irrigation Water Quality Indices

Qualitative research method Qualitative research methods like group discussion with farmers‟ and one to one interaction with the general public including landless labourers was used to collect community perceptions. A checklist was designed comprising of questions relating to (i) people‟s observation in changing water quality over a period of time (ii) causes for the changes in water quality (iii) its implications on multiple uses like agriculture, livestock, drinking, other domestic uses and biodiversity and (iv) specific issues affecting women due to water quality changes. The qualitative information generated through group discussion and one to one interaction is analyzed and presented in the subsequent section. Community perception on changing water quality During group discussion with farmers they expressed that a decade before the tank water was crystal clear in its physical appearance and tasted very good which was directly used for drinking, cooking, and bathing, washing and feeding animals. However, they could observe gradual deterioration in water quality since the year 2000 and became worst in the last five years. Major factor attributed by the community is that in addition to the fresh water sources the domestic and dyeing wastewater from Podatturpet village is directly discharged into Pandravedu tank through a drainage canal hence it is the worst affected tank in the chain. Cause of the problem Figure-2. Comparison of Irrigation water quality indices

HYDRO 2014 International

They further explained that there are about 150 dyeing units in Podatturpet village. Weaving and dyeing is one of the

MANIT Bhopal

Page 173

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

predominant nonfarm activities in this village.Previously weaving alone was done in Podatturpet and dyeing was done in Kanchipuram which is located in a different taluk.After some time people started dyeing in their own units and the untreated effluent was let into a barren land in Podatturpet itself. It posed lot of health issues therefore a lined channel of 5 km is constructed by the local people to fetch to discharge untreated effluent from the dyeing units to a nearby tank namely Thamaraikullam. From there effluent water goes into a small pond called Thangal which in turn drains into Pandravedu tank. The waste water that runs through the channel is of dark brown colour and has bad odour. Implications of the waste water

Earlier they use to rear fish in the tank water and harvest when the water reduces during summer. Major varieties were Koravai (snake head), Kelluthi (cat fish), Keandai (carp) and Veral (Murrel) but in the last few years they could harvest only tilapia and could not find other species. Tilapia is the only variety which survives in poor quality water. Some of the farmers expressed that the colour of the fish has also changed and if it is consumed it causes vomiting and diarrhea. Secondly due to continuous availability of water, harvesting fish has become an issue therefore fish rearing is almost stopped in Pandravedu tank. Due to contamination of tank water, the culture of fish rearing and consuming is total affected. Drinking and domestic uses

According to the farmers and general public views mixing of wastewater into fresh water tank has various implications on productive uses of water like agriculture, livestock rearing, fishing and other uses like ground water recharge and biodiversity which is presented below: Agriculture

From the farmer‟s and the landless agricultural laborerspoint of viewsthe physical quality of tank water is affected because of the untreated effluent from the dyeing units.The quality of water has deteriorated due to both drainage water from houses and effluent from the dyeing unit is sent together without treatment. The taste of water has changed and people are not using it for drinking. The bore wells in and around the tank is also being impacted of the same problem. Panchayat erected four bore wells around the tank and supplied it for drinking to the village by storing in the overhead tanks and establishing a common distribution system. But during last year the colour and odour of the water pumped into the overhead tank was dark brown and hence the Panchayat decided to change the source point.Accordingly bore hole is erected near Kosasthalaiyaru River and pumped to overhead and supplied for three days per week which is not sufficient and they depend on mineral water for drinking. Even the milk gets spoiled if the tank water is used directly. The people face irritation in their skin if they use the tank water to bath or wash their clothes. So they use other small fresh water ponds called as Thoppaiamman and Vannarakulam for washing clothes. Ground water recharge

There are four irrigation channels irrigating the agricultural lands. Five years back the cropping pattern was paddy, paddy followed by chilli, groundnuts, ragi and other dry crops depending on tank water availability. Generally the tank receives water during monsoon and dries out in summer. In fact even the tank water spread area and tank bund was used by farmers to grow short term vegetable crops during summer. But from last five years due to constant flow of wastewater from upstream village Pandravedu tank had become perennial but with poor water quality.Therefore the farmers have no option of cultivating summer crops except of paddy that too very few specific varieties like ADT 37 which is a fat type of rice.Farmers are using both surface water from tank and ground water through open and bore wells conjunctively as a coping strategy. Farmers expressed that the lands irrigated with tank water alone resulted in stuntedcrops and the soil is also affected. Comparatively the middle and tail end farmers are better as the water quality changes in the natural process through conveyance. Farmers feel that the entire ayacut is being affected due to the polluted tank water and the paddy yield is also reduced from 40 bags/acre to 15 bags/acre. The worst implication is the rice grown by the farmers is not consumed by them due to the fact that it will cause health problems so they buy rice from outside. But earlier,a portion of the produce was stored for their household consumption. Livestock Similarly livestock which is the secondary source of living for farmers and landless community, tank water was the main source for cattle drinking and cleaning. But the farmers now suspect that the cattle fall sick when it drinks water directly from tank. Also they expressed that milk production is gradually declining but in depth studies to be done to analyze the cause and effect relationship. They also attributed that due to the odour some livestock is not drinking it. So they are very particular and providing only the drinking water supplied by Panchayat. Fishing

HYDRO 2014 International

Unlike any other tanks this tank is also recharging the groundwater and majority of the farmers‟ are using it for various purposes.However, water in open wells is polluted and is not used presently for domestic purposes.Previously ground water was at 200 feet depth and it was good. But now that water is also polluted and farmers go in for bore for a depth of 300 feet.In addition to groundwater recharging generally tanks also contribute for conservation of biodiversity. Biodiversity Farmers expressed that a decade before this tank maintained a very healthy environment including floral and faunal biodiversity but gradually it is declining due to wastewater. For example there were lot of crabs in the tank as well in paddy field after monsoon especially during November but now they could not find crabs in tank as well as fields. Community used crabs as medicine mainly to treat over cold and breathing problems. Similarly they expressed that some of changes are observed in floral diversity. Another important fact is that earlier when the water quantity reduces during summer they maintain the system, do social forestry and other activities and all these are affected due to continuous flow of wastewater. As a result the entire environment and ecosystem is getting affected.

MANIT Bhopal

Page 174

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 reduce the total disinfectant dose while managing minimum residual chlorine across the system.

4. CONCLUSION The water quality index shows that the water is unfit for drinking and status of four tanks is eutrophic and needs proper care and interventions to improve its quality. The values of irrigation water quality indices prove that the tank water quality has deteriorated and has become unfit for irrigation. Untreated waste waterfrom dyeing units is a major cause for the pollution of Pandravedu tank. Continuous disposal of wastewater without proper treatment makes the tank water unfit for any use. The wastewater before let into the tank should undergo the necessary treatment and the industry should strictly follow the same. Urbanisation of the villages in and around the tank resulted in discharging the sewage directly into the other three tanks. The sewage system in the villages should be well designed and the domestic sewage should be treated properly. REFERENCES: i. Lenin Babu K and Mansi S, Estimation services of Rejuvenated irrigation tanks. A case study in mid Godavari Basin (http://publications.iwmi.org/pdf/H042911.pdf ) ii. Ramesh K and Elango L (2011, July) Groundwater qualityand its suitability for domestic and agricultural use in Tondiar river basin, Tamil nadu India; Environmental Monitoring Assessment: DOI 10.1007/s10661-011-22313/Springer Science business Media B.V. iii. Raihan F, Alam J B (2008) Assessment of groundwater quality in Sunamganj of Bangladesh; Iranian Journal of Environmental Health Science and Engineering , Vol. 5, No.3, pp. 155 – 166. iv. Sakthivadivel R, Gomathinayagam P and Tushaar, S (2010, July 31) Rejuvenating Irrigation Tanks through local institutions, Economic and political. v. Yogendra K, Puttaiah E.T (2007) Determination of water quality index and suitability of an urban water body in Shimoga Town, Karnataka (Paper presented at the 12th World Lake Conference), 2007.

Booster Chlorination Strategy For Managing Chlorine Disinfection In Drinking Water Distribution System – A Review Roopali V. Goyal 1, Dr. H.M. Patel2 Research Scholar , The M.S. University of Baroda, Vadodara , Assistant Professor, Civil Engineering Department, Sardar Vallabhbhai Patel Institute of Technology Vasad, Dist . Anand 388 306. Gujarat, India. 2 Head and Professor, Civil Engineering Department, Faculty of Technology & Engg, The M.S University of Baroda, Vadodara. 390 001, Gujarat, India. Email: [email protected], [email protected]

1

ABSTRACT The amount of residual chlorine in a Drinking water distribution system (DWDS) is commonly used as an indicator of water quality supplied to the consumers. Adequate amount of residual chlorine ensures the microbiological safety, and excess chlorination leads to taste, odour, or by-product problems. Compared to conventional methods that apply disinfectant only at the source, in booster chlorination, chlorine is supplied at strategic locations throughout the distribution network can

HYDRO 2014 International

The objective of this paper is to review the work of various researchers who have effectively applied the strategy of booster chlorination for managing chlorination by modeling of booster chlorination using various modeling tools. Further the available literature is extended to explore the work done by various investigators who have applied the different optimization methods for (i) Optimal scheduling of injection rates of chlorine and optimal operation of booster stations and (ii) Optimal location of booster stations in the water distribution network. In addition to the normal operation of booster stations a limited amount of research that has explored the application of booster stations to the contamination incident problem and other applications is also included in this review. After reviewing the work of all researchers it is found that coupling of water quality modeling tool with advanced optimization methods can serve as important decision making tool for management of water quality in the DWDS. Keywords: Booster chlorination, Drinking water distribution system, Optimization methods. 1. INTRODUCTION: Inadequate chlorine residual in drinking water distribution increases potential for the breakthrough of organisms and can ultimately result in public health and regulatory compliance problems. As chlorine is reactive, it reacts with natural organic and inorganic matter in water which decreases the chlorine concentration with time called chlorine decay. The long term chlorine decay in distributed drinking water and in natural waters receiving chlorinated discharges can be modeled by using first order kinetics ( Johnson 1978 ; Hass and Kara 1984; Rossman 1994; Powell et al. 2000) is given by, C = Co e(−K t) (1) Where, Ct= Chlorine concentration at time t, mg/l Co= Initial chlorine concentration, mg/l T= Time ( hour) K= First order reaction rate coefficient (hr -1) As seen from the above equation the residual chlorine concentration is the function of initial chlorine concentration, travelling time and decay coefficient. To maintain the adequate residual chlorine at the farthest end more amount of chlorine is supplied at source in conventional methods to compensate the loss of chlorine. But this can generate higher disinfection by products ( DBPs), and bring odour and taste complaints. Booster chlorination is the best strategy to maintain the balance between lower and upper limit of the residual chlorine concentration in which, disinfectant is applied at strategic locations within the distribution system to compensate the losses that occur as it decays over time (Boccelli et al.,1998; Tryby et al., 2002). Many researchers have worked on the modeling of booster chlorination using the water quality modeling tool for the prediction of residual chlorine concentration in DWDS as it is

MANIT Bhopal

Page 175

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

the prerequisite for the modeling of the Booster chlorination. Although booster disinfection is commonly practiced, a standardized procedure for the location and operation of booster stations has not been adopted in the water utility community. Booster stations are often located near areas with low levels of disinfectant residual, and they are operated with regard to the local goals of increased residual which often ignores the systemlevel interactions (Haxton et al. 2011). In area of water distribution system analysis, Optimization models are used for calibration, design, and operation purpose using various kinds of algorithms. The coupling of such water quality model with advanced optimization methods can serve as an important decision support model for the water supply authority for scheduling and mass rate application of chlorine at storage reservoir for maintaining chlorine with range in DWDS at all the nodes. 2. MODELING AND OPTIMIZATION OF BOOSTER CHLORINATION: For the effective modeling of the Booster Chlorination station, the accurate prediction of the residual chlorine concentration is required, for which many water quality modeling tools are available. The usability of these models was greatly improved in the 1990s with the introduction of the public domain EPANET model (Rossman, 1994). The model considers first-order reactions of chlorine to occur both in the bulk flow and the pipe wall as mentioned in equation 1. It is used by most of the researchers to find out the residual chlorine concentration in DWDS (Boccelli et al,1998; Tryby et al., 2002; Munavalli and Kumar 2003; Prasad et al. 2004; Tryby et al. 1999; Uçaner and Ozdemir 2003; Propato and Uber 2004a,b; Ostfeld and Salomons 2005, 2006; Kang and Lansey 2010; Haxton et al. 2011). The booster stations are introduced in EPANET by water quality sources nodes where the quality of external flow entering the network is specified. EPANET can model the four types of sources. (i) A concentration source fixes the concentration of any external inflow entering the network at a node (ii) A mass booster source adds a fixed mass flow to that entering the node from other points in the network.(iii)A flow paced booster source adds a fixed concentration to that resulting from the mixing of all inflow to the node from other points in the network (iv)A set point booster source fixes the concentration of any flow leaving the node. (EPANET user‟s Manual, 2000). A new version of EPANET, the EPANET Multi-Species Extension or EPANET MSX (Shang et.al., 2008) which can be utilized for the modeling of two source chlorine decay uses the same first order chlorine decay equation as EPANET is also utilized by different researchers (Carrico and Singer 2009; Parks and Van Briesen 2009; Ohar, Z. and Ostfeld, A. 2010, 2014; Haxton et al. 2011) for the prediction of residual chlorine. There is wide application of optimization methods for various engineering applications including Booster Chlorination Station. The optimization methods can be utilized for minimizing of the mass rate of chlorine applied at booster station, optimization of location of booster station and its operation with the constraint of minimum residual chlorine at the locations of DWDS. Available Literature on the application of various methods of

HYDRO 2014 International

optimization for the booster chlorination stations is divided into two major categories (i) Optimal scheduling of disinfectant injection and operation of Booster Station (ii) Optimal location of Booster Stations. 3. OPTIMAL SCHEDULING OF DISINFECTANT INJECTION AND OPERATION OF BOOSTER STATION: The purpose of optimum scheduling of chlorine injection is to minimize the total dose of chlorine at source and booster stations at the same time to satisfy the constraint of maintaining the minimum residual chlorine at all the locations of DWDS. Boccelli et al. (1998) formulated a linear optimization model for the scheduling of disinfectant injections into water distribution systems. They used EPANET water quality model to quantify disinfectant transport and decay as a function of the booster dose schedule using the principle of linear superposition and firstorder reaction kinetic to avoid the computational burden of water quality simulations during optimization and booster station operation problem . Tryby et al. (2002) extended the linear programming (LP) booster disinfection scheduling model presented by Boccelli et al. (1998) to incorporate booster station location as a decision variable within the optimization process. The formulation was similar to the general, mixed-integer linear programming, fixedcharge facility location problem, and was solved using a branchand-bound solution procedure using coupling the data using EPANET water quality simulator. Munavalli and Mohan Kumar (2003) formulated a optimal scheduling model in terms of a nonlinear optimization problem to determine the chlorine dosage at the water quality sources using (GA) approach in which decision variables (chlorine dosage) were coded as binary strings and solved by linking EPANET with a genetic algorithm (GA). For the linear chlorine reaction kinetics (first-order reaction kinetics) the principle of linear superposition was utilized to compute dynamic chlorine concentrations without running the dynamic water quality simulation model. Uçaner and Ozdemir (2003) studied, the locations, injection rates and scheduling of chlorine booster stations using genetic algorithms by coupling the hydraulic solution and chlorine concentration distribution using EPANET software. Prasad et al. (2004) investigated the booster facility location and injection scheduling problem formulated as a multi objective genetic algorithm optimization model using the theory of linear super position in water quality modeling for calculating concentration profiles at network nodes. A multi objective genetic algorithm called NSGA-II was used in solving the twoobjective problem. Ostfeld and Salomons (2004) presented the methodology and application of a genetic algorithm (GA) scheme, tailor-made to EPANET for simultaneously optimizing the scheduling of existing pumping and booster disinfection units, as well as the

MANIT Bhopal

Page 176

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

design of new disinfection booster chlorination stations, under unsteady hydraulics.

carried out to find out the optimal locations of booster stations is presented in the following paragraph

Propato and Uber (2004 a) formulated a linear least-squares problem to determine the optimal disinfectant injection rates that minimize variation in the system residual space-time distribution with assumption of known locations of booster stations . To investigate the performance and limitations of the proposed LLS problem was applied on a Cherry hill/Brushy plains DWDS .

Tryby M. and Uber J. (1999) developed a mixed integer linear programming method to provide optimal locations and operating data for booster disinfection stations in drinking water distribution systems. The problem formulation was related to the general fixed charge facility location problem, requiring that a branch and bound solution procedure be used.

Propato and Uber (2004b) extended their previous work to include the locations of the booster stations as decision variables and formulated a mixed-integer quadratic programming ( MIQP) problem to locate booster stations and to identify their dosage schedules for maintaining disinfectant residual in DWDS. Solution of the problem was done via the branch-andbound technique with quadratic programming sub problems. Ostfeld and Salomons (2006) presented two different optimization objectives for optimal pump operation and booster disinfection. The proposed objectives were (1) minimization of the cost of pumping and the booster stations operation and (2) maximization of the chlorine injected in order to maximize the system protection. The problem was solved using a GA linked with EPANET. Gibbs et al. (2010) studied the booster disinfection dosing problem, including daily pump scheduling, for a real system in Sydney, Australia using GA to optimize the operation of the Woronora WDS. Kang and Lansey (2010) formulated a real-time optimal valve operation coupled with booster disinfection problem as a single objective optimization model. The problem was solved using a genetic algorithm (GA) linked with EPANET. Ohar Z and Ostfeld A. (2010) extended the authors previous work on the usage of chlorine - TTHM multi species model for optimal design and operation of booster chlorination stations. An alternative model formulation was suggested by adding constraints requiring that the concentrations of all species at the beginning and end of the design period be the same Ohar, Z. and Ostfeld, A. (2014) formulated and solved model to set the required chlorination dose of the boosters for delivering water at acceptable residual chlorine and TTHM concentrations for minimizing the overall cost of booster placement, construction, and operation under extended period hydraulic simulation conditions through utilizing a multi-species approach. The developed methodology linked a genetic algorithm with EPANET-MSX. 4. OPTIMAL LOCATION OF BOOSTER STATIONS:

Constans S. et al (2000) proposed linear programming formulations to determine the optimal locations where disinfectant must be added and optimize the injection patterns. Solution of the proposed optimization problem not only gave the best booster stations locations and injection patterns, but also calculated the corresponding chlorine patterns at all the nodes of the network. Avi Ostfeld (2005) determined the optimal location of a set of monitoring stations aimed at detecting deliberate external terrorist hazard intrusions through water distribution system nodes: sources, tanks, treatment plant intakes.The methodology implemented in a non commercial program entitled optiMQ-S linking optiGA and EPANET. Lansey et al. (2007) assumed first-order reaction kinetics and formulated an integer linear programming optimization problem to determine the optimal location of booster stations as well as their injection rates. The problem was solved using a GA. Wang Hongxiang et al (2010) formulated an optimization model in the presence of partial coverage based on the maximum covering location problem for locating optimal booster chlorination stations in water distribution systems. A hybrid PSO, combined with GA algorithms, was proposed to get the solution which was applied to a hypothetical network . Wang Hongxiang ( 2010) introduced an optimization model to identify optimal booster chlorination stations in water distribution systems in the presence of partial coverage based on the maximum covering location programming model (MCLP). Ant Colony Optimization Algorithms was applied to optimize the booster chlorination stations model. To improve the optimization ability of ACOAs and avoid getting in the local optimal solution, the Max-Min ACOAs were adopted, and a sensitivity-based visibility factor was applied to the ACOAs to a case study . Table no 1 gives the summary of various optimization methods used for the optimal scheduling, operation and location of booster stations. Table no 2 gives the summary of various objectives proposed by different researchers. Table 1. Optimization methods for optimal scheduling, operation and location of Booster Station

Optimal Locations of the booster station is equally important as the operations and scheduling of chlorine doses. The work

HYDRO 2014 International

MANIT Bhopal

Page 177

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 efficient and realistic than boosting on a preset schedule by assuming that the sensor network is detecting a low concentration of chlorine due to contamination or unpredictable demand. Brian Carrico and Philip C. Singer (2009) checked the effect of conventional and booster chlorination on chlorine residuals and Trihalomethans (THM) formation in drinking water distribution systems using EPANET and EPANET -MSX model.

Table 2. Objective functions used for the optimization methods

Haxton et al. (2011) studied the problem of locating booster stations to support booster disinfection in the context of a contamination incident with objective to locate a given number of booster stations using two different ways of formulating a booster station optimization. The first optimization formulation was using multi-species EPANET-MSX software to evaluate the effects of chlorine utilization and contaminant reactions. The second optimization formulation used an algebraic model for modeling the flow of contaminants and chlorine in the network. Nilufar Islam et al. ( 2013) proposed an innovative scheme for maintaining adequate residual chlorine with optimal chlorine dosages and numbers of booster locations was established based on a proposed WQI for The City of Kelowna , Canada water distribution network using EPANET software and later coupled with an optimization scheme. Table no 3 narrates the major findings of various researchers. Table 3. Major Findings of Various researchers by application of Booster Chlorination

5. BOOSTER CHLORINATION RESPONDING TO A CONTAMINATION INCIDENT AND OTHER APPLICATIONS: Various investigators worked on the different field to check the effect of applications of booster chlorination towards contaminant events and formation of disinfection by-products. Some of the studies are mentioned here. Propato and Uber (2004c) applied the booster chlorination strategy to two example networks under a worst-case deliberate intrusion scenario. Results saw that the risk of consumer exposure is affected by the residual maintenance strategy employed. They found that addition of a booster station at storage tanks may improve consumer protection without requiring excessive disinfectant. Parks and Van Briesen (2009) tested the hypothesis that a booster disinfection system used in conjunction with a sensor network boost-response system could provide substantial protection to allow for uninterrupted high quality water service during an intrusion event using EPANET EPANET-MSX to perform the water quality simulations. The hypothesis was evaluated that a reactive booster schedule would be more

HYDRO 2014 International

6. DISCUSSIONS AND CONCLUDING REMARK: After reviewing the work of most of the researchers it is found that coupled water quality modeling tool with advanced optimization methods can serve as important decision making tool for the operation of booster chlorination station to manage

MANIT Bhopal

Page 178

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

effective residual chlorine in the DWDS. Investigators utilized different methods of optimization for optimal scheduling, operation and locations of booster stations to maintain adequate levels of residual chlorine throughout the DWDS. Many researchers have linked the water quality model such as EPANET or EPANET- MSX with optimization methods to achieve the balance between the upper and lower limit of residual chlorine. As seen from summary it is observed that linear programming model , mixed integer linear programming and Genetic Algorithm is widely used by many researchers. Limited research papers are found with applications of evolutionary algorithms like Particle Swarm Optimization (PSO). The investigation carried out by various researchers suggests that the application of booster chlorination strategy can maintain the balance between the upper and lower limits of residual chlorine. Studies of most of the researchers show that the booster chlorination can reduce the amount of disinfectant required to satisfy concentration constraints, when compared to conventional disinfection only at the source. This reduced concentration may help in reduction of harmful disinfection byproduct formation. Thus, the application of linked water quality and optimization model serve as the important decision supporting tool for the water supply mangers for effective management of residual chlorine in DWDS. This will ultimately provide the protection against the pathogens and harmful disinfection by-products to consumers. REFERENCES i. Boccelli, D. L., Tryby, M. E., Uber, J. G., Rossman, L. A., Zierolf, M. L., and Polycarpou, M. M. (1998). Optimal scheduling of booster disinfection in water distribution systems. Journal of Water Resources Planning and Management, 124(2), 99-111. ii. Boccelli, D. L., Tryby, M. E., Uber, J. G., & Summers, R. S. (2003). A reactive species model for chlorine decay and THM formation under rechlorination conditions. Water Research, 37(11), 2654–2666. iii. Brian Carrioca, Phillip C Singer(2009) Impact of Booster Chlorination on Chlorine Decay and THM production: Simulated Analysis. ASCE Journal of Environmental Engineering 135( 10 ), 928-935. iv. Constans, S., Bremond, B., and Morel, P. (2000) Using Linear Programs to Optimize the Chlorine Concentrations in Water Distribution Networks. Building Partnerships: Joint Conference on Water Resource Engineering and Water Resources Planning and Management 2000 Minneapolis, Minnesota, United States pp. 1-12. v. Haxton, T., Murray, R., Hart, W., Klise, K., and Phillips, C. (2011) Formulation of Chlorine and Decontamination Booster Station Optimization Problem. World Environmental and Water Resources Congress, 199-205. vi. J.D., Johnson( 1978) Measurement and Persistence of Chlorine Residuals in Natural Watersin Water Quality Modeling by Clark, 2012. vii. Haas C.N., S.B. Karra (1984) Studies on Chlorine Demand Constants." Journal of WPCF 56(2) 170-173. viii. Kang, D., and Lansey, K. (2010). Real-Time Optimal Valve Operation and Booster Disinfection for Water Quality in Water Distribution Systems. Journal of Water Resources Planning and Management, 136(4), 463473. ix. Lansey, K., Pasha, F., Pool, S., Elshorbagy, W., and Uber, J. (2007). Locating satellite booster disinfectant stations. Journal of Water Resources Planning and Management, 133(4), 372-376. x. Matthew S. Gibbs., Graeme C. Dandy., and Holger R. Maier ( 2010). Calibration and Optimization of the Pumping and Disinfection of a Real Water Supply System .Journal of Water Resources Planning and Management, 136(4), 493-501. xi. Munavalli, G. R., and Kumar, M. S. M. (2003). Optimal scheduling of multiple chlorine sources in water distribution systems. Journal of Water Resources Planning and Management, 129(6), 493-504.

HYDRO 2014 International

xii. Nilufar Islam, Rehan Sadiq, Manuel J. Rodriguez ( 2013) Optimizing booster chlorination in water distribution networks: a water quality index approach,Journal of Environmental Monitoring and Assessment , 185( 10) , 8035-8050. xiii. Ostfeld, A., and Salomons, E. (2004).Optimal layout of early warning detection stations for water distribution systems security, Journal of Water Resource Planning and Management, 130(5), 377–385. xiv. Ostfeld, A., and Salomons, E. (2005).Securing Water Distribution Systems Using Online Contamination Monitoring, Journal of Water Resources Planning and Management, 131( 5) xv. Ostfeld, A., and Salomons, E. (2006) Conjunctive optimal scheduling of pumping and booster chlorine injections in water distribution systems, Engineering Optimization, 38(3), 337-352. xvi. Ohar, Z. and Ostfeld, A. (2010) Alternative Formulation for DBP's Minimization by Optimal Design of Booster Chlorination Stations, World Environmental and Water Resources Congress 2010: pp. 4260-4269. xvii. Ohar, Z. and Ostfeld, A. (2014) Optimal design and operation of booster chlorination stations layout in water distribution systems,Water Research 58( 1) , 209–220 xviii. Parks, S. L. I., and VanBriesen, J. M. (2009)Booster Disinfection for Response to Contamination in a Drinking Water Distribution System, Journal of Water Resources Planning and Management, 135(6), 502-511. xix. Prasad, T. D., Walters, G. A., and Savic, D. A. (2004) Booster disinfection of water supply networks: Multiobjective approach, Journal of Water Resources Planning and Management, 130(5), 367-376. xx. Propato, M., and Uber, J. G. (2004a) Linear least-squares formulation for operation of booster disinfection systems, Journal of Water Resources Planning and Management, 130(1), 53-62. xxi. Propato, M., and Uber, J. G. (2004b) Booster system design using mixed-integer quadratic programming, Journal of Water Resources Planning and Management, 130(4), 348-352. xxii. Propato, M., and Uber, J. G. (2004c) Vulnerability of water distribution systems to pathogen intrusion: How effective is a disinfectant residual?, Environmental Science & Technology, 38(13), 3713-3722. xxiii. Powell J.C., N.B. Hallam, J.R. West, C.F. Forester, and J.Simmsm (2000) Factors which control Bulk Chlorine Decay Rates. Water Research 34( 1), 117-126. xxiv. Rossman L.A., Robert Clark, Walter Grayman. (1994) Modeling Chlorine Residuals in Drinking Water Distribution Systems. ASCE Journal of Environmental Engineering 120( 4) 803-820. xxv. Rossman L. A. (2000). EPANET 2.0 - User Manual. United States Environmental Protection Agency - EPA. Cincinnati, USA. xxvi. Shang, F., Uber, J. G., and Rossman, L. A. (2008). EPANET multispecies extension users‘ manual, EPA/600/S-07/021, U.S. EPA, Cincinnati. xxvii. Tryby, M. and Uber, J. (1999) Development of a Booster Chlorination Design Using Distribution System Models. 29th Annual Water Resources Planning and Management Conference, Tempe, Arizona, United States,WRPMD'99: 1-9. xxviii. Tryby, M. E., Boccelli, D. L., Uber, J. G., and Rossman, L. A. (2002) Facility location model for booster disinfection of water supply networks, Journal of Water Resources Planning and Management, 128(5), 322-333. xxix. Ucaner M.and Ozdemir (2003) Application of Genetic Algorithms for Booster Chlorination in Water Supply Networks, World Water & Environmental Resources Congress , Philadelphia, Pennsylvania, United States: American Society of Civil Engineers. xxx. Wang, Hongxiang ,Guo Wenxian ; Xu Jianxin ; Gu Hongmei (2010), A Hybrid PSO for Optimizing Locations of Booster Chlorination Stations in Water Distribution Systems. International Conference on Intelligent Computation Technology and Automation (ICICTA), China Univ. of Water Resources & Electr. Power, China, Volume 1 , 126- 129 xxxi. Wang, Hongxiang. ( 2010) Ant Colony Optimization for Booster Chlorination Stations of Water Distribution Systems, International Conference on Computer Application and System Modeling ( ICCASM). China, VI 166-VI 170.

MANIT Bhopal

Page 179

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Hydrogeochemical Stuidies Of Groundwater In And Around Metropolitan City Vadodara, Gujarat, India M.K. Sharma C.K. Jain National Institute of Hydrology, Roorkee – 247667, India E. mail: [email protected] ABSTRACT : Geo-environmental conditions have a marked influence on the groundwater quality. Hydrogeochemical studies relevant to the water quality explain the relationship of water chemistry to aquifer lithology. Such relationship would help not only to explain the origin and distribution of dissolved constituents but also to elucidate the factors controlling the groundwater chemistry. In the present investigation, hydrogeochemical study was carried out in and around the metropolitan city Vadodara, Gujarat, India to identify and delineate the important geochemical processes which were responsible for the evaluation of chemical composition of groundwater. The study area is a part of Indo-gangetic Plains, composed of Pleistocene and subrecent alluvium. The groundwater in the study area occurs under both the unconfined and confined conditions. Groundwater conditions in the alluvial terrains are considerably influenced by varying lithology of subsurface formations. The rainfall is main recharge source of groundwater body besides infiltration from river, canals and return flow from irrigation. Thirty five groundwater sources viz; open wells, tubewells, piezometric wells, bore wells and hand pumps in and around Vadodara city in pre- and post-monsoon seasons during 2008 and 2009 were collected and analysed for major constituents. Data has been processed as using Piper Trilinear Diagram and it was observed that majority of the groundwater samples of the study area belong to Ca-Mg-Cl-SO4 or Na-K-Cl-SO4 hydrochemical facies in both pre- and post-monsoon seasons. Gibbs ratio plot indicate that the chemistry of groundwater in the study area is controlled mainly by the chemical interaction between aquifer rocks and groundwater, and to some extent by processes like evapo-transpiration etc. The process of evaporation might have incorporated some components of sodium and chlorine ions. The scatter plots of ions show that the relatively high contribution of (Ca+Mg) to the total cations (TZ +) and high (Ca+Mg)/(Na+K) ratio indicate that carbonate weathering is a major source of dissolved ions in the groundwater of the study area. The plot of (Ca+Mg) vs HCO3 for most of the samples in study area indicates an excess of Ca+Mg over HCO3 inferring an extra source of Ca and Mg. This requires that a portion of the (Ca+Mg) has to be balanced by other anions like SO 4 and/or Cl. Plot of (Ca+Mg) vs HCO3+SO4 shows the ion exchange process activated in the area, which may be due to the excess bicarbonate. The plot of Na vs Cl indicates contribution of silicate weathering through the release of Na. Key words: Groundwater, Hydrogeochemical process, Vadodara, Gibbs Plot, Scatter Plot 1. INTRODUCTION Ground water plays an important role in our life support system as it is being used for different designated uses specially for

HYDRO 2014 International

drinking purpose. Groundwater situation in different parts of India is diversified because of variation in geological, climatological and topographic set-up. The prevalent rock formations, ranging in age from Archaean to Recent, which control occurrence and movement of groundwater, are widely varied in composition and structure. Further, significant variations of landforms from the rugged mountainous terrains of the Himalayas, Eastern and Western Ghats to the flat alluvial plains of the river valleys and coastal tracts, and the aeolian deserts of Rajasthan are also responsible non-uniform distribution of ground water. The rainfall patterns too show similar region-wise variations. The topography and rainfall virtually control run-off and groundwater recharge (Master Plan, 2002). Growing demand of water in various sectors viz; agriculture, industrial and domestic sectors, has brought problems of overexploitation of the groundwater resource, continuously declining groundwater levels, sea water ingress in coastal areas, and groundwater pollution in different parts of the country. The falling groundwater levels in various parts of the country have threatened the sustainability of the groundwater resource, as water levels have gone deep beyond the economic lifts of pumping. Geo-environmental conditions have a marked influence on the groundwater quality. Hydrogeochemical studies relevant to the water quality explain the relationship of water chemistry to aquifer lithology. Such relationship would help not only to explain the origin and distribution of dissolved constituents but also to elucidate the factors controlling the groundwater chemistry. Kumar et al. (2006) also studied the hydrogeochemical processes in NCT Delhi to identify the geochemical processes and their relation with groundwater quality as well as to get an insight into the hydrochemical evaluation of groundwater and reported that salinity and nitrate are two major problem from drinking point of view. The prevailing hydrochemical processes operating in the study area are simple dissolution, mixing, weathering of carbonate minerals (kankar) and of silicate, ion exchange, and surface water interaction. Limited reverse ion exchange has been noticed in a few parts of the study area especially in post-monsoon periods. Periodic seasonal switch-over has been clearly noticed in these hydrogeochemical processes that control groundwater quality of the area. Reddy and Kumar (2010) carried out hydrogeochemical studies in Penna-Chitravahi river basins in Southern India to identify and delineate the geochemical processes responsible for the evolution of chemical composition of ground water and reported that the groundwater in general is of Na +-Cl-, Na+-HCO3-, Ca2+Mg2+-HCO3- and Ca2+-Mg2+-Cl- type . Na+ among cations and Cl- and/or HCO3- among anions dominate the water; Na+ and Ca2+ are in the transitional state with Na+ replacing Ca2+ and HCO3- Cl- due to physicochemical changes in the aquifer and water rock interactions. Further, Gibbs plots indicate that the evolution of water chemistry is influenced by water-rock interaction followed by evapotranspiration process. Vijaykumar et al. (2010) studied hydrogeochemistry in the part of Ariyalur

MANIT Bhopal

Page 180

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

region, Perambalur district, Tamil Nadu, India and reported that Ca+Mg, SO4+Cl and HCO3+CO3 are high facies during pre- and post-monsoon season and evaporation process dominates the groundwater chemistry as explained by Gibbs plot. The quality of water for irrigation was estimated by USSL classification indicating high salinity and low sodium hazard, satisfactory for plants having moderate salt tolerance on soils. Obiefuna and Orazulike (2011) characterized groundwater in semiarid Yola area of northeastern Nigeria employing chemical indicators and reported that alkaline earths (Ca+Mg) significantly exceed the alkali (Na+K) and week acids (HCO3+CO3) exceed the strong acids (Cl+SO4), suggesting dominance of carbonate weathering followed by silicate weathering. Chemical fertilizers and anthropogenic activities are contributing to sulphate, nitrate and chloride concentrations in surface and ground water of the study area. Srinivasamoorthy et al. (2012) made an attempt to identify the major geochemical process activated for controlling the ground water chemistry of Sarabanga minor basin of river Cauvery, situated in Salem district, Tamil Nadu, India and inferred that water chemistry is guided by complex weathering process, ion exchange along with influence of Cl ions from anthropogenic impact. In the present paper, hydrogeochemical study in and around the metropolitan city Vadodara, Gujarat, India is carried out to identify and delineate the important geochemical processes which were responsible for the evaluation of chemical composition of groundwater by collecting groundwater samples in pre- and post-monsoon season. 2. STUDY AREA The metropolitan city Vadodara is the graceful city of Gujarat State. It is bounded by 22°18′ N latitude and 73°16′ E longitude (Fig.1). Vadodara urban agglomeration covers an area of about 140 km2. The rivers Jambua, Surya, Vishwamitri and Dhadhar, which flow through central part of the district and empty into Gulf of Khambat, are also part of Mahi Basin. The climate of the metropolitan city is moderate tropical type. The temperature of the city varies from 8˚C to 46˚C. The average annual rainfall is recorded as 900 mm. The study area is a part of Indo-gangetic Plains, composed of Pleistocene and subrecent alluvium. The earliest geological evolution of the basement rocks, exposed in northern and eastern parts, had been controlled by the Precambrian orogenies (Arvalli and Delhi cycles), and the older crystalline rocks ideally shows folds, faults and magmatism related to the two orogenies. After Precambrian orogenies, major geological events of Vadodara district were confined to Mesozoic and Cenozoic Eras which can be related with the breaking up of the Gondwana land and the subsequent northward drift of the Indian sub-continent, involving formation of sediments and Deccan Trap Volcanism with uplifts and subsidence along the two major lineaments – Narmada and Cambay rift system. The groundwater in the study area occurs under both the un-confined and confined conditions. Groundwater conditions in the alluvial terrains are considerably influenced by varying lithology of subsurface formations. The rainfall is main recharge source of groundwater body besides

HYDRO 2014 International

infiltration from river, canals and return flow from irrigation. There is no yield of water upto 50 feet, sandy aquifer was found from 50 to 70 feet. The principal industrial areas within Vadodara Urban areas are at Makarpura and Nandesari. 3. MATERIAL AND METHODS Thirty five groundwater samples from open wells, tubewells, piezometric wells, bore wells and hand pumps in and around Vadodara city (Fig. 1) were collected for physico-chemical analysis in polypropylene bottles in pre- and post-monsoon seasons during 2008 and 2009. All the samples were stored in sampling kits maintained at 4oC and brought to the laboratory for detailed chemical analysis. All general chemicals used in the study were of analytical reagent grade (Merck/BDH). Deionized water was used throughout the study. The physicochemical analysis was performed following standard methods (APHA, 1995).Ionic balance was calculated, the error in the ionic balance for majority of the samples was within 5%. 4. RESULTS AND DISCUSSIONS 4.1 Physico-chemical characteristics of groundwater The hydro-chemical data of groundwater samples of premonsoon, 2008 is presented in Table 1. The pH values in the groundwater of metropolitan city of Vadodara mostly fall within the range 7.6 to 8.6. The pH values for most of the samples are well within the limits prescribed by BIS (2012) for various uses of water including drinking and other domestic supplies. The electrical conductivity and dissolved salt concentrations are directly related to the concentration of ionized substance in water and may also be related to problems of excessive hardness and/or other mineral contamination. The conductivity values in the groundwater samples of the metropolitan city vary widely from 760 to 5480 S/cm with almost 80% of the samples having conductivity value above 1000 S/cm. The maximum conductivity value of 5480 S/cm was observed in the sample of Harni. In the metropolitan city of Vadodara, the values of total dissolved solids (TDS) in the groundwater varies from 486 to 3507 mg/L. Almost all the samples were found above the acceptable limit but within the maximum permissible limit of 2000 mg/L and only 14% of the samples exceed the maximum permissible limit of 2000 mg/L. Water containing more than 500 mg/L of TDS is not considered desirable for drinking water supplies, though more highly mineralized water is also used where better water is not available. For this reason, 500 mg/L as the acceptable limit and 2000 mg/L as the maximum permissible limit has been suggested for drinking water (BIS, 2012). Water containing TDS more than 500 mg/L causes gastrointestinal irritation (BIS, 2012). The presence of calcium and magnesium along with their carbonates, sulphates and chlorides are the main cause of hardness in the water. A limit of 200 mg/L as acceptable limit and 600 mg/L as permissible limit has been recommended for drinking water (BIS, 2012). The total hardness values in the study area range from 79 to 1144 mg/L. About 20% of the

MANIT Bhopal

Page 181

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

samples fall within acceptable limit of 200 mg/L and 29% sample cross the permissible limit of 600 mg/L. In groundwater of the study area, the values of calcium range from 12 to 313 mg/L. The values of magnesium vary from 12 to 127 mg/L. The acceptable limit for calcium and magnesium for drinking water are 75 and 30 mg/L respectively (BIS, 2012). Further, only few samples exceed maximum permissible limit of calcium as 200 mg/L and magnesium as 100 mg/L. The concentration of sodium in the study area varies from 54 to 1110 mg/L. High sodium values in the city may be attributed to base-exchange phenomena causing sodium hazards. Such groundwater with high value of sodium is not suitable for irrigation purpose. The concentration of potassium in groundwater of the study area varies from 1.0 to 77 mg/L. As per EEC criteria, ten samples exceed the guideline level of 10 mg/L.

45 mg/L and six samples even cross the permissible limit of 45 mg/L. In higher concentrations, nitrate may produce a disease known as methaemoglobinaemia (blue babies) which generally affects bottle-fed infants. The higher nitrate concentration in the metropolitan city at few locations may be attributed due to combined effect of contamination from domestic sewage, livestock rearing landfills and runoff from fertilized fields. The fluoride content in the groundwater of the study area varies from 0.00 to 1.26 mg/L. Almost all the samples of the metropolitan city fall within the acceptable limit of 1.0 mg/L and none of the samples exceeded the maximum permissible limit of 1.5 mg/L. From the above discussion, it is clearly indicated that in the groundwater of metropolitan city of Vadodara, the concentration of total dissolved solids exceeds the acceptable limit of 500 mg/L in almost all the samples but within the maximum permissible limit of 2000 mg/L. From the hardness point of view, about 20% of the samples fall within acceptable limit of 200 mg/L and 29% sample cross the permissible limit of 600 mg/L. The chloride content exceeds the desirable limit in more than 40% of the pre-monsoon samples. Sulphate contents are within the desirable limits in about 89% samples. The nitrate content in more than 84% samples is well within the permissible limit. The concentration of fluoride in almost all the samples is well within the desirable limit. The violation of BIS limit could not be ascertained for sodium and potassium as no permissible limit for these constituents has been prescribed in BIS drinking water specifications. Table 1. Hydro-chemical characteristics of the Groundwater during Pre-monsoon 2008 Parameters

Figure 1. Map showing location of sampling sites The concentration of chloride varies from 20 to 1464 mg/L. More than 60% samples of the metropolitan city falls within the desirable limit of 250 mg/L and only three samples of the city exceeds the maximum permissible limit of 1000 mg/L. The concentration of sulphate in the metropolitan city varies from 6 to 600 mg/L. Bureau of Indian standard has prescribed 200 mg/L as the desirable limit and 400 mg/L as the permissible limit for sulphate in drinking water. In the study area, 89% of the samples analysed fall within the desirable limit of 200 mg/L and only two samples exceed the maximum permissible limit of 400 mg/L. The nitrate content in the metropolitan city of Vadodara varies from 0.0 to 252 mg/L. About 84% of the samples of the metropolitan city of Vadodara fall within the permissible limit of

HYDRO 2014 International

pH Conductivity, S/cm TDS, mg/L Hardness, mg/L Chloride, mg/L Sulphate, mg/L Nitrate, mg/L Fluoride, mg/L Sodium, mg/L Potassium, mg/L Calcium, mg/L Magnesium, mg/L

Mini mum

Maxim um

Aver age

7.6 760

8.6 5480

8.0 2013

486 79 20 6.0 0.0 0.0 54 1.0 12 12

3507 1143 1464 600 252 1.3 1110 77 313 127

1288 435 320 112 36 0.6 250 11.7 103 43

BIS (2012) Limit Accepta Permis ble sible 6.5 8.5 500 200 250 200 45 1.0 75 30

2000 400 1000 400 1.5 200 100

4.2 Mechanism Controlling the Groundwater Chemistry Geo-environmental conditions have a marked influence on the groundwater quality. Hydrogeochemical studies relevant to the water quality explain the relationship of water chemistry to aquifer lithology. Such relationship would help not only to explain the origin and distribution of dissolved constituents but also to elucidate the factors controlling the groundwater chemistry. Gibbs (1970) proposed a hypothesis to elucidate the major natural mechanisms controlling world water chemistry. Three mechanisms – atmospheric precipitation, rock dominance and the evaporation-crystallization process – are the major factors controlling the composition of dissolved salts of the

MANIT Bhopal

Page 182

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

world waters. Other second-order factors, such as relief, vegetation and composition of material in the basin dictate only minor deviations within the zones dominated by the three prime factors. Gibbs plot is a diagrammatic representation of the mechanisms responsible for controlling the chemical composition of various bodies of water on the surface of the earth. The major cations that characterize the end-members of the world surface waters are Ca for freshwater bodies and Na for high-saline water bodies. Gibbs plotted the weight ratio Na/(Na+Ca) on the x-axis and the variation in total salinity on the y-axis (Fig. 2). This ordered arrangement can serve as a basis for discussion of the several mechanisms that control world water chemistry. The first of these mechanisms is the atmospheric precipitation. The chemical compositions of low-salinity waters are controlled by the amount of dissolved salts furnished by precipitation. These waters consist mainly of the rivers having sources in thoroughly leached areas of low relief in which the rate of supply of dissolved salts to the rivers is very low and the amount of rainfall is high – much greater in proportion to the low amount of dissolved salts supplied from the rocks. In addition, the composition of this precipitation differs from that of rockderived dissolved salts. The second mechanism is the rock dominance controlling world water chemistry. The waters of this rock-dominated end-members are more or less in partial equilibrium with the materials in their basins. Their positions within this grouping are dependent on the relief and climate of each basin and the composition of each basin. The third major mechanism that controls the chemical composition of the earth‟s surface waters is the evaporation-fractional crystallization process. This mechanism produces a series extending from the Ca-rich, medium-salinity (freshwater), „rock source‟ endmember grouping to the opposite, Na-rich, high-salinity endmember.

Figure 2. Gibbs plot (Source: Gibbs, 1970) Almost all collected groundwater samples from study area in both seasons fall in rock dominance zone followed by evaporative zone suggesting precipitation induced chemical weathering along with dissolution of rock forming minerals. It may be inferred that the chemistry of groundwater in the study area is controlled mainly by the chemical interaction between aquifer rocks and groundwater, and to some extent by processes like evapo-transpiration etc. The process of evaporation might have incorporated some components of sodium and chlorine ions. 4.3 Classification of Ground Water Data has been processed as using Piper Trilinear Diagram and it was observed that majority of the groundwater samples of the study area belong to Ca-Mg-Cl-SO4 or Na-K-Cl-SO4 hydrochemical facies in both pre- and post-monsoon seasons. Such water has permanent hardness and does not deposit residual sodium carbonate in irrigation use and generally creates salinity problems both in irrigation and drinking uses. 4.4 Scatter Plots between Ions The scatter plot of (Ca+Mg) vs TZ+ shows that all the points fall above 1:1 equiline (Fig. 3). The relatively high contribution of (Ca+Mg) to the total cations (TZ+) and high (Ca+Mg)/(Na+K) ratio indicate that carbonate weathering is a major source of dissolved ions in the groundwater of the study area (Fig. 3). The scatter plot of (Na+K) vs TZ+ shows that all the points fall above 1:1 equiline with a low ratio indicating a relatively low contribution of dissolved ions from silicate weathering (Fig. 4). Na+, K+ and dissolved silica in the drainage basin are mainly derived from the weathering of silicate minerals, with clay minerals as by-products. The plot of Na vs Cl indicates most of

HYDRO 2014 International

MANIT Bhopal

Page 183

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

the points lie below the 1:1 equiline reflecting contribution of silicate weathering through the release of Na. The plot of (Ca+Mg) vs HCO3 for most of the samples of the study area indicates an access of alkalinity over Ca+Mg content (Fig. 5). The excess of Ca+Mg over HCO3 in some of the sample of the upper part of basin indicate an extra source of Ca and Mg. This requires that a portion of the (Ca+Mg) has to be balanced by other anions like SO4 and/or Cl. The plot of (Ca+Mg) vs HCO3+SO4 is a major indicator to identify the ion exchange process activated in the study area. If ion exchange is the process, the points shift to right side of the plot due to excess of HCO3+SO4. If reverse ions exchange is the process, points shift left due to excess Ca+Mg. Plot of (Ca+Mg) vs HCO3+SO4 shows that most of the plotted points clusters around the 1:1 equiline and fall in HCO3+SO4 indicating the ion exchange process which may be due to excess bicarbonate (Fig. 5).

Figure 5. Scatter plot of (Ca+Mg) vs HCO3 and (Ca+Mg) vs (HCO3+SO4) (Pre- and Post-monsoon) 5. CONCLUSION Hydrogeochemical studies relevant to the water quality successfully explain the relationship of water chemistry to aquifer lithology. It is concluded that the problem of hardness in groundwater at few location was attributed due to dissolution of rock forming minerals and dominance of carbonate weathering. The ion exchange process is dominating in the study area, which may be due to excess bicarbonate. High concentration of sodium and chloride may be attributed to the process of evaporation and contribution of silicate weathering through the release of Na. REFERENCES

Figure 3. Scatter plot of (Ca+Mg) vs TZ+ and (Ca+Mg) vs (Na+K) (Pre- and Post-monsoon)

i. APHA (Clesceri LS, Greenberg AE, Trussel RR, 1995) Standard Methods for the Examination of Water and Wastewater, APHA, Washington DC. ii. BIS (2012) Indian Standard Drinking Water – Specification (Second Revision). IS:10500:2012, Bureau of Indian Standards, New Delhi. iii. Gibbs Ronald J. (1970) Mechanisms controlling world water chemistry. Science 170(3962): 1088-1090. iv. Kumar Manish, Ramanathan AL., Rao MS, Kumar Bhishm (2006) Identification and evaluation of hydrogeological processes in groundwater environment of Delhi, India. Environ. Geol. 50(7): 1025-1039. v. Master Plan (2002) Master Plan for Artificial Recharge to Groundwater in India Central Ground Water Board, New Delhi, February 2002, p. 115. vi. Obiefuna GI, Orazulike DM (2011) The hydrochemical characteristics and evolution of groundwater in semiarid Yola area, Northeast, Nigeria. Res. J. of Environ. Earth Sci. 3(4): 400-416. vii. Piper AM (1944) A Graphical Procedure in the Geochemical Interpretation of Water Analysis. Trans. Am. Geophysical Union, 25: 914-923. viii. Reddy AG, Kumar KN (2010) Identification of the hydrogeochemical processes in ground water using major ion chemistry: a case study of PennaChitravahi river basins in Southern India. Environmental Monitoring Assessment 170(1-4): 365-382. ix. Srinivasamoorthy K, Vasanthavigar M, Chidambaram S, Anandhan P, Manivannan R, Rajivgandhi R (2012) Hydrochemistry of groundwater from Sarabanga minor basin, Tamil Nadu, India. Proceedings of the International Academy of Ecology and Environmental Sciences. 2(3): 193-203. x. Vijaykumar V, Vasudevan S, Ramkumar T, Shrinivasamoorthy K, Venkatramanan S, Chidambaram S (2010) Hydrogeochemistry in the part of Ariyalur region, Perambalur district, Tamil Nadu, India. J. Applied Geochemists 12(2): 253-260.

Figure 4. Scatter plot of (Na+K) vs TZ+ and Na vs Cl (Pre- and Post-monsoon)

HYDRO 2014 International

MANIT Bhopal

Page 184

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 optimal number of sensors and their locations in the network is desirable.

EVALUATION OF VARIOUS OBJECTIVES IN MULTIOBJECTIVE SENSOR PLACEMENTS IN WATER DISTRIBUTION SYSTEMS S. Rathi1 R. Gupta2 1 Research Scholar, Visvesvaraya National Institute of Technology, Nagpur-440010, India. 2 Professor, Visvesvaraya National Institute of Technology, Nagpur-440010, India. Email: [email protected] ABSTRACT Online monitoring of water quality in distribution network through sensors are of pronounced interest for early detection of contamination event. Since the online monitoring of network is costly affair, the limited numbers of sensors are placed at crucial locations to cover the entire network. Several objectives have been proposed to decide the location of sensors. However, including all of them in deciding location of sensors is a difficult task. Sensor locations are obtained by considering single or few objectives at a time. How far the other objectives not considered during the design are satisfied can be obtained by analysis of sensor network design. This paper aims at explaining evaluation of various objectives for a set of known sensor locations. The objectives evaluated are Demand Coverage, Detection Likelihood, Time of Detection, Population Exposed, Extent of contamination, Volume of contaminated water consumed, Number of failed detection and Risk. The evaluation of above objectives is carried out by considering: (i) hydraulic simulation for dominating demand pattern; and (ii) both hydraulic and water quality simulation over a period of time. EPANET is used for both hydraulic and water quality simulation. The methodology for evaluating various objectives is explained with an illustrative network. The values of various objectives evaluated through water quality simulations provided more realistic and accurate results as compared to that obtained through only hydraulic simulations. However, water quality simulation require more efforts and computation time along with calibrated network to rely on the modeled output. Keywords: contamination, monitoring, distribution system, water quality.

objectives,

water

Lee and Deininger et al. (1992) were perhaps the first to suggest a methodology for location of monitoring stations (MSs) in a WDN using the objective of maximizing the demand coverage (DC) for routine monitoring of water quality. The demand coverage was defined as the percentage of total demand monitored by the set of MSs. The demand coverage does not quantify the impact of contamination events. Kessler et al. (1998) suggested total volume of contaminated water consumed by population as an objective to be restricted to a desired level while selecting location of MSs. Kumar et al. (1999) suggested time of detection as level of service (LOS). In the last decade, the purpose of water quality monitoring has completely changed and early warning system with several other objectives like population exposed to contamination, extent of contamination, detection likelihood, number of failed detections, risk and redundancy of monitoring system etc. were suggested to protect the human from deliberate contamination events. These objectives have been considered independently or jointly by different researchers to propose algorithms for location of monitoring stations/sensors (Chastain 2006; Ostfeld et al. 2004; Watson et al. 2004; Wu and Walski 2006; Berry et al. 2005, 2006; Propato 2006; Ostfeld et al. 2008; Peris and Ostfeld 2008; Aral et al. 2010; Weickgenannt et al. 2010; Krause et al.2008; Dorini et al. 2010, Hart and Murray 2010; Shen and McBean 2011; Kansal et al. 2012). It is observed that various researchers have considered different objectives in the design of sensor network. Some of the objectives like maximizing detection likelihood would probably locate the sensors at the far end of the system or at the downstream network nodes in order to detect more number of contamination events while the objective like minimizing expected time of detection would locate the sensors as close as possible to the source of contamination. Thus, optimizing sensor locations with different objectives will give different sensor locations. Further, different objectives can be evaluated by: (i) considering only hydraulic simulation, in which network is analyzed for flow and velocities for most dominating demand pattern and it is assumed that contamination in any concentration is detected by sensor as it reaches the sensor node; and (ii) water quality simulations to predict the more realistic temporal evaluation of contaminant concentration.

1. INTRODUCTION: Water distribution network (WDN) is an important part of the city infrastructure and its primary aim is to provide safe and adequate drinking water to consumers. A network consists of several pipes connected to each other and other components used to control and measure flows and pressures. Water contamination can occur at any time due to several reasons. The reasons for deterioration of water quality in WDN may be classified as natural, accidental or intentional. In order to detect contamination event at the earliest and to reduce the impact of contamination event, online water quality monitoring in a WDN through sensors is desirable. However, installation of sensors and continuous monitoring is costly affair, therefore selection of

HYDRO 2014 International

The sensor network designed for one or more objectives may required to be checked for its efficacy for other objectives not considered in the design. Further, in GA based designs, few alternative designs are required to compared for fulfillment of different objectives during the design itself. This paper aims at explaining evaluation of various objectives for a set of known sensor locations. The objectives evaluated are Demand Coverage (DC), Detection Likelihood (DL), Time of Detection (TD), Population Exposed (PE), Extent of Contamination (EC), Volume Consumed (VC), Number of Failed Detection (NFD), and Risk.

MANIT Bhopal

Page 185

International Journal of Engineering Research Issue Special3 2. PERFORMANCE DEFINITIONS

OBJECIVES

AND

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 THEIR

Let S be the number of sensors installed in a WDN at different nodes. The total number of nodes are J and number of flow patterns are P. Let us considered that contamination takes place only at the nodes and for any event x, the probability of occurrence during flow pattern p is αxp. The contamination event x at node j will certainly be detected, if j is one of the sensor node s. If j is not a sensor node then contamination event x may be detected at one or more of downstream sensor nodes, if there exists flow paths from j to downstream sensor nodes s. The event x will remain undetected if there are no sensor nodes on the downstream of contamination node. Demand Coverage (DC) - The term DC is defined as the percentage of network demand monitored by a particular and/or set of sensor nodes. If the quality of water is good at any node, it can be presumed good at upstream nodes, if sufficient quantity of water has passed through upstream nodes. An upstream node is assumed as covered by a sensor node if a desired fraction of flow has passed through that node. In general the demand coverage of sensor network would be P

DC 

J

 a p 1 j 1 P J

j

p 1 j 1

t xp for detected events and  for

undetected events. The TD is an important parameter of sensor network. It can be noted that other parameters quantifying the impact of contamination event are dependent on TD. Population Exposed (PE) –It is defined as the number of people exposed to the contaminant before detection by a sensor. In case when only hydraulic simulation is carried out, it is assumed that sensor is capable of detecting any small concentration of contaminant. Thus, population exposed during contamination event x would be the addition of population of all the nodes which gets contaminated in time txp. Therefore, population exposed is given by J

P

PE   xp x1 p 1



jp jcontaminated nodes

(3) Where,  jp

 population associated with node j during pattern p.

In case of water quality simulation, PE is mathematically expressed as

 q j, p

 q

where, DT (x,p) = Travel time

J

p

PE   xp x 1 p 1

j, p

J

C

 jp

xpj jcontaminated node

(1)

(4)

where aj =1 if node j is covered by set of sensor nodes, else aj = 0; q is the nodal demand. It can be noted that DC indicates the property of sensor nodes. Time of Detection (TD) –The detection time for a particular contamination scenario is given by the time elapsed between the start of contamination event and its detection by the first sensor location. Thus, the detection time for any event x at node i which is detected first at node j would be the minimum travel time required by contaminant to reach from node i to node j during flow pattern p and represented as txp. There could be some scenarios in which contaminant may not be detected by any sensor. The Time of Detection for undetected events may be considered as 24 hrs or more (say  ) based on time of simulation (Watson et al. 2004) or when it is indirectly detected in public. The time of detection for the sensor network can been represented by including or excluding the undetected events. In the simplest way, it is the average time necessary for a sensor to detect a substance.

Where, Cxpj - Contamination concentration indicator variable; Cxpj = 1, if concentration is more than threshold concentration, 0 otherwise. It can be observed that during water quality simulation for a contamination event the concentration of contaminant at sensor node may be less than threshold concentration. Herein, the event remains undetected at sensor nodes. However, the population at all the nodes at which concentration is more than the threshold concentration are included in population exposed. Extent of Contamination (EC) – It is defined as the length of pipe contaminated by a contamination event. Length of pipe contaminated during contamination event x would be the addition of contaminated length of all pipes which gets contaminated in time txp under the flow pattern p. It is mathematically defined as J

p

EC  ip x 1 p 1

TD   xp DT ( x, p)

J



jp jcontaminated node

(5)

xJ pP

(2)

HYDRO 2014 International

MANIT Bhopal

Page 186

International Journal of Engineering Research Issue Special3

 jp 

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

L

population exposed. Following these definitions Risk of contamination can be expressed as

e epipes between contaminated nodes

(6) Where, Le – Length of pipe e. The expression similar to Eq. (4) also can be written for the case of water quality simulation for other objectives. Volume Consumed (VC) - The amount of contaminated water consumed by the population before detection by a monitoring station. Mathematically, it can be written a p

J

VC   xp x 1 p 1

J

q

jp jcontaminated node

( DT ( x, p)  t xp ) (7)

Where, qjp - demand at contaminated node j; (DT(x,p) - txp) – Consumption time – The duration before detection by the time water reaches to the sensor node defined as duration before detection over which contaminated node consumes contaminated water injected at a specific node for pattern p. (Detection time at a sensor node minus minimum flow time to the contaminated nodes from the contaminant injected node); DT (x, p) –travel time to a detection point and tijp - Minimum flow time between x and j for p. Number of Failed Detections (NFD) - The proportion of attacks that are undetected by all monitoring stations. Mathematically it is expressed as p

J

NFD   xpbi 0 p x 1 p 1

(8) Where, bi0p = 1, if contamination event is undetected; else 0. Detection likelihood (DL) – It is defined as the probability of detection of a contaminant or it is defined as complement of NFD. For a given sensor network design i.e. by knowing the known number and locations of sensors, J

p

DL   xpbijp x 1 p 1

(9) Where,

bijp =1, if contamination event is detected else 0.

Risk – Weickgenannt et al. (2010) defined Risk as the product of the probability of not detecting the contaminant intrusion and the corresponding consequence in terms of water demand consumed. Berry et al. (2005) defined Risk as fraction of

HYDRO 2014 International

R( S )  (1 

1 J 1 J max  ( S p , K l ))(  min  ( S P , kl ))  x x 1 x x 1 S P S S P S (10)

where, S - set of sensor locations; R(S) - associated contamination risk; Sp - elements in S (Sensor location); x number of contamination scenarios; kl - a contamination scenario index; ξ (Sp, kl) - a binary function with variables Sp (which is a sensor location) and kl (which is a scenario); ξ(Sp,kl) = 1 if the sensor at Sp can detect the scenario kl and 0 otherwise; χ(Sp, kl) - Volume of water that is contaminated prior to network shutdown following the intrusion detection at a specific sensor. Monitoring Stations response delay (MSRD) – It is defined as possible monitoring Stations response delay in revealing a hazard intrusion. Herein, we assume that MSRD = 0 means it is assumed that monitoring stations are capable of providing real time detection data. 3.0 Common Assumptions 3.1 In hydraulic as well as water quality simulation: Following assumptions have been made to evaluate the objective values. 1. One contamination event is considered at any time. The contaminant intrusion is considered at the nodal point only. 2. The probabilities of contamination at all the nodes are assumed to be equal. 3. Sensor locations are considered only at the nodal points in the network. 4. Sensors are assumed to be perfect in the sense that above a specified concentration, the sensor is 100 % reliable and below that concentration the sensor always fails to detect the contaminant and they are accurate i.e no false positives and no false negatives. It is also assumed that the alarm is raised by the sensors at detection time and without considering any response delay. 3.2 Additional assumptions during hydraulic simulation only: 1. Hydraulic analysis is carried out for only one demand pattern, i.e. peak demands. 2. The contaminant travels in the pipeline with the velocity of water. Further, it is assumed that contamination is detected by sensor as it reaches the sensor node how-so-ever small is the concentration, thus ignoring the effect of dilution on contaminant concentrations. 3. Sensor protects downstream populations. A population is considered exposed if it could be reach by a flow path from the attack point without passing a sensor. 4. The contaminated water moves in the pipeline and travel in different pipes connected at the junctions. All points on

MANIT Bhopal

Page 187

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

downstream of attack point are assumed to be contaminated until contamination is reached at one of the sensors, i.e. up to time when the contaminated water will reach at least one sensor.

Figure 1. Example water distribution system Table 1. Pipe details peak demand hour results

3.3 Additional assumptions during water quality simulation: 1.

2.

3.

4. 5.

The pollutants are assumed to be conservative type where contaminant does not react with water and its dynamics is determined by water flows, dilution and mixing from the intrusion site to consumers nodes. Water quality simulation is performed to predict contaminant concentration with time resulting from a particular contamination scenario and various objectives are evaluated by developing a pollution matrix with mass rates of 5000 mg/min with duration of injection is of 2 hours. It is assumed that contamination event occur at peak demand time where more number of people consuming water at that time. The hazard concentration threshold is taken as 0.3 mg/l. The contaminant reaches in the network at different nodes with different concentrations from the contaminated source node and effect will continues till contaminated water reaches to one sensor in a set with concentration greater than the threshold concentration. The effect of contamination will be used for determination of objectives up to that time.

Table 2. Demand pattern

4. ILLUSTRATIVE EXAMPLE A single source WDN (Kessler et al. 1998) as shown in Figure.1 is considered for the evaluation of various objectives for a set of known sensor locations. The network has 12 pipes and 8 consumer locations- the number of consumers at each location (given in parenthesis) is shown in Figure 1, a source, a storage tank and a pump. The average nodal demands are given in Lps. The pipe diameters are given in Table 1. The demand multiplier (ratio of actual demand to average demand) for different periods in 24 hours are given in Table 2. Total length of the pipe in the network is 19364 meters and total consumers of the network are 7600. The pipe 10 is 3209 m long, while all other pipes are 1609 m in length. Hazen-Williams coefficient for all pipes is 100. Other details can be obtained from the Kessler et al. (1998).

5. EVALUATION OF VARIOUS OBJECTIVES Various performance objectives are evaluated for two sets of known sensor locations ─ (1) Sensors at nodes 32 and 23; and (2) Sensors at nodes 32, 23 and 31. Further, objective function values are obtained by considering: (1) only hydraulic simulation; and (2) both hydraulic and water quality simulation. 5.1 Evaluation considering hydraulic analysis: Case 1 : Sensors at nodes 32 and 23. 5.1.1 Evaluation of DC: Lee and Deininger (1992) suggested a methodology for maximizing DC which is based on development of water fraction matrix and coverage matrix based on chosen coverage criteria. In order to determine the upstream nodes covered by a monitoring station, the coverage criteria is used and defined as the minimum percentage of total water received at a monitoring node that should have passed through an upstream node to be able to consider it “covered”. The lower the coverage criteria, the more the demand coverage of monitoring nodes increases.

HYDRO 2014 International

MANIT Bhopal

Page 188

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Let us consider two coverage criteria of 60 % and 30 %. With 60% coverage criteria only those upstream nodes through which 60% of the total flow has passed will be included, while with 30 % coverage criteria all the upstream nodes through which 30% total quantity of water passed will get included. Thus, demand coverage of monitoring station is based on coverage criteria, which brings subjectivity in the design as its value has to be fixed by the designer according to his own experience. Therefore, here to evaluate the demand coverage objective we used a simple methodology (Rathi et al. 2014) which avoids subjectivity owing to coverage criteria as it considers all the nodes on the shortest path from source to sensor node as covered based on the assumption that major flows are along shortest path. The set of sensor at nodes 32, 23 are given. First, identify the shortest path for nodes 32 and 23. Shortest path for 32 is 10-1112-22-32 and for 23 is 10-11-12-22-23. Now calculate DC by adding the demands of all nodes on shortest path. Therefore, DC of 32 = 0+7.57+7.57+10.09+7.57 = 32.8 Lps and similarly DC for 23 = 5.046 (without adding demand of previously covered nodes twice). Therefore, total DC of 32 and 23 = 37.846 Lps (68.18% shown in Table 5 and 6) out of a total of 55.508 Lps. It can be noted that DC is the attribute of sensor network that is not affected by number or probabilities of contamination events. 5.1.2 Evaluation of TD: To evaluate TD for the sensor network, the detection time for individual contamination events are required. A general travel time matrix as shown in Table 3 is developed which can be used for evaluation of other objectives also as discussed later. The element in travel time matrix is the shortest travel time (in Hrs.) from the contaminated node to the other nodes.

5.1.3 Evaluation of PE, EC, DL and NFD: To determine PE, EC, DL and NFD for sensors at nodes 32, 23 we make use of general travel time matrix shown in Table 3. Consider contamination at node 10. The event is first detected at sensor node 32 in 6.98 hrs. Therefore, all the nodes to which travel time is less than 6.98 hrs. will get contaminated. Considering the row 1 in Table 3, all the nodes except the two sensor nodes have travel time less than 6.98 hrs and therefore they are included in the list of contaminated nodes shown in Table 4. The PE and EC are also given in Table 4. Average values of PE and EC with equal probability of contamination at all nodes are obtained as 2328 and 4847.336 meters. Whether a contamination scenario is detected or not is shown in Col. 5. Thus, total 9 out of total 10 number of events are detected by sensors at nodes 32 and 23. Therefore, detection likelihood is 90 % and NFD as complement of DL is 10 %. Table 4. Calculation of objectives for Sensor location at nodes 32, 23

Now, to evaluate TD for sensors at nodes 32, 23 various contamination scenarios are considered at different nodes. For example, from the shortest travel time matrix it is observed that for contamination event at node 10, event is be detected by both the sensors at nodes 32 and 23 in time 6.98 hrs and 8.53 hrs, respectively. Therefore, the detection time for this event is 6.98 hrs. The contamination at all nodes except that at node 2 is detected at least by one of the sensor node. The event at node 2 is undetected by sensors at nodes 32, 23. The average time of detection is calculated by considering only the detected events with equal probability of occurrence and found as 5.94 hours. The TD is also evaluated by considering both detected and undetected events in which detection time for the undetected events is the time when such events are indirectly noticed in public. Herein, detection time for undetected events is taken as twice of the maximum simulation duration. The average TD is obtained as 10.14 hrs. Table 3. Calculation of objectives for Sensor location at nodes 32, 23

HYDRO 2014 International

5.1.4 Evaluation of VC: Contaminated volume consumed is the actual consumption up to the event is detected. It is calculated by aggregating the product of the nodal demands at contaminated node by the time difference between time of detection at sensor node and the time required by contaminant to reach the contaminated demand node from the point of intrusion. Thus, in this example if contamination takes place at node 10 and first detected at sensor

MANIT Bhopal

Page 189

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

node 32, the contaminated nodes are 10,11,12, 2, 21,22,31, and 13. These nodes are arranged in ascending order of travel times as nodes 10, 11, 12, 2, 21, 22, 31, and 13. Minimum travel time from node 10 to 32 is 6.988 hours. The contaminated water consumed by the time water just reaches node 32 will be given by: {[0×(6.98-0) + 7.57×(6.98-1.3) + 7.57×(6.98-1.88) + 0×(6.98-1.93) + 7.57×(6.98-2.1) + 10.09×(6.98-3.8) + 5.046×(6.98-4.03) + 5.046× (6.98-5.84)} × 3600 = 203640 litres. Multiplier outside the brackets are the demands (L/s) at nodes 10, 11, 12, 2, 21, 22, 31, and 13. In this way, consumption is calculated by assuming the contamination event at each node and assuming equal probability of attack to all nodes gives a volume consumption of 220128 litres.

with hydraulic simulation over 24 hr period. The obtained values of the objectives are shown in Tables 6 and 7. Table 6. Evaluation of various objectives for a set of known sensor locations using Water quality analysis

5.1.5 Evaluation of Risk: Risk values for PE (%) are evaluated considering fraction of population exposed. The average PE is 2328 (Table 4) and the total population is 7600. Therefore, PE (in %) is 30.63 % (=2328x100/7600). Risk for VC (%) is 32.37 %, (i.e. 220128 Litres of contaminated water consumed out of total volume consumption of 679937.4 Litres).

Table 7. Evaluation of Risk objective

Case 2. Sensors at nodes 32, 23 and 31. The evaluation of various objective are also carried out similarly for case 2. The obtained values are shown in Table 5 along with those obtained for case 1 for easy comparison. Table 5. Evaluation of various objectives for a set of known sensor locations using hydraulic analysis

Performance objectives evaluated through water quality simulations provides more realistic results as they are obtained by considering variation in demands over time and concentration of pollutant. However, water quality simulation require more efforts and computation time. From Table 5 and 6 it can be observed that the difference between the various objective values under hydraulic simulation and water quality simulation are not much. 5. CONCLUSIONS

It can be observed from table 5 that with one additional sensor at node 31, coverage increases and other parameters such as PE, EC, TD and VC decreases. The DL and NFD remains the same as the contamination event at node 2 still remains undetected with addition of sensor at node 31. Risk values for PE (%) and VC (%) are 24.33 % and 11.59 %, respectively. 5.2 Evaluation considering water quality simulation along with hydraulic simulation: Performance objectives are evaluated for both the cases of known sensor locations using water quality simulations along

HYDRO 2014 International

This paper aims at explaining evaluation of various performance objectives of a WDN equipped with a set of sensors at known locations. The objectives evaluated are Demand Coverage (DC), Detection likelihood (DL), time of detection (TD), population exposed (PE), extent of contamination (EC), volume consumed (VC), Number of failed detection (NFD), and Risk. It is observed that DC is an attribute of sensor network that is not dependent on number of contamination events and their locations. The PE, EC, and VC are the attributes governed by TD. With the increase in average TD, these parameters decreases. The evaluation of above objectives is carried out by considering only hydraulic simulation and also with water quality simulation. The values of objectives evaluated after performing water quality simulations provides more realistic and accurate results as compared to considering simply hydraulic simulations. However, water quality simulation requires more efforts and computation time along with calibrated network to rely on the modeled output.

MANIT Bhopal

Page 190

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Such evaluation are also required in sensor network design using GA with multiple objectives where several alternative designs are compared based on performance objectives and their fitness is quantified. REFERENCES: i. Aral MM, Guan J, Maslia LM (2010) Optimal Design of Sensor Placement in Water Distribution Networks. Journal of Water Resources Planning and Management 136 (1): 5-18. ii. Berry JW, Fleischer L, Hart WE, Phillips CA, Watson J-P (2005) Sensor placement in municipal water networks. Journal of water Resources Planning and Management 131 (3): 237-243. iii. Berry J, Hart WE, Phillips CA, Uber JG, Watson, J-P (2006) Sensor placement in municipal water networks with temporal integer programming models. Journal of Water Resources Planning and Management 132 (4): 218224. iv. Chastain JR Jr. (2006) Methodology for locating monitoring stations to detect contamination in potable water distribution systems. Journal of infrastructure system 12 (4): 252–259. v. Dorini G, Jonkergouw P, Kapelan Z, Savic, D (2010) SLOTS: Effective algorithm for sensor placement in water distribution systems. Journal of Water Resources Planning and Management 136 (6), 620-628. vi. Hart WE, Murray R (2010) Review of sensor placement strategies for contamination warning systems in drinking water distribution systems. Journal of Water Resources Planning and Management 136 (6): 611-619. vii. Lee BH, Deininger RA (1992) Optimal locations of monitoring stations in water distribution system. Journal of Environmental Engineering 118 (1): 4-16. viii. Kansal ML, Dorji T, Chandniha SK and Tyagi A (2012) Identification of optimal monitoring locations to detect accidental contaminations. Proc. World Water and Environmental Resources Congress 2012, ASCE, Albuquerque, NM, 758-776. ix. Kessler A, Ostfeld A and Sinai G (1998) Detecting accidental contaminations in municipal water networks. Journal of Water Resources Planning and Management 124 (4): 192–198. x. Krause A, Leskovec J, Guestrin C, VanBriesen J, Faloutsos C (2008) Efficient sensor placement optimization for securing large water distribution networks. Journal of Water Resources Planning and Management 134 (6): 516– 526. xi. Kumar A, Kansal ML, Arora G (1999) Detecting accidental contaminations in municipal water networks. Journal of Water Resources Planning and Management 125 (5): 308–310. xii. Ostfeld A, Salomons E (2004) Optimal layout of Early Warning Detection Stations for Water Distribution Systems Security. Journal of Water Resources Planning and Management 130 (5): 377-385. xiii. Ostfeld A, et al (2008) The battle of the water sensor networks (BWSN): A design challenge for engineers and algorithms Journal of Water Resources Planning and Management 134 (6): 555-568. xiv. Preis A, Ostfeld A (2008) Multiobjective contaminant sensor network design for water distribution systems. Journal of Water Resources Planning and Management 134 (4): 366-377. xv. Propato M (2006) Contamination Warning in Water Networks: General Mixed-Integer Linear Models for Sensor Location Design. Journal of Water Resources Planning and Management 132 (4): 225-233. xvi. Rathi S, Gupta R (2014) Location of sampling stations for water quality monitoring in water distribution networks. Journal of Environmental Science and Engineering (in press). xvii. Shen H, McBean E (2011) Pareto optimality for sensor placements in a water distribution system. Journal of Water Resources Planning and Management 137 (3): 243-248. xviii. Watson J-P, Greenberg HJ, Hart WE (2004) A multiple objective analysis of sensor placement optimization in water networks. Proc. World Water and Environmental Resources Congress ASCE, Reston, VA. xix. Weickgenannt M, Kapelan Z, Blokker M, Savic DA (2010) Risk based sensor placement for contaminant detection in water distribution systems. Journal of Water Resources Planning and Management 136 (6): 629-636. xx. Wu ZY, Walski T (2006) Multi objective optimization of sensor placement in water distribution systems Proc. 8th Annual Water Distribution Systems Analysis Symp., ASCE, Reston, VA.

HYDRO 2014 International

Water Quality Assessment Of Dal Lake, Kashmir, J&K. Shabina Masoodi Associate Professor, SSM College of Engineering and Technology, Parihaspora, Pattan, Kashmir, J&K. 193121 Email: [email protected] ABSTRACT: Dal Lake is one of the prized lakes of world; it is part of India‟s beautiful national heritage and has been the centre of Kashmir‟s civilization. It has suffered a lot due to the impact of pollution and the present paper is an attempt to assess its water quality. The water quality of the Dal Lake has been seriously altered over a period of time because of human interventions which include agricultural activities within and on the periphery of the lake, urbanization and mushrooming of hotels besides waste discharge into it. The lake thus has turned Eutrophic and is under great stress. Since the lake water is also been harvested for public distribution (potable purposes) this problem has gained significance keeping in view the public health. The zones at the periphery and close to the effluent discharge depict temporal variations. Around fifty percent of the observed maximum specific conductivity, dissolved oxygen, nitrate-nitrogen, ammonical–nitrogen and total phosphorus have been noticed in the spring season. Summer season has twenty five percent of such observations and the remaining twenty five percent are distributed in autumn and winter seasons. This may be possibly due to the start of activities in the catchment, mixing or re-suspension. A comparison of values over a period of time shows that the Dal Lake has passed through several stages of eutrophic evolution. Extensive data establishes far reaching changes in the physico-chemical environment. Dal Lake receives large quantities of nitrogen and phosphorus from incoming sewage drains from non-point sources like seepages and diffused runoff. Of the total phosphorus and inorganic nitrogen inflow from all sources, the quantity contributed by the drains works out to be thirty five percent. Similarly a sizeable quantity of total phosphates and nitrogen are added to the lake from non point sources. Various engineering interventions like catchment management, dredging, de-weeding, sewerage treatment plants etc have been taken but their efficacy is under assessment since the results are not very positive for the health of the Lake. Keywords: Water quality, Human interventions, Waste discharge, Eutrophication, Engineering intervention. 1. INTRODUCTION: The valley of Kashmir is bordered to the South and West by Pir Panjal ranges and to the North and East by the Himalayan foot hills. Numerous freshwater lakes are found within the state of Jammu and Kashmir which covers an altitudinal range of 600m and 500m. These lakes have been originated as a result of earthquakes, damping of glaciers, weathering, denudation, floods and meandering of alluvial deposits. DAL LAKE is one such prized moderate altitude lake located within the geographical coordinates of 340 6 N 740 45‟ East of Srinagar

MANIT Bhopal

Page 191

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

spreading over an area of 25 sq Km (1895 AD) and reduced to merely 11.5 Sq Km (2009). It is at an altitude of 1587msl.

Dal Lake has been the centre of Kashmir civilization and is one of the most beautiful spots of tourist attraction. This shallow-post glacial freshwater body is bounded on Southwest and West by Srinagar city, and its remaining sides are surrounded by gentle terraced slopes at the base of precipitous mountains. Dal Lake is unique because of: 

Floating Gardens with the lake.





Habitation within the lake.

HYDRO 2014 International

Biodiversity.

The Dal Lake lies in the flood plains of river Jhelum whose broad meanders have cut swampy low lands out of the Karewa terraces. The inflow Telbal nallah channel enters the lake from the North bringing water from the high altitude Mansar Lake. During its downward journey the inflow stream collects large quantities of silt from the denuded catchment and deposits it in the lake. Numbers of ephemeral water channels, surface drains enter the lake from the human settlements discharging large quantities of wastes. An estimated load of 12.30 x10 6m3 of liquid waste with 18.17 tons and 25 tons of phosphorus and inorganic nitrogen is enriching the lake annually. Within Lake Basin itself a number of freshwater springs (mostly choked at present) act as permanent source of water to the lake. Towards the South-west

MANIT Bhopal

Page 192

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

side an outflow channel Tsuant-Kul discharges lake water into Jhelum river at Gawkadal. The outflow is regulated by a sluice gate to prevent the entry of Jhelum water into the lake during the floods. On the Eastern side the Nigeen basin of the lake is connected by nallah or a channel dug by Afghan Governor Amir Khan up to Khushalsar lake, which in turn connects with Anchar lake. This channel also serves an additional outflow channel, particularly during floods. Influx of waste and silt and excessive weed growth in the Lake has affected the quality of its water and the present study it aimed at assessing it. 1. MARGINAL DREDGING AND ITS IMPACT ON WATER QUALITY The main purpose of dredging is to increase the area of open water to improve water circulation, navigational routes, to create more attractive mosaic and to define margins. As part of the Dal Lake conservation proposals under taken National Lake Conservation Program, NLCP as per the proposals of IRAM consultants, marginal dredging along the shore lines of Dal near Nishat basin and Habak basin was done using suction cutter dredgers. Similar peripheral dredging was also undertaken in the Nigeen basin of the lake. Another consultant AHEC (Roorkee) also had favored marginal dredging but with the remarks that there should be pre-implementation evaluation of lake settings, proper equipment and disposal sites and its effect on lake ecology and long term productivity should be continuously evaluated. AHEC identified 38 channels within the lake which were clogged or reduced in width and proposed to excavate them. Similarly 57 fresh water springs were identified around the lake whose water got polluted during the intervening period they reached the lake. The post dredging changes and a comparative limnology of Dal Lake reported a decrease in Nitrate-nitrogen and total phosphorous content after dredging while increase in Ammonical and ortho-phosphates. The plankton diversity did not show any significant change in dredged and un-dredged sites. Table1. Comparative changes in Physio-Chemical parameters at dredged and Un-dredged sites in Dal Lake Kashmir.

Fresh water lakes usually are abound of aquatic vegetation and constitute one of the important components of biodiversity. It is also an established fact that the aquatic plants (Macrophytes) are the bio-indicators of pollution and have an important role in removal of nutrients from the lake sediment and help in pollution abatement. At the same time excessive growth of aquatic weeds impede boat transport hinder irrigation and increase sediment deposition besides effect the lake aesthetics. Thus the most sound and reasonable management approach is to control their growth. In Dal Lake the lake dweller have been doing de-weeding through traditional pole method where in they would whirl the wooden pole in such a skilled way that they would extract the weeds and use them for preparation of vegetation gardens or as bio fertilizers. They would also take out the bottom mud and use it for vegetable garden preparations. But when the weed infestation in the lake basins increased beyond proportion the authorities concerned had to deploy mechanical harvesters which also became an issue of controversy among the lake scientists. According to the consultants the de-weeding in Dal should be selective. AHEC, Roorkee (2000) states; based on the information available, it is recommended that de-weeding has to be selective and limited to certain areas only especially areas which are useful to repeated harvesting. According to the consultants de-weeding should be limited to backwaters, areas where exotic water ferns, water lilies abound and areas where water skilling or swimming takes place. They further suggest that in areas selected for de-weeding it is very important that only 40% - 50% weed is removed and the rest is left untouched. Efforts should be directed towards harvesting undesirable plant species such as Ceratophyllum demersum, Nymphaea Stellata Salvinia natans and Hydrocharis morus ranae. According Trisal (1977, 1987) Typha Agustata and Phragmites communis were the chief occupants of littoral zone of Dal and Nigeen Lake and extended all along the Eastern part of the Southern side of the Hazratbal basin. In the Nishat basin and Nigeen basin the emergents are scattered towards the shorelines and formed large stands in the arms of the lake basin. According to the author rooted floating leaf macrophytes (Aquatic plants) occupy 29.2% of total area of the lake free floating aquatic ones were distributed throughout the lake area in sheltered pockets. Submerged aquatic species due to their aggressive capacity cover the maximum area of 57.6% in all the basins of the lake. Zutshi and Tickoo (1990) while studying the impacts of mechanical de-weeding in Dal Lake recorded the reduction in Seechi transparency of water and attributed it to the suspension of sediments due to impact of harvesters. The authors however noted the increase in dissolved oxygen content by 23 % in the surface waters and by 36% in bottom waters. They further recorded significant temporal change in nitrate nitrogen but little horizontal and vertical difference as a result of de-weeding.

2. DE-WEEDING AND ITS IMPACT ON WATER QUALITY.

HYDRO 2014 International

Kundangar (1996) while studying the impact of waste waters on the vegetation pattern of Dal Lake reported surprising changes in

MANIT Bhopal

Page 193

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

the Dal Lake basins and reverted the increase in abundance of some eutrophic species. He attributed the luxuriant entering the lake besides the enrichment of sediments through leaching of fertilizers in the immediate agricultural lands surrounding the lake. Kundangar (2003) while studying the impact of de-weeding in Dal Lake estimated liquid wastes carrying 18.7 tons of phosphorus and 25 tons of inorganic nitrogen into the lake which results in increase in fertility of lake waters and resulting in accelerated weed growth. They also added that major part of phosphorus and nitrates coupled with other nutrients get locked up in the roots and rhizomes of the aquatic weeds. Thus these aquatic weeds play significant role in keeping the water crustily more or less in stable condition. But these aquatic weeds on decaying during autumn-winter go on enriching the sediment with nutrients and play an active role in re-growth of aquatic weeds in the next spring. The authors recorded a slight shift in pH of water in Nehru Park and Nigeen basin (Table 2). After de-weeding the authors concluded that with overall 55% of manual aquatic weed removal in various basins of Dal Lake, there was decrease in specific conductivity, iron, and phosphorus. The authors also recorded that the full scale de-weeding (8-100%) enhance the release of nutrients from the enriched sediment and result in serious and hazardous algal blooms in a Lake ecosystem particularly in Dal Lake. The authors stressed on long term studies to establish a set of standards both for water quality and biodiversity changes as a result de-weeding practices in the Lake ecosystem. Table 2a: Pre and post de-weeding changes in water quality by Mechanical de-weeding in Dal Lake Kashmir (after Kundangar 2003)

Table 2b: Pre and Post de-weeding changes in water quality by Manual de-weeding in Dal Lake (after Kundangar 2003).

3. SEWERAGE AND SEWAGE TREATMENT AND ITS IMPACT ON WATER QUALITY Sewerage and sewage treatment constitutes a major component of the Dal lake conservation plan for preventing the pollution of the lake. The Dal Lake receives water from fifteen major drains besides inflow from the Telbal and Bota-kadal Nallahs. The drains bring in 40 mld of sewage and join the lake at locations identified. Two alternative plans for sewage treatment were envisaged. One proposed conceptualized a centralized sewage treatment where in all the waste will be collected by sewers (gravity mains) and trunk sewers with 15 immediate pumping stations (IPS) and a main pumping station, at Brarinambal. This unit of about 41 mld will treat the sewage through an activated sludge process and release treated waste effluent through Brainambal cut into Jhelum. This system through theoretically very sound has some inherent weakness, such as power dependence (in pumping and treatment) large size trunk sewers and large distance of transport. The power scenario in Srinagar town is dismal and utilizing it for pumping sewage as against domestic requirements seems as far cry. Moreover, failure of system or any component will put the entire machinery out of gear. To obviate these difficulties a decentralized system is preferred and has been proposed, which could do away with a large amount of pumping and trunk sewers. The bulk of the sewage will flow by gravity and pumping will be restored to only when there is no alternative. The STP‟s will be provided at least at six sites in Dal Lake and two or three at Nigeen. The treated effluent of three STP‟s will flow out of the lake and the rest after tertiary treatment will be discharged into the lake (around 40%). The total sewage generated in all three zones worked out to be 36.7 mld in the year 2017. A total of nine IPS, one in zone one, six (under construction) in zone 2 and two (existing) in zone 3 are proposed. The decentralization has resulted in a significant reduction in the cost of sewers and of operation and maintenance. Sewerage treatment.

HYDRO 2014 International

MANIT Bhopal

Page 194

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

There are numerous options available to treat the waste water. These include dispersed and attached growth aerobic systems. (Activated Sludge process, Aerated Lagoon, Oxidation ditch, Trickling filter and Rotating Biological Discs), suspended and attached growth anaerobic systems (up flow anaerobic Sludge Blanket, expanded bed, fluidized bed) and pond processes. In recent past the artificial wetland compartment technology has also gained momentum in the developed countries where in aquatic plant species are exploited for waste water treatment. According to the AHEC consultancy the FAB (Fluidized Aerobic Bed and Bio-filters) technology was considered and recommended for Dal Lake. FAB technology consists of screening, grit removal, biological treatment (bioreactors), tertiary treatment of clarifloculator (with alum), centrifuge and chlorination. The six units were proposed of which five have been made operational. Habak 3.2mld STP 1 (a) STP 1 (b)

REC

7.5mld

STP 1 (c)

Nallah Amir Khan

5.4mld

STP 2

BrariNambal

9.5mld

STP 3 (a)

Hotel Heemal

6.6mld

STP 3 (b)

Laam

4.5mld

Total

36.7mld

The treated effluent of STP 1 (c) and 3 (a) is discharged in channels leaving Dal Lake via Amir Khan. Dalgate exit and Brari-Nambal cut). Thus only 40% of the total of 36.7 mld finds its way into the Dal Lake. Controversy regarding FAB Technology Kundangar (2003) while maintaining the FAB based sewage treatment plant, of one of the hotels in the immediate vicinity of Dal Lake recorded reversed trend i.e, instead of expected decrease in nutrients, a significant increase was observed in the treated sewage. According to the author 90-98% increase was recorded in ortho-phosphate and total phosphorus respectively while 32% increase was recorded in nitrate-nitrogen during winter months. In their studies during April 2008 (Table 3a) regarding the functioning of FAB based STP reported 44% increase in nitratenitrogen content of the treated sewage indicating the malfunctioning of the STP‟s installed. Table 3(a) Efficiency of nutrient removal through FAB – STP (April 2008)

Water quality of the Dal Lake has been seriously altered over a period of time because of human interventions which include agricultural activities within and on the periphery of the lake, urbanization and mushrooming of hotels besides waste discharge. The lake thus has turned Eutrophic and is under great stress. Since the lake water at Nishat and Nigeen is also harvested for public distribution (Potable purposes), the quality of water has therefore assumed a great significance keeping in view the public health. The zones at the periphery and close to the effluent discharge depict temporal variations. Around 50% of the observed maximum specific conductivity, dissolved oxygen, nitratenitrogen, ammonical–nitrogen, PO4 and total phosphorus have been noticed in the spring season. Summer season has 25% of such observations and the remaining 25% are distribution in autumn and winter seasons. This may possibly be due to the start of activities in the catchment, mixing or re-suspension (LAWDA, 2000 report). A comparison of values over a period of time (Table 4) shows that the Dal Lake has passed through several stages of trophic evolution. Extensive data establishes far reaching changes in the physico-chemical environment. Dal Lake receives large quantities of nitrogen and phosphorus from incoming sewage drains, Telbal Nallah and that of Bhota Kadal as well as from non-point sources like seepages and diffused runoff. The lake being peculiar in having human habitations within the lake either in hamlets (Islands), boats, house boats etc of the total phosphorus inflow 156.62 tons from all sources, the quantity contributed by the drains works out to be 56.36 tons. In the case of inorganic nitrogen (NO3 and NH3-N) these figures are 241.18 tons and 77.60 tons with a flow of 11.70 million cum/yr. Similarly 4.5 tons of total phosphates and 18.14 tons of nitrogen are added to the lake from non point sources. Table 4a: Water Quality changes in Hazratbal Basin of Dal Lake over a period of time

. Table 4b: Water Quality changes in Nishat Basin of Dal Lake over a period of time

CONCLUSION-WATER QUALITY ASSESSMENT

HYDRO 2014 International

MANIT Bhopal

Page 195

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Table 4c: Water Quality changes in Nehru Park basin of Dal Lake over a period of time

xii. 2003, De-weeding practices in Dal Lake & impact assessment.Kundangar. xiii. 2004, Thirty years of Ecological Research on Dal Lake, Kundangar. xiv. 2004, Groundwater quality of downtown Srinagar, Adnan, Neelofer, Nuzhat and Kundangar. 2005. xv. 2004, Bacterial Dynamics of Dal Lake, a Himalayan temperate fresh water lake, Adnan & Kundangar. xvi. 2005. Ecology of peripheral springs of Dal lake, Kashmir Adnan & Kundangar. xvii. 2009, Monitoring of Dal-Nigeen Lakes & other water bodies (J&K PCB). xviii. 2009. Three decades of Dal Lake, Adnan & Kundangar. xix. 2010, Sanative role of macrophytes in Aquatic Ecosystems, Adnan. xx. 2011, Water quality changes in Nigeen Lake, Shariqa Maryam. xxi. 2011. Ecological studies & uses of valued aquatic plants in Kashmir wet lands, Adnan, Afsha & Kundangar. xxii. 2012, Impact of mechanical de-weeding on Macrozoobenthic community in Dal Lake, Basharat, Rajini, AR Yousuf &Ashwani.

Spatial Water Quality Analysis Of Nagalamadike Watershed Of Pavagada Taluk, Tumkur District Karanataka Using Geo Informatic Tools

Table 4d: Water Quality changes in Nigeen Lake over a period of time

Nandeesha1, Ravindranath.C2, T.Gangadaraiah3, and S.G Swamy4 1 Professor, Civil Engineering Department, Siddaganga Institute of Technology, Karnataka, India 2 Research Scholar, Civil Engineering Department, Siddaganga Institute of Technology, Karnataka, India 3 Professor Civil Engineering Department, Siddaganga Institute of Technology, Karnataka, India 4 Fellow KSCST Bangalore Karnataka, India [email protected] [email protected] [email protected] [email protected] ABSTRACT

REFERENCES: i. 1978, Pollution of Dal Lake, Enex. ii. 1990. Impact of mechanical de-weeding on Dal lake eco system, Zutshi & Tickoo. iii. 1993. Effects of weed cutting on species, composition and abundance of plankton population, Zutshi & Tickoo. iv. 1996, Impact of waste water on the vegetational pattern of Dal Lake, Kundangar. v. 1996. Aeration of Dal lake (an interim report) HRL. vi. 1997, Dal Lake conservation & rehabilitation. (J&K LAWDA). vii. 1998, Technical report on Dal Lake (J&K LAWDA). viii. 1999, Technical report on Dal Lake (J&K LAWDA). ix. 2000, Technical report on Dal Lake (J&K LAWDA). x. 2000. DPR conservation and management plan for Dal –Nigeen lake-AHEC Roorkee. xi. 2001, Post dredging changes & comparative limnology of Dal Lake, Kundangar.

HYDRO 2014 International

Ground water samples from 25 locations of the watershed bounded by latitude N 1405‟to 14015‟ and longitude E 77015‟ to77025‟ were collected. The samples collected are distributed over Precambrian rocks such as closepet granite and gneissic terrines. Red sandy and loamy soil covers the major area of the watershed. The samples were analyzed for pH, Electrical Conductivity (EC), Total Dissolved Solids (TDS), Total hardness, Fluorides, Iron, Nitrite, Sodium and Chloride. The results of all the samples analyzed as per standard method and compared with the BIS and WHO, drinking water standards out of 25 samples 23 samples of Fluoride showed more than permissible limit, and 15 samples of nitrate showed more than permissible limit, and 20 samples of sodium shoved more than permissible limit the permissible range of Fe, pH, EC, Cl, TDS, TH, are in permissible limit. The most of the samples are lie within the permissible limits. Arc View Ver.9.2 software and ERDAS Ver. 9.1 was used to get watershed map, land use/land cover map, litho logical map and Iso contour maps of major parameter are generated and overlayed on the thematic map to study the spatial variation of the parameters in the watershed and causes for the pollution from various sources.

MANIT Bhopal

Page 196

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

KEY WORDS: Spatial variation, permissible limit, Arc View

Ver.9.2, ERDAS Ver. 9.1

1. Introduction Due to the ever increasing demand for potable and irrigation water and inadequacy of available surface water the importance of ground water is increasing everyday. In the natural Hydrological cycle the rainwater gives us sample of good quality of the water but as the Urbanization and Industrialization the natural cycle of the water is disturbed resulting in less rainfall or runoff of the good quality of the water into sea as there is no open space left in cities to allow rain water to get absorbed in earth due to concretization. Drinking water is a basic requirement for life and a determinant of standard of living. Around 22 per cent of households in India lack of access to safe drinking water sources, like tap, hand pump and tube well (Census 2001). Hence, significant efforts are being made by the central and state governments for increasing the coverage of households with adequate and safe drinking water supply, along with sanitation services, which coincide with the Millennium Development Goals. In the recent past, several parts of our country have been experiencing drought conditions very often due to vagaries of the nature, mainly monsoon. In Karnataka, Tumkur district comes under this category. Depending on the ground water resources available even at the times of severe drought conditions, when major part of this surface water resources are exhausted, it has been conceived to develop ground water base irrigation system in certain part of the district. Nearly 2/3rd of the state receives less than 750mm of rainfall. Many parts of the south and north interior Karnataka depends on ground water for its domestic and agricultural needs.

Fig 1 Location map of study area Sample collection points and location of study area

Study area The Nagalamadike gram panchayat is located in eastern part of the pavagada taluk 10.9 km from the main pavagada town and 99 km from Tumkur town. The gram panchayat has a total area of 74.6 sq. km. and a population of 1500. The area consists of 14 micro watersheds that constitute a mini watershed. This is situated in the Pennar river basin. The sources of water in this area include bore well, hand pump, water tanks etc. The study area is reported to be facing a lot of problems regarding the quality of water. The residing people are facing acute problems of fluorosis which is due to deficient of excessive quality of water. Thus an effort has been made to survey the study area and analyze the quality of water by sampling and presenting the results in an interesting and attractive way so that the need for reforms is highlighted. The technology involved in this project plays a major role in the analysis. The use of sophisticated instruments such as the Water Analyzer 371, Colorimeter DDR 2010, flame photometer is used for the analysis and AAS (Atomic Absorption Spectrophotometer) have made the tests very simpler and quicker. Moreover the use of G.P.S. devices such as GARMIN 12 channel made it much easier to locate a particular water source so that any person can identify the point. Arc GIS Ver. 9.2 is used for representation of results.

HYDRO 2014 International

Fig 2: Sample Location map study area. Details of the latitude and longitude points of Nagalamadike watershed, sample collection of 25 points shown in table no1.

Table 1. Details of sample locations

MANIT Bhopal

latitute and langitute of

Page 197

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Methodology For studying the chemical quality of groundwater 25 groundwater samples were collected and the sample locations are shown in fig 2. Water samples collected from bore wells in use and samples collected in the one litre pre-washed polythene bottles, were analysed in the chemical labaroatoty of the Department of Civil Engineering, SIT, Tumkur and the results are given in table no 3.

Chemical Analysis of Ground water: Groundwater is the main source of water that meets the agricultural, industrial and household requirements. Population growth, socioeconomic development, technological and climate changes has increased the demand for potable water manifolds in the past few years (Alcamo et al. 2007).One of the internationally accepted human rights is the access to safe drinking water which is the basic need for human health and development (WHO 2001). The general health and life expectancy of the people is reported to be adversely affected due to lack of the availability of clean drinking water in many developing countries of the world (Nash and McCall 1995). In irrigation, the poor water quality not only affects the crop yield but also affects the physical conditions of the soil (Ayers and West cot 1994). Since the dependence on groundwater has increased tremendously in India due to vagaries of monsoon and scarcity of surface water in recent years, therefore groundwater quality and surface water needs to be monitored and managed. The water sample is analysed by using BIS 1983 permissible limits which is shown in table no 2.

HYDRO 2014 International

Fig.3. Water Analysis Methodology chart. The above methodology is used to find the chemical contamination of water samples of 25 location in Nagalamadike watershed of the Pavagoda taluk of Tumkur district karanataka state india.

MANIT Bhopal

Page 198

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Soil map: In the pavagada taluk the soil is consisting of fine grained and loamy soil. The soil map is shown in fig 4

fig 6 and also the fluoride concentration is as shown in same figure. Here the spatial distribution map of fluoride from the Iso contour map of the study are shows the southwest zone having a rich content of fluoride and the central ,norhtenzone consisting limited quantity the lithology of these shows the granite belt and northern shows the PGC belt.

Fig: 4 soil map of pavagada taluk Lithology map of study area: In the Lithology map the study area consists of Granite and PGC.is shown in fig 5

Fig:6 Lu/Lc map and overlay of Iso concentration map of Fluoride. Table 2 : Permissible limits (BIS-1983) of potable water in the study area

PARAMETER

HIGHEST DESIRABLE LIMIT (in ppm)

MAXIMUM PERMISSIBLE LIMIT (in ppm)

FLUORIDE

0.6-1.2

1.5

NITRATE

45

NO RELAXATION

TOTAL HARDNESS CHLORIDE

300

600

250

1000

pH

6.5-8.5

8.5-9.5

IRON

0.3

1.0

SODIUM

0-60

100

Fig: 5 Lithology map of the study area

Land use and Land cover map of study area: In the study area the five groups of land use and land cover is as shown in

HYDRO 2014 International

MANIT Bhopal

Page 199

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Results and Discussions: Chemical concentration analysis of water samples is collected from the study area. In the study area 25 water samples are water collected from various locations and analyzed in chemical lab the following details is shown in table 3.

permissible limit as shown in table 2, and fluoride isoconcentration map is shown in fig 8. Fluoride is more in south west region and remaining zone is less.

Table 3: Chemical analysis of water samples.

Fig 8. Spatial distribution map of fluoride (Iso contour map of fluoride) Iso concentration map of Nitrate: Spatial distribution of Nitrate and Iso contour map is prepared using of Arc GIS tool. Nitrate content is more in most of the water samples out of 25 samples the Nitrate present in 15 samples is more than the permissible limit as shown in table 2, Nitrate concentration is shown in figure no 9. In North east and south east ,the nitrate contamination is more due to more application of artificial manure (NPK) in agriculture.

Fig:7 Chemical concentration of water sample Iso Concentration map of fluoride: Spatial distribution of fluoride and Iso contour maps is prepared using of Arc GIS tool. Fluoride content is more in most of the water samples out of 25 samples the fluoride present in 23 samples is more than the

HYDRO 2014 International

Fig: 9 Spatial distribution map of Nitrate (Iso contour map of Nitrate)

MANIT Bhopal

Page 200

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Iso concentration map of Iron: Spatial distribution of the iron is prepared using Arc GIS and concentration of Iron is ranges from 0 to 0.4 is within the permissible limit the Iron concentration map is shown in fig 10.the iron concentration is distributed almost equal in all places.

Iso concentration map of Total Hardness: Spatial distribution of Iso counter map is developed by using the Arc GIS tool is shown in fig 12 .TDS is more in the south central and north central part of the study area and remaining area is less.

Fig 12: Spatial Distribution map of Total Hardness (Iso contour map of Total Hardness) Fig 10: Spatial Distribution map of Iron (Iso contour map of Iron) Iso concentration map of pH scale: Spatial distribution map pH is developed by using Arc GIS tool and pH is ranges from 6.28 to 8.26. All the samples in the study area falls within the permissible range.

Fig: 11 Spatial distribution map of pH (Iso contour map of pH)

HYDRO 2014 International

Iso concentration map of Electrical Conductivity: Spatial distribution and Iso counter map is developed by using the Arc GIS tool and, Electrical Conductivity shown in fig 13. The electrical conductivity is more in Northern part of the study area where as remaining part the electrical conductivity is less.

Fig 13: Spatial distribution map of Electrical Conductivity (Iso contour map of Electrical Conductivity) 3.6: Iso concentration map of Cl: Spatial distribution and Iso counter map is developed by using Arc GIS tool and concentration level shown in fig 14. The chloride is more in the

MANIT Bhopal

Page 201

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

north central part and south part of the study area and remaining area is less.

dissolved materials. In natural waters, salts are chemical compounds comprised of anions such as carbonates, chlorides, sulphates, and nitrates (primarily in ground water), and cat ions such as potassium (K), magnesium (Mg), calcium (Ca), and sodium (Na). In ambient conditions, these compounds are present in proportions that create a balanced solution. If there are additional inputs of dissolved solids to the system, the balance is altered and detrimental effects may be seen. Inputs include both natural and anthropogenic source.

Fig 14: Spatial distribution map of chloride (Iso contour map of chloride) Iso concentration map of Sodium: Spatial distribution of Sodium and Iso contour maps is prepared using of Arc GIS tool. Sodium content is more in most of the water samples out of 25 samples the Sodium content in 20 samples is more than the permissible limit as shown in table 2.the sodium content is more in north central part and remaining of the study area is less.

Fig 16: Spatial distribution map of Total Dissolved Solids (Iso contour map of TDS) CONCULSIONS

Fig 15: Spatial Distribution map of sodium (Iso contour map of sodium)

Iso concentration map of Total Dissolved Solids: Spatial distributionof TDS and Iso counter map is developed by using Arc GIS tool. TDS concentration is shown in fig 16.The total dissolved solids (TDS) in water consist of inorganic salts and

HYDRO 2014 International

It is observed that the study area is basically composed of hard and compact lithologies and to add to the conclusions the distribution of rainfall in the state with time and space is highly variable. Moreover, limited surface water resources and non uniform rainfall as increased the dependence on the ground water resources. This mounting pressure has resulted in excess utilization of the ground water resource. Thus, the ground water resources have reached critical stages. Geographic Information Systems are rapidly developing as primary technologies for the investigation of large scale patterns and processes. The use of Arc GIS software not only improves the analytical capabilities for water resource management but also the ability to communicate work results and research findings to the decision makers and general public. The advantage of GIS software‟s has made it possible to update, modify or revalidate data at any location. This tool will help the public and decision makers to understand, assess and actively participate in issues pertaining to water bodies‟ pH, Electrical Conductivity, Iron content, Total Hardness and Chloride content in all the samples is within the maximum permissible range. Fluoride content, nitrate content, and sodium is more in most of the water samples. Samples exceeding Fluoride limit- 23/25Samples exceeding nitrate limit15/25Samples exceeding sodium limit- 20/25as per permissible table.

MANIT Bhopal

Page 202

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Suggestions: To prevent the entry of nitrate in the groundwater sources, the use of chemical fertilizers in agriculture should be minimized and the use of natural manure should be encouraged. The people of the area should make the awareness programmers about water quality management and rain water harvesting, artificial groundwater recharge, etc. Frequent quality checks of water , Soil analysis, Rock analysis shall be made for betterment of water quality analysis. Scope of further study: Sampling is to be more representative because of the vast area covered and more samples are needed to be taken to give more accurate results. The samples have to be analyzed for bacteriological parameters, heavy metals such as lead, and also radioactive metals to know more about the affects of water for various purposes .Spatial distribution maps have to be overlaid on geomorphologic information. By overlaying the map of the study area over the drainage map, soil map and lithology map the drainage pattern; soil of the area can be assessed respectively for future generation.

xvi. Kazi T.G., Arain M.B., Jamali M.K., Jalbani N., Afridi H.I., Sarfraz R.A., Baig J.A., and Shah A.Q., (2009), Assessment of water quality of polluted lake using multivariate statistical techniques: A case study, Ecotox. Environmental Safety, 72(20), pp 301-309 xvii. S.F. Mulgundmath (1974) , Dept of Mines and Geology, Bangalore. A report on ―GROUND WATER RESOURCES OF TUMKUR TALUK, TUMKUR DISTRICT‖. xviii. Statistical abstract. (2008). State statistical abstract. Chandigarh, India: Government of Haryana Publication. xix. Todd, D. K., & Mays, L. W. (2005). Groundwater hydrology (3rd ed.). New York: Wiley. xx. U.S. Salinity Laboratory (USSL) (1954). Diagnosis and improvement of saline and alkali soils; USDA Handbook No. 60. pp. 160 Richards LA (ed) (1954) xxi. WHO (2001). Water health and human rights, world water day http://www.worldwaterday.org/wwday/2001/thematic/ hmnrights.html xxii. WHO (2008). Guidelines for drinking water quality incorporating Ist and 2nd addenda Vol.1 Recommendations, (3rd edit) http://www.who.int/water_sanitation_health/dwq/ gdwq3rev/en. xxiii. Wilcox, L. V. (1948). The quality of water for irrigation use, USDA Technical Bulletin No 962, pp. 1–40

Water Pollution In Ganga River Susmita Saha Asst. Professor Sagar Institute of Research & Technology Email: [email protected]

References: i. Abbasi, S.A., (2002), Water quality indices, state of the art report, National Institute of Hydrology, scientific ii. Contribution no. INCOH/SAR-25/2002, Roorkee: INCOH, pp 73. iii. Ahmed, S., David, K.S. and Gerald, S., (2004), Environmental assessment: An innovation index for evaluation water iv. quality in streams, Environment Management., 34 pp 406-414. v. Bajpai, A., Vyas, A., Verma, N. and Mishra, D.D. (2009). Effect of idol immersion on water quality of twin Lakes of vi. Bhopal with special reference to heavy metals. Poll. Res., 28(3):433-438. vii. Bhavana, A., Shrivastava, V., Tiwari, C.R. and Jain, P. (2009). Heavyvmetal contamination and its potential risk with viii. special reference to Narmada River at Nimar region of M.P. (India). Res. J. of Chem. &Env. 13 (4), 23-27. ix. Chaudhary, B. S., Kumar, M., Roy, A. K., & Ruhal, D. S. (1996). Applications of RS and GIS in groundwater investigations in Sohna Block, Gurgaon District, Haryana, India. International Archives of Photogrammetry and Remote Sensing, 31(B-6), 18–23. Eaton, F. M. (1950). Significance of carbonates in irrigation water. Soil Science, 69, 123–133. doi:10.1097/00010694-195002000-00004. x. ―DISTRICT PROFILE AND RESOURCES ATLAS OF TUMKUR DISTRICT‖. – N.R.D.M.S Centre, Z.P, Tumkur ―Ground Water quality evaluation of Tumkur town- By Ajay K.C., Pawan kumar P.M. ,Sanjeev Saurabh. Year 2006-07 xi. ―Ground water quality assessment using GIS‖:-by Channabasabanna A. Year 2005-06 xii. Goyal, S. K., Chaudhary, B. S., Singh O., Sethi, G. K., & Thakur, P. K. (2010) GIS Based Spatial Distribution Mapping and Suitability Evaluation of Groundwater Quality for Domestic and Agricultural Purpose in Kaithal Distirct, Haryana State, India. Environmental Earth Science. In press, doi:101007/s12665-010-0472-z. xiii. Indian Standard Specification for Drinking Water (1983), IS-105001983, Indian Standards Institution, New Delhi, xiv. Jain, C. K., & Sharma, M. K. (2000). Regression analysis of groundwater quality of Sagar District, Madhya Pradesh. Indian Journal of Environmental Health, 42(4), 159–168. xv. Lloyd, J. W., & Heathcote, J. A. (1985). Natural inorganic hydrochemistry in relation to groundwater: An introduction. Oxford, New York: Clarendon Press, Oxford University Press.

HYDRO 2014 International

Abstract : There is a universal reverence to water in almost all of the major religions of the world. Most religious beliefs involve some ceremonial use of "holy" water. The purity of such water, the belief in its known historical and unknown mythological origins, and the inaccessibility of remote sources, elevate its importance even further. In India, the water of the river Ganga is treated with such reverence. The river Ganga occupies a unique position in the cultural ethos of India. Legend says that the river has descended from Heaven on earth as a result of the long and arduous prayers of King Bhagirathi for the salvation of his deceased ancestors. From times immemorial, the Ganga has been India's river of faith, devotion and worship. Millions of Hindus accept its water as sacred. Even today, people carry treasured Ganga water all over India and abroad because it is "holy" water and known for its "curative" properties. However, the river is not just a legend, it is also a life-support system for the people of India. It is important because the densely populated Ganga basin is inhabited by 37 percent of India's population. The entire Ganga basin system effectively drains eight states of India. About 47 per cent of the total irrigated area in India is located in the Ganga basin alone. It has been a major source of navigation and communication since ancient times. The IndoGangetic plain has witnessed the blossoming of India's great creative talent. Keywords: Pollution in Ganga, Pollution free by Ganga Action Plan, Treatment of water of Ganga. 1. INTRODUCTION

MANIT Bhopal

Page 203

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The Ganga rises on the sourthern slopes of the Himalayan ranges(fig 1.1) from the Gangotri glacier at 4,000 m above mean sea level. It flows swiftly for 250 km in the mountains, descending steeply to an elevation of 288 m above means sea level. In the Himalayan region the Bhagirathi is joined by the tributaries Alaknanda and Mandakini to form the Ganga. After entering the plains at Haridwar, it winds its way to the Bay of Bengal, covering 2,500 km through the provinces of Uttar Pradesh, Bihar and West Bengal ,. In the plains it is joined by Ramganga, Yamuna, Sai, Gomti, Ghaghara, Sone, Gandak, Kosi and Damodar along with many other smaller rivers. The Ganga river carries the highest silt load of any river in the world and the deposition of this material in the delta region results in the largest river delta in the world (400 km from north to south and 320 km from east to west). The rich mangrove forests of the Gangetic delta contain very rare and valuable species of plants and animals and are unparalleled among many forest ecosystems.

In the recent past, due to rapid progress in communications and commerce, there has been a swift increase in the urban areas along the river Ganga. As a result the river is no longer only a source of water but it is also a channel, receiving and transporting urban population lives in the towns of the Ganga basin. Out of the 2,300 towns in the country, 692 are located in this basin, and of these, 100 are located along the river bank itself. The belief the Ganga river is “holy” has not, however, prevented over-use, abuse and pollution of the river. All the towns along its length contribute to the pollution load. It has been assessed that more than 80 per cent of the total pollution load (in terms of organic pollution expressed as biochemical oxygen demand (BOD)) arises from domestic sources, i.e. from the settlement along the river course. Due to over-abstraction of water for irrigation in the upper regions of the river, the dry weather flow has been reduced to a trickle. Rampant deforestation in the last few decades, resulting in topsoil erosion in the catchment area, has increased silt deposits which, in turn, raise the river bed and lead to devastating floods in the rainy season and stagnant flow in the dry season. Along the main river course there are 25 towns with a population of more than 100,000 and about another 23 towns with populations above 50,000. In addition there are 50 smaller towns with population above 20,000. There are also about 100 identified polluting areas. Fifty-five of these industrial units have complied with the regulations and installed effluent treatment plants (ETPs) and legal proceedings are in progress for the remaining units. The natural assimilative capacity of the river is severely stressed. The principal sources of pollution of the Ganga river can be characterized as follows: 

 

2. MATERIAL AND METHODS



The purity of the water depends on the velocity and the dilution capacity of the river. A large part of the flow of the Ganga is abstracted for irrigation just as it enters the plains at Haridwar. From there it flows as a trickle for a few hundred kilometers until Allahabad, from where it is recharged by its tributaries. The Ganga receives over 60 per cent of its discharge from its tributaries. The contribution of most of the tributaries to the pollution load is small, except from the Gomti, Damodar and Yamuna rivers, for which separate action programmes have already started under Phase II of “The National Rivers Conservation Plan”.

 

Domestic and industrial wastes. It has been estimated that about 1.4 x 106m3d-1 of domestic wastewater and 0.26 x 106 m3 d-1 of industrial sewage are going into the river. Solid garbage thrown directly into the river. Non-point sources of pollution from agricultural run-off containing residues of harmful pesticides and fertilizers. Animal carcasses and half-burned and unburned human corpses thrown into the river. Defecation on the banks by the low-income people. Mass bathing and ritualistic practices.

Causes of pollution in Ganga It provides water to drinking purpose and irrigation in agriculture about 40% of India‟s population in 11 states. After 27 years and Rs. 1000 crore expenditure on Ganga river, it has a critical situation. In modern times, it is known for being much polluted, 30 polluted nalas flows in Ganga river from Varanasi city within seven kilometers. 2.1 Human Waste

HYDRO 2014 International

MANIT Bhopal

Page 204

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The river flows through 29 cities in which cities population living above ten lakh. A large proportion damp the solid and liquid wastes in Ganga river like domestic usage (bathing, laundry and public defecation), Sewage wastes, unburnt dead bodies through in Ganga river. Patna and Varanasi cities are more responsible to water pollution of Ganga. 2.2 Industrial Waste Countless industries lies on the bank of the Ganga river from Uttrakhand to West Bengal like chemical plants, textile mills, paper mills, fertilizer plants and hospitals waste. These industries are 20% responsible to water pollution and run off solid waste and liquid waste in the Ganga river. It is very dangerous to water quality, their chemical properties and riverine life. 2.3 Religious factor Festivals are very important and heartiest to every person of India. Owing festival seasons a lot of peoples come to Ganga Snans to clean themselves. After death of the people dump their asthia in Ganga river it is a tradition of India because they think that Ganga gives mukti from the human world. Khumbha Mela is a very big festival of the world and billion peoples come "Ganga Snans at Allahabad, Hardwar in India. They throw some materials like food, waste or leaves in the Ganges for spiritualistic reasons. 2.4 Riverine Life The Ganga river pollution increased day by day and from this pollution marine life have been going to lost in near future and this polluted water disturb the ecosystem of the river. And irrigation and Hydroelectric dams give struggle to life in their life cycle. 2.5 Bio Life Some dams are constructed along the Ganges basin. Dams are collected a huge volume of water and this is hazard for wild life which are moving in Ganga river. The Kotli Bhel dam at Devprayag will submerge about 1200 hectors of forest. In India wildlife has been warning that the wild animals will find it difficult to cope with the changed situation. 2.6 Human beings An analysis of the Ganges water in 2006 showed significant associations between water-borne/enteric disease occurence and the use of the river for bathing, laundry, washing, eating, cleaning utensils, and brushing teeth. Exposure factors such as washing clothes, bathing and lack of sewerage, toilet at residence, children defecating outdoors, poor sanitation, low income and low education level also showed significant associations with enteric disease outcome. Water in the Ganges

HYDRO 2014 International

has been correlated to contracting dysentery, cholera, hepatitis, as well as severe diarrhea which continue to be one of the leading causes of death of children in India. 2.7 The Ganga Action Plan The Ganga Action Plan or GAP was a program launched by Rajiv Gandhi in April 1986 in order to reduce the pollution load on the river. Under GAP I, pollution abatement schemes were taken up in 25 Class-I towns in three States of U.P., Bihar and West Bengal. GAP I was declared complete on 31.03.2000 with an expenditure of Rs. 452 crore. As GAP I addressed only a part of the pollution load of Ganga, GAP II was launched in stages between 1993 and 1996, 59 towns along the main stem of river Ganga in five States of Uttarakhand, U.P., Jharkhand, Bihar and West Bengal are covered under the Plan and included the following tributaries of the Ganges, Yamuna, Gomti, Damodar and Mahananda. According to Hindustan Newspaper, January 11, 2013, the Prime Minister has been monitoring the availability of adequate water from Tehri Dam in river Ganga at Allahabad during the Kumbh Mela. Directions have been given to control the pollution load flowing in river Yamuna during the Kumbh Mela period. Tehri Hydro Development Corporation (THDCIL) has agreed to release 250 cumecs water from 21st December 2012 to 20th February 2013 to 28th February 2013 in view of demand of water for Allahabad „Kumbh Snans‟. Instructions have also been given by PMO that Delhi Jal Board should ensure that the performance of the 72 MGD STP (Sewage Treatment Plant) at Keshavpur renovated / commissioned recently is stabilized so that it functions optimally and the effluent meets the norms. The Delhi Government has been asked to ensure that the performance of the STPs and CETPs (Common Effluent Treatment Plants) is optimized to meet the effluent quality norms. At Sangam, Allahabd, the Biochemical Oxygen Demand (BOD) of Yamuna and Ganga is generally less than 6 mg/ltr but the main issue is of the color of effluents discharged by the pulp and paper industries into the river Ram Ganga and Kali (both tributaries of Ganga). Monitoring of water quality in river Ram Ganga and river Kali and their tributaries is being initiated on a daily basis by the State Boards of Uttrakhand and Uttar Pradesh with the coordination of CPCB. Action will be taken against the industries for violating the norms. Spiritual dip in holy Ganga at Kumbh is not clean. The pollution level in the sacred river has risen since Kumbh started at the historical city of Allahabad on January 14, 2013 and the water is not fit for bathing purposes, latest evaluation by country‟s pollution watchdog the Central Pollution. The level of the Biochemical Oxygen Demand (BOD) level – used to measure of the level of organic pollution in the water – had increased to 7.4 milligram per litre at the main bathing place, known as Sangam, since the Kumbh started.

MANIT Bhopal

Page 205

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

A day before the Kumbh, the pollution level was 4.4 milligram per litre slightly more than the national standard for bathing quality of water of 3 miligram per litre. “Higher the BOD level worse it is for one‟s skin,” said a CPCB expert. High exposure to dirty water can result in skin rashness and allergies. The official reason for the sudden rise of contaminants in the river was sudden increase in flow of human waste because of increased bathing during Kumbh. Around 10 million people have already visited the Kumbh and the UP government has employed around 10,000 sweepers to keep the city clean. Off the record officials admit that their drive to check sewage from industries in Ganga upstream of Allahabad has not worked as dirty sewage was still flowing into the river. The Board has been asked by the environment ministry to monitor the pollution level in Ganga under its National Ganga Basin River Authority and conduct periodic check on pollution industries along the river bank. But, the dirt in the river is not a deterrent for people to take a dip at Allahabad. Hindus believe that the Ganga water has ability to clean and purify itself, a claim not scientifically proven. And, this belief has drive millions to the world biggest Hindu congregation and another 15 million are expected to visit in the 55-day long festival to end on March 10.

state governments, under the supervision of the GPD. The GPD was to remain in place until the GAP was completed. The plan was formally launched on 14 June 1986. The main thrust was to intercept and divert the wastes from urban settlements away from the river. Treatment and economical use of waste, as a means of assisting resource recovery, were made an integral part of the plan. The GAP was only the first step in river water quality management. Its mandate was limited to quick and effective, but sustainable, interventions to contain the damage. The studies carried out by the CPCB in 1981-82 revealed that pollution of the Ganga was increasing but had not assumed serious proportions, except at certain main towns on the river such as industrial Kanpur and Calcutta on the Hoogly, together with a few other towns. These locations were identified and designated as the “hot-spots” where urgent interventions were warranted. The causative factors responsible for these situations were targeted for swift and effective control measures. This strategy was adopted for urgent implementation during the first phase of the plan under which only 25 towns identified on the main river were to be included. The studies has revealed that: 

3. RESULT AND ANALYSIS



3.1 Scientific awareness



There are 14 major river basins in India with natural waters that are being used for human and developmental activities. These activities contribute significantly to the pollution loads of these river basins. Of these river basins the Ganga sustains the largest population. The Central Pollution Control Board (CPCB), which is India‟s national body for monitoring environmental pollution, undertook a comprehensive scientific survey in 1981-82 in order to classify river waters according to their designated best uses. This report was the first systematic document that formed the basis of the Ganga Action Plan (GAP). It detailed land-use patterns, domestic and industrial pollution loads, fertilizer and pesticide use, hydrological aspects and river classifications. This inventory of pollution was used by the Department of Environment in 1984 when formulating a policy document. Realizing the need for urgent intervention the Central Ganga Authority (CGA) was set up in 1985 under the chairmanship of the Prime Minister. The Ganga Project Directorate (GPD) was established in June 1985 as a national body operating within the National Ministry of Environment and Forest. The GPD was intended to serve as the secretariat to the CGA and also as the Apex Nodal Agency for implementation. It was set up to co-ordinate the different ministries involved and to administer funds for this 100 per cent centrally-sponsored plan. The programme was perceived as a once-off investment providing demonstrable effects on river water quality. The execution of the works and the subsequent operation and management (O&M) were the responsibility of the

HYDRO 2014 International



75 per cent of the pollution loads was from untreated municipal sewage. 88 per cent of the municipal sewage was from the 25 Class I towns on the main river. Only a few of these cities had sewage treatment facilities (these were very inadequate and were often not functional) All the industries accounted for only 25 per cent of the total pollution (in some areas, such as Calcutta and Kanpur, the industrial waste was very toxic and hard to treat).

3.2 Attainable objectives The board aim of the GAP was to reduce pollution and to clean the river and to restore water quality at least to Class B (i.e. bathing quality: 3 mg l-1 BOD and 5 mg l-1 dissolved oxygen). This was considered as a feasible objective and because a unique and distinguishing feature of the Ganga was its widespread use for ritualistic mass bathing. The other environmental benefits envisaged were improvements in, for example, fisheries, aquatic flora and fauna, aesthetic quality, health issues and levels of contamination. The multi-pronged objectives were to improve the water quality, as an immediate short-term measure, by controlling municipal and industrial wastes. The long-term objectives were to improve the environmental conditions along the river by suitably reducing all the polluting influences at source. These included not only the creation of waste treatment facilities but also invoking remedial legislation to control such non-point sources as agricultural run-off containing residues of fertilizers and pesticides, which are harmful for the aquatic flora and fauna. Prior to the creation of the GAP, the responsibilities for pollution

MANIT Bhopal

Page 206

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

of the river were not clearly demarcated between the various government agencies. The pollutants reaching the Ganga from most point sources did not mix well in the river, due to the sluggish water currents, and as a result such pollution often lingered along the embankments where people bathed and took water for domestic use. 3.3 The strategy The GAP had a multi-pronged strategy to improve the river water quality. It was fully financed by the Central Government, with the assets created by the Central Government to be used and maintained by the industrial wastes. All possible point and non-point sources of pollution were identified. The control of point sources of urban municipal wastes for the 25 Class I towns on the main river was initiated from the 100 per cent centrallyinvested project funds. The control of urban non-point sources was also tackled by direct interventions from project funds. The control of non-point source agricultural run-off was undertaken in a phased manner by the Ministry of Agriculture, principally by reducing use of fertilizer and pesticides. The control of point sources of industrial wastes was done by applying the polluterpays-principle. A total of 261 sub-projects were sought for implementation in 25 Class I (population above 100,000) river front towns. This would eventually involve a financial outlay of Rs 4,680 million (Indian Rupees), equivalent to about US$ 156 million. More than 95 per cent of the programme has been completed and the remaining sub-projects quality, although noticeable, is hotly debated in the media by the certain non-governmental organizations (NGOs). The success of the programme can be gauged by the fact that Phase II of the plan, covering some of the tributaries, has already been launched by the Government. In addition, the earlier action plan has now evolved further to cover all the other major national river-basins in India, including a few lakes, and is known as the “National Rivers Conservation Plan”. 3.4 Prevention of pollution of river Ganga Training cum Awareness programme on Saltless Preservation of Hides / skins was organized by CPCB at Lucknow and Kanpur, which was attended by representatives from slaughter houses, tannery & allied units and officers of UPPCB. The programme was oriented towards the ongoing efforts pursuing basin-wise approach for reduction of dissolved solids in wastewater from leather processing industries in particular by invoking salt less preservation of hides / skins. CPCB has initiated a Techno-Economic Feasibility for setting up of Common Recovery Plant & Common Effluent Treatment Plant for Pulp & Paper Industries identified clusters at Muzaffar nagar, Moradabad and Merut. CPCB also made a reconnaissance survey from Gomukh to Uluberia (West Bengal) for identified the point source and its impact on River. This reconnaissance survey is conducted in association with Shri Rajinder Singh, Member, NGRBA.

HYDRO 2014 International

CPCB issued direction to UPPCB and Uttrakhand PCB in the matter of Prevention and Control of Pollution from agro based Pulp & Paper Sector Mills. As a result 31 industries have been issued directions in U.P., 25 digester sealed at Uttrakhand, 8 industries were directed and 4 were stop chemical pulping. CPCB conducted monitoring of 26 industrial units in the strength of river Ganga between Kannauj to Varanasi in the month of September 2010. Of these 7 were found closed during inspection, 2 were complying to the prescribed discharge norms, 9 were requiring minor improvements, 4 have been issued directions (under section 5 of Environment Protection Act 1986) for closure, 3 have been issued directions for corrective measures (under section 5 of Environment Protection Act 1986) and I have been issued Show Cause notice for closure (under section 5 of Environment Protection Act 1986). 3.5 Integrated improvements of urban environments Apart from the above, the GAP also covered very wide and diverse activities, such as conservation of aquatic species (gangetic dolphin), protection of natural habitats (scavenger turtles) and creating riverine sanctuaries (fisheries). It also included components for landscaping river frontage (35 schemes), building stepped terraces on the sloped river banks for ritualistic mass-bathing (128 locations), improving sanitation along the river frontage (2,760 complexes), development of public facilities, improved approach roads and lighting on the river frontage. 3.6 Applied research The Action Plan stressed the importance of applied research projects and many universities and reputable organizations were supported with grants for projects carrying out studies and observations which would have a direct bearing on the Action Plan. Some of the prominent subjects were PC-based software modeling, sewage-fed pisciculture, conservation of fish in upper river reaches, bioconservation in Bihar, monitoring of pesticides, using treated sewage for irrigation, and rehabilitation of turtles. Some of the ongoing research projects include land application of untreated sewage for tree plantations, aquaculture for sewage treatment, disinfection of treated sewage by Gamma radiation. Expert advise is constantly sought by involving regional universities in project formulation and as consultants to the implementing agencies to keep them in touch with the latest technologies. Eight research projects have been completed and 17 are ongoing. All the presently available research results are being consolidated for easy access by creation of a data base by the Indian National Scientific Documentation Centre (INSDOC). 3.7 Public participation The pollution of the river, although classified as environmental, was the direct outcome of a deeper social problem emerging from long-term public indifference, diffidence and apathy, and a

MANIT Bhopal

Page 207

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

lack of public awareness, education and social values, and above all from poverty. In recognition of the necessity of the involvement of the people for the sustainability and success of the Action Plan, due importance was given to generating awareness through intensive publicity campaigns using the press and electronic media, audio visual approaches, leaflets and hoardings, as well as organizing public programmes for spreading the message effectively. In spite of full financial support from the project, and in spite of a heavy involvement of about 39 well known NGOs to organizing these activities, the programme had only limited public impact and even received some criticism. Other similar awarenessgenerating programmes involving school children from many schools in the project towns were received with greater enthusiasm. These efforts to induce a change in social behaviour are meandering sluggishly like the Ganga itself.

The Action Plan started as a “cleanliness drive” and continues in the same noble spirit with the same zeal and enthusiasm on other major rivers and freshwater bodies. Its effectiveness could however be enhanced if these efforts could be integrated and well-accepted within the long-term objectives and master plans of the cities, which are consultancy under preparation without adequate attention to the disposal of wastes. More information on polluted groundwater resources in the respective river basins will prove useful, because the existing levels of depletion and contamination of groundwater resources, which are already overexploited and fairly contaminated, will increase the dependency in the future on the rivers, as the only economical source of drinking water. This aspect has not been seriously considered in any long-term planning. 4.2 Recommendations  

3.8 River water quality monitoring Right from its inception in 1986, the GAP started a very comprehensive water quality monitoring programme by obtaining data from 27 monitoring stations. Most of these river water quality monitoring stations already existed under other programmes and only required strengthening. Technical help was also received for a small part of this programme from the Overseas Development Agency (ODA) of the UK in the form of some automatic water quality monitoring stations, the associated modeling software, training and some hardware. The monitoring programme is being run on a permanent basis using the infrastructure of other agencies such as the CPCB and the Central Water Commission (CWC) to monitor data from 16 stations. Some research institutions like the Industrial Toxicology Research Centre (ITRC) are also included for specialized monitoring of toxic substances. The success of the programme is noticeable through this record of the water quality over the years, considered in proportion to the number of improvement schemes commissioned. To evaluate the results of this programme an independent study of water quality has also been awarded to separate universities for different regional stretches of the river. 4. CONCLUSION 4.1 The future Apart from the visible improvement in the water quality, the awareness generated by the project is an indicator of its success. It has resulted in the expansion of the programme over the entire Ganga basin to cover the other polluted tributaries. The GAP has further evolved to cover all the polluted stretches of the major national rivers, and including a few lakes. Considering the huge costs involved the central and state governments have agreed in principle to each share half of the costs of the projects under the “National Rivers Action Plan”. The state governments are also required to organize funds for sustainable O&M perpetuity. Initially, the plan was fully sponsored by the central government.

HYDRO 2014 International

    

A white paper on the status of Ganga and GAP. Self purifying power of the river should be ascertained. People should be warned that the river is not worth aachman and bathing. Army should be involved in cleaning the river in Cantonment stretches. A Ganga Restoration Fund should be constituted. Additional resources should be generated by charging the Ganga usesrs, through sand mining etc. Campaign like clean Ganga, sare Ganga should be introduced.

References i. Cleaning-up the Ganges: A cost-Benefit Analysis of the Ganga Action Plan by A Markandya and M.N. Murty. ii. On the Banks of the Ganga: When Wastewater Meets a Sacred River by Kelly D Alley. iii. The River Goddess (Tales of Heaven & Earth S.) By Vijay Singh (Author) and Pierre De Hugo (Illustrator) iv. Tare, Dr. Vinod. ―Pulp and Paper Industries in Ganga River Basin: Achieving Zero Liquid Discharge‖. Report Code: 14_GBP_IIT_EQP_S& R_04_Ver 1_Dec 2011. v. K. Jaiswal, Rakesh. ―Ganga Action Plan-A critical analysis‖. (May, 2007). vi. A report ―Status Paper on River Ganga‖ State of Environment and Water Quality, National River Conservation Directorate Ministry of Environment and Forests Government of India, Alternate Hydro Energy Centre Indian Institute of Technology Roorkee, (August, 2009). vii. Singhania, Neha. ―Cleaning of the Ganga‖. Journal Geological Society of India, Vol. 78, pp.124-130, August 2011. viii. Das, Subhajyoti. ―Cleaning of the Ganga‖. Journal Geological Society of India, Vol 78, pp. 124-130, August 2011. ix. A report of Central Pollution Control Board, Ministry of Environment and Forest ―Ganga Water Quality Trend‖, Monitoring of Indian Aquatic Resources Series, Dec., 2009. x. A report of Water Resources Planning Commission, ―Report on Utilization of Funds and Assets Created through Ganga Action Plan in States under GAP‖, May, 2009. xi. http://en.wikipedia.org/wiki/pollution_of_the-Ganges xii. Report for improvement in GAP, March 1999 MOE&F. xiii. Ganga : A Journey Down the Ganges River by Julian Crandall Hollick, Published October 15th 2007 by Island Press. xiv. Jaya Ganga : In Search of the River Goddess By Vijay Singh.

MANIT Bhopal

Page 208

International Journal of Engineering Research Issue Special3 xv.

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The Ganges By Raghubir Singh.

Flood Frequency Analysis Using A Novel Mathematical Approach Bidroha Basu1V.V. Srinivas2 Research Scholar, Department of Civil Engineering, Indian Institute of Science, Bangalore - 560012, India. 2 Associate Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore-560 012, India 1

ABSTRACT Regional frequency analysis (RFA) is often considered to estimate design flood quantile at target site(s) in river basins when there is paucity of data. The analysis involves use of flood related information from a homogeneous region (group of sites that are hydrologically similar to the target site) to arrive at the estimate. Conventionally RFA is based on Index-flood approach in L-moment framework. Very recently, shortcomings associated with assumptions of Index-flood approach motivated authors to develop a novel mathematical approach to RFA. The approach involves (i) identification of an appropriate frequency distribution to fit the random variable (flood) being analysed for homogeneous region, (ii) use of a proposed transformation mechanism to map observations of the variable from original space to a dimensionless space where the form of distribution does not change, and variation in values of its parameters is minimal across sites, (iii) construction of a growth curve in the dimensionless space, and (iv) mapping the curve to the original space for the target site by applying inverse transformation to arrive at required quantile(s) for the site. Effectiveness of the proposed approach in predicting quantiles for ungauged sites is demonstrated through a case study on watersheds in Godavari basin, India, using a jackknife procedure. Formation of homogeneous regions is based on region-of-influence method. Results are compared with those obtained by using conventional index-flood procedure.Results indicate that the proposed approach outperforms conventional index-flood approach. Keywords:Regional Frequency Analysis, Design flood, Lmoment, Region-of-influence 1. INTRODUCTION Estimation of design quantile of hydro-meteorological events such as floods at target locations in river basins having sparse/no records is one of the major challenges for hydrologists. To obtain the required design quantile, Regional Frequency Analysis (RFA) gained wide recognition The analysis involves (i) use of a regionalization approach for identification of locations that are similar to the target location (site), in terms of mechanisms influencing the variable being analyzed, to form a

HYDRO 2014 International

homogeneous region, and (ii) use of a RFA approach to fit a distribution to information pooled from the region for arriving at design estimate. Among the various RFA approaches developed in the past, conventional index-flood (CIF) approach (Dalrymple, 1960) gained wide recognition. The CIF approach considers the following assumptions: (i) Records of the variable at each site in a region are identically distributed; (ii) Records at each site are serially independent; (iii) There is no dependence between records at different sites; and (iv) Frequency distribution of the variable is identical across sites in the region, except for a site-specific scaling factor called index-flood. Of these assumptions, the first three are generally valid for analysis of a random variable representing hydro-meteorological extreme event, but the fourth is specific to only index-flood related approach. Implementation of the CIF approach involves normalization of records of the variable for each site by dividing them by the site‟s scaling factor and combining information from those normalized records to construct a „dimensionless distribution function‟ (growth curve) that is assumed to be unique for all the sites in the region. Required quantiles at the target site are estimated by multiplying the growth curve by sitespecific scaling factor, which is often chosen as mean of the variable. For the index-flood approach to be effective, the aforementioned assumptions (i)-(iv) should be valid for the records before and after normalization. Validity of the first three assumptions can be ensured by considering the scaling factor to be a population statistic. However, as population statistic is unknown in real world scenario, modelers chose sample statistic for normalization. In real world scenario, the scale and shape parameters of sites in a homogeneous region may not be close enough to be considered identical, even if the type of frequency distribution is the same for all the sites in the region. The shortcomings associated with CIF approach motivated the authors to develop a newmathematical approach to RFA. The RFA is deemed to be effective if knowledge of location, scale as well as shape parameters of all the sites is utilized in the analysis, to properly characterize the growth curve (dimensionless distribution function) that represents the region. The proposed approach involves: (i) identification of an appropriate frequency distribution to fit the random variable being analyzed for the homogeneous region, (ii) use of a proposed transformation mechanism to map observations of the variable from original space to a dimensionless space where the form of distribution does not change, and variation in values of location, scale as well as shape parameters of the distribution is minimal across sites, thus satisfying all the assumptions of index-flood approach, (iii) construction of a growth curve in the dimensionless space, and (iv) mapping the growth curve to the original space for the target site by applying proposed inverse transformation to arrive at required quantile(s) for the site. The reminder of this paper is structured as follows: Methodology for new mathematicalRFA approach is presented in section 2.1 and that of CIF approach is provided in section 2.2. Effectiveness of the new mathematical approach is demonstrated by application to real world data in section 3. Finally, summary and conclusions are given in section 4.

MANIT Bhopal

Page 209

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

2. METHODOLOGY 2.1 Methodology for new mathematical approach to RFA This section presents methodology of a novel mathematical approach that was recently proposed by authors (Basu and Srinivas, 2013). Let N denote the number of sites in a region that is homogeneous with respect to a random variable X depicting peak flows. Let x denote an observation (data point) corresponding to X . Implement the following steps to arrive at regional quantile function for a target site in the region. (i) Identify an appropriate regional frequency distribution to fit X . In real world scenario, the distribution can be identified using observations (data) corresponding to sites in the region by an effective regional goodness-of-fit test. (ii) Map observations corresponding to X from the original space to those corresponding to a random variable Y in a dimensionless space, such that frequency distribution of X and Y remain the same, and variation in at-sites values of location, scale as well as shape parameters of the distribution is minimal. Use equation (1) for mapping when X follows Generalized Logistic (GLO), Generalized Extreme Value (GEV), Generalized Pareto (GPA) or Generalized Normal (GNO) distributions, and use equation (2) for mapping when X followsPearson type-3 (PE3) distribution.

1  kX  x   X   ln 1   , x  X , y  Y kX  X  x X y , x  X , y  Y X y

Where

X

equation

denotes location parameter, (1)

denote

parameters, whereas

X

respectively

X

scale

and

k X in

and

shape

in equation (2) represents scale

parameter of the frequency distribution of X . Equation of cumulative distribution function (CDF) of X corresponding toGLO, GEV, GPA, GNO and PE3 distributions can be found in Hosking and Wallis (1997). The CDF of Y that follows GLO, GEV, GPA or GNO distributions, and the corresponding values for L-moments and parameters are given in Table 1, while those for PE3 distribution are provided in Table 2. It may be noted that the values of location, scale and shape parameters for GLO, GEV, GPA, and GNO populations are 0, 1, and 0 respectively. Further values of location and scale parameters for PE3 population are 0 and 1 respectively, whereas the value of shape parameter is the same as that in the original space. Details pertaining to derivation of population parameter values and the corresponding equations for population growth curves in the dimensionless space can be found in Basu and Srinivas (2013, Appendix).

HYDRO 2014 International

(iii) Compute L-statistics corresponding to each of the sites in the dimensionless space using values obtained from mapping of observations and use those as the basis to estimate regional average L-statistics. (iv) Estimate location, scale and shape parameters of regional frequency distribution using the regional average Lstatistics and construct growth curve

ˆy  F  in the

dimensionless space. (v) To arrive at regional quantile function for the target site, map the growth curve to the original space by applying proposed inverse transformation equation. Use equation (3) if regional frequency distribution is among GLO, GEV, GPA or GNO, and equation (4) if it is PE3.

x  F    X 

 X k X

1  exp k  ˆy  F  X

x  F    X   X  ˆy  F  Where

 X

denotes location parameter,

 X

and

k X in

equation (3) represent respectively scale and shape parameters, and

 X

in equation (4) represents scale

parameter corresponding to the target site. The subscript X indicates that all the parameters are estimated in the original space. Those parameters can be reliably estimated using observations at the target site if record length for that site is large enough. However, if the site is ungauged or has inadequate data, the required parameters can be estimated based on regional information by various methods. One option is to estimate (1) those parameters using regional average values of L-statistics. An alternate option is to estimate those parameters by using regression relationships developed between each (2) of them and site-specific attributes that influence the variable being analyzed. The site-specific attributes should be those that are readily available even for ungauged locations. For example, catchment area, slope, drainage density and soil characteristics could be considered as attributes in the case of RFA of floods. Table 1. Formulations related to GLO, GEV, GPA and GNO frequency distributions for the random variable Y . FY  y  is cumulative distribution function, 1Y , 2Y and 3Y are the first three L-moments,  Y ,  Y , and kY denote, location, scale and

shape parameters respectively, and y  F  is population growth

MANIT Bhopal

curve.

Page 210

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 Implementation of CIF approach involves the following steps: (i) Normalize peak flow values corresponding to each gauged site in the region by dividing them by the sites‟ scaling factor, which is considered to be the mean annual peak flow. (ii) Estimate L-statistics (L-mean; coefficient of L-variation, Lskewness, L-kurtosis) corresponding to each of the sites using the respective normalized records. (iii) Compute regional average L-statistics by taking weighted average of at-site values of those statistics computed in step (ii), with weights being proportional to sites‟ record length. (iv) Use the regional average L-statistics as the basis to identity an appropriate regional frequency distribution by regional goodness-of-fit test (Hosking and Wallis, 1997). (v) Let

Table 2. Formulations related to random variable Y in case of PE3 frequency distribution. FY  y  is cumulative distribution function, 1Y , and 2Y are the first two L-moments,  Y , Y and  denote parameters related to distribution of random variable Y and y  F  is population growth curve.

q  denote

CDF (quantile function) corresponding to

the fitted distribution. Refer to it as growth curve. (vi) Determineregional quantile function ungauged site

k

as,

Qk  F   q  F   k ,

Qk 

for the

F   0,1

where q  F  is ordinate of growth curve corresponding to non-exceedance probability

F ,and  k

is scaling factor

(index-flood) corresponding to the ungauged site. The factor is estimated using regression relationship developed between the scaling factor and catchment attributes corresponding to gauged sites in the region. Attributes should be those that influence peak flows in catchments of the study area and which can be determined even for ungauged locations. Typical examples of attributes include variables related to catchment‟s physiography, shape, soil, drainage,climate, land-use/land-cover, and geographic location.

3. CASE STUDY 3.1.Description of study area and data Effectiveness of the new mathematicalRFA approach in predicting quantiles for ungauged sites is demonstrated through a case study on watersheds in Godavari river basin, India, using a jackknife procedure. The river basin extends from 16°16' and 23°43' north latitude and 73°26' and 83°07' east longitude, and has an area of 3,12,813 km2 (Figure 1). The river originates near Trayambak in the state of Maharashtra at an elevation of 1067 m, and flows from west to east and confluences with Bay of Bengal near Rajahmundry in Andhra Pradesh. The river has its catchment in Maharashtra, Karnataka, Madhya Pradesh, Chhattisgarh, Orissa and Andhra Pradesh states. Boundary of the river basin was extracted from watershed atlas (AISLUS, 1990).

2.2. Methodology approach to RFA

for

conventional

HYDRO 2014 International

index-flood

(CIF)

Information on annual maximum flows at 50 sites (gauges) in the Godavari river basin, their location (latitude and longitude) and contributing drainage areas was collated from Central Water Commission (CWC) offices in Hyderabad and Nagpur, India. Watershed corresponding to each of the gauges was delineated from 90m resolution Shuttle Radar Topography Mission

MANIT Bhopal

Page 211

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

(SRTM) digital elevation model (DEM)using ArcHydro tools in ArcGIS environment. Attributes of the watersheds, namely average elevation (above mean sea level), perimeter, length of longest stream, main stream slope, drainage density, compactness coefficient, circularity ratio, form factor and elongation ratio were computed using tools in ArcGIS. In addition, area weighted annual rainfall was computed for each of the watersheds using one-degree resolution gridded daily rainfall data available for the period 1951-2004 from India Meteorological Department (IMD). Information on nature, areal extent and spatial distribution of soils in the study region was extracted from soil map obtained from National Bureau of Soil Survey and Land Use Planning (NBSS&LUP). Further, information pertaining to land-use/landcover was extracted from Earth Science Data Interface (ESDI) at the Global Land Cover Facility (GLCF) available at web site: http://glcfapp.umiacs.umd.edu. The extracted information includes areas classified as built-up, agricultural, forest, water bodies and waste lands.

Figure 1.Location of gauges considered for thepresent study in Godavari river basin 3.2. Results and Discussion Database of attributes prepared for watersheds corresponding to 50 sites in the Godavari river basin was scrutinized to identify irredundant attributes that are fairly well correlated with mean of Annual Maximum Flows (AMFs). The attributes identified based on this analysis were drainage area, perimeter, main channel slope and average watershed elevation. Those four attributes together with two location indicators (latitude and longitude) were chosen as attributes for regionalization.Among the six attributes, values corresponding to „drainage area‟ were quite large and their distribution was highly skewed. Consequently, those values were transformed using logarithmic transformation. Subsequently values (or transformed values) corresponding to each of the six attributeswere standardized by

HYDRO 2014 International

subtracting by its respective mean and then dividing by their standard deviation. The resulting values are referred to as scaled attributes. Jackknife procedure was implemented to demonstrate effectiveness of the new mathematicalRFA approach in predicting quantiles for ungauged sites. It involved considering one site at a time (from among 50 sites) to be ungauged, and preparing pooling group (region) for the ungaugedsite based on „Region of Influence‟ (ROI) (Burn, 1990) approach. The ROI approach isone of the widely used approaches for regionalization, though none of the available regionalization approaches is proven to be universally superior.To prepare pooling group for the ungauged site using ROI approach, other gauged sites were ranked in ascending order of their Euclidean distance to the ungauged site in the six-dimensional space of the scaled attributes. Following this, those sites were considered one at a time (in order of their distance), and assigned to the pooling group until collective record length of all the sites in the group exceeded 500 station-years. This ensures that pooled information is adequate to determine quantiles corresponding to return period T up to 100-years, as per 5T rule (Institute of Hydrology,1999), and adequate sites are available to develop regression relationship using information in the group for estimating first Lmoment (index-flood) for the ungauged site. The foregoing analysis yielded 50 pooling groups, each corresponding to one of the 50 sites in the study area that was assumed to be ungauged. To arrive at regional quantile function for ungauged site corresponding to each of the 50 pooling groups, the RFA was performed on each pooling group using the new mathematical approach (MA) and the CIF approach described in section 2. The regional quantile function constructed for an ungauged site using each of the approaches was compared with the “true” quantile function (CDF) corresponding to the site for five return periods (T = 25, 50, 75, 100 and 200 years) in terms of three performancemeasures (R-bias, AR-bias, and R-RMSE). The “true” quantile function was constructed by fitting the best-fit frequency distribution to AMF data available for the ungaugedsite, following the conventional practice (e.g., Cunderlik and Burn, 2006). The best-fit at-site frequency distribution was found to be GLO for 10 sites, GEV for 4 sites, GNO for 8 sites, PE3 for 15 sites, and GPA for 13 sites using Lmoment based goodness-of-fit test (Hosking and Wallis, 1997) with 90% confidence level. Values of the performancemeasures indicate that errors are significantly lower for the MA when compared to that for CIF method (Table 3). To gain further insight, scatter plots between the “true” at-site quantile estimates and regional quantile estimates based on MA and CIF were prepared for various return periods. They showed that points corresponding to PA are less deviated with respect to the solid 1:1 line than those corresponding to CIF approach. Results corresponding to a typical return period (T = 100 years) are presented in Figure 2, for brevity. Overall the results indicate that the proposed approach offers significant improvement over the CIF method for RFA.

MANIT Bhopal

Page 212

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Table 3.Performance measures R-bias, AR-bias and R-RMSE computed based on errors in flood quantiles estimated corresponding to 50ungauged sites.

REFERENCES: i. AISLUS, 1990 Watershed atlas of India, All India soil and land use survey, Ministry of Agriculture, Government of India. ii. Basu, B., and V. V. Srinivas (2013), Formulation of a mathematical approach to regional frequency analysis, Water Resour. Res., 49, doi:10.1002/wrcr.20540. iii. Burn, D.H. (1990), Evaluation of regional flood frequency analysis with a region of influence approach, Water Resour. Res., 26(10), 2257-2265, doi:10.1029/WR026i010p02257. iv. Cunderlik, J. M., and D. H. Burn (2006), Switching the pooling similarity distances: Mahalanobis for Euclidean, Water Resour. Res., 42(3), W03409, doi:10.1029/2005WR004245. v. Dalrymple, T. (1960), Flood frequency analysis, U.S. Geol. Surv. Water Supply Pap., 1543-A, 11 – 51. vi. Hosking, J. R. M., and J. R. Wallis (1997), Regional frequency analysis: An approach based on L-moments, Cambridge University Press, New York, USA. vii. Institute of Hydrology (1999), Flood Estimation Handbook, vol. 3, Wallingford, UK.

Performance Comparative Of Wavelets And Savitzky-Golay Filter On Bathymetry Survey Data Figure 11. Comparison of at-site (true) quantile estimates with regional quantile estimates for ungauged sites based on new mathematical approach and CIF methods for 100-year return period. The solid 1:1 line corresponds to the case where at-site and regional estimates are equal. A method is considered to be effective if points corresponding to the method are closer to the 2

solid line. R (coefficient of determination) corresponds to the dash-dot trend line fitted to points in a plot. 4. SUMMARY AND CONCLUSIONS The key assumption of the conventional index-flood approach is that it requires location, scale and shape parameters of frequency distributions of normalized records to be identical for all the sites in a homogeneous region. For practical applications, this assumption is always violated, which leads to ineffective quantile estimation for ungauged sites using conventional index flood approach. To overcome the shortcoming of CIF approach, a novel mathematical approach is proposed for RFA in Lmoment framework. Transformation mechanisms corresponding to various commonly used frequency distributions are proposed to facilitate mapping the random variable being analyzed from original space to a dimensionless space where distribution of the random variable does not change, and deviations of regional estimates of all the parameters (location, scale, shape) of the distribution with respect to their population values as well as atsite estimates are minimal. The location, scale and shape parameters corresponding to GLO, GEV, GPA and GNO populations are analytically derived to be 0, 1 and 0 respectively, in the dimensionless space. Experiments on real world data showed that the new mathematicalapproach offers significant improvement over CIF, method in RFA. Further improvement in results could be possible by considering Mahalanobis distance to form ROI (Cunderlik and Burn, 2006), instead of Euclidean distance considered in this study.

HYDRO 2014 International

M.Selva Balan1 Arnab Das2 Chief Research Officer, Central Water and Power Research Station, Khadakwasla, Pune 411024, India 2 Commander, Indian Navy, Military Institute of Technology, Girinagar, Pune-41125. India Email: [email protected]

1

ABSTRACT : Bathymetry survey is one of the most reliable and practical way to assess the reservoir capacity as well as to estimate the sediment volume. Accurate estimation of reservoirs volume is of crucial importance to make optimum utilization of stored water and to plan the reservoir operations. This also will enable the dam authorities to plan the dredging techniques. The correct knowledge of the volume of dams facilitate in planning the amount of water discharge and silt removal. The volume is determined using the area which is extracted from the satellite imagery and depth collected through echo sounder by running a boat along survey lines. A precise, linear indication of the depth of water as well as the sediment deposit in a specific part of water body is what always required. Presently there are a wide variety of ways to produce a signal that tracks the depth of water bodies. The Ultrasonic signal offers the benefits of shorter wavelength which resolves smaller details and inaudibility so humans are unaffected, hence most commonly used for the depth estimation. This signal is affected by various underwater noises which results in inaccurate depth estimation. In case of finding the layer width below the sediment the reflected ultrasound signal gets severely affected by the underwater noises. The objective of this paper is to provide noise reduction methods for underwater acoustic signal. In present work, the signal processing is done on the data collected using TC2122 dual frequency echo transducer. There are two signal processing techniques which are applied on a case study: The first method is denoising algorithm based on Stationary wavelet transform (SWT) and second method is Savitzky-Golay filter. The results are evaluated based on the criteria of peak signal to noise ratio and volume estimation is done by combining the data related to different locations of the

MANIT Bhopal

Page 213

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

reservoir and plotting them inside the boundary extracted from satellite imagery. However the results obtained with SavitzkyGolay filter matches acceptable level of interpolation and also matches the depth measured at site. This paper shows the performance of two newly developed techniques applied on depth data which was acquired with underwater noise. 3D Surfer plots of the reservoir whose depth and volume estimation has to be done are shown with different processing for the performance comparison. Keywords: Reservoir Sedimentation, Bathymetry Survey, Savitzky-Golay filter, Wavelet transform 1. INTRODUCTION: Irrigation and Agriculture are the main occupations of the people of India for thousands of years. Amongst the natural resources of a country, fresh water reservoirs i.e. dams, lakes etc are of utmost significance. The water stored by the dams can also be used to prevent floods and facilitate forestation in the catchments areas of the reservoirs. The measurement of capacity of reservoir is of crucial importance to regulate the water discharge from the reservoir for meeting the demands of irrigation and drinking water supply. The volume measurement is done using area and depth of the reservoir. Hence area and depth of the reservoir are to be calculated very precisely. Depth measurement of water bodies has developed remarkably in the last few decades with the adaptation of new ultrasonic techniques, which is proven successful among other methods based on image processing, airborne laser and mechanical systems. Photo bathymetry method, discussed by M. Selva Balan, et all (2013) based on image processing, digitally processes the aerial pictures to correlate light intensity with depth. This method is fast depth below the water cannot be measured with it. So it remains a tool for assessing the present area and approximate volume. An airborne laser system utilizes method of estimating the time delay between the surface and bottom reflections of the transmitted laser light. These systems are efficient, high speed and have good coverage but water clarity is the primary constraints as well as initial and operational cost are higher. Depth measurement methods based on acoustic uses ultrasonic signal and are classified as single beam and multiple beam eco sounding. The ultrasonic signal is transmitted towards the bottom of the reservoir and time interval required for the signal to reflect and travel back to the transducer is measured. Prior knowledge of velocity of ultrasonic signal in water and the time taken gives the distance travelled which is the depth of the reservoir. Multibeam eco sounding comprises of multiple narrow single beam transducers mounted near to each other and focussed at equally spaced angles for covering a large space beneath the boat. In this paper single beam eco sounding is utilized as it is simple and inexpensive.

Celsius in temperature, salinity which is a measure of the quantity of dissolved salts and other minerals in water and the total amount of dissolved solids in water. As shown in International hydrographic Bureau, (2005) the pressure also has a significant impact on the sound velocity variation and has a major influence on the sound velocity in deep water. When an ultrasonic wave is transmitted through water, it is expected to reach the bottom and then reflect back, but instead of this, it changes the characteristics (i.e. picks up noise) due to the medium as well as the reflective surface. However submerged trees and rocks create large spikes, which are mainly due to multipath effect. This gives a false bottom anticipation which doesn‟t provide the accurate results. The reflected signal when graphically plotted clearly indicates the unwanted sharp peaks, which are normally interpolated with standard mathematical techniques as given in Surfer manual ver. 8. The focus of this study is to analyse the reflected signal received through the sediment particles, which are corrupted badly than the surface reflections. The raw depth signal is denoised by applying signal processing techniques, which is then processed on Surfer ver.8 software to plot the 3D images of the reservoir bed. These sharp peaks could be the reflections from the suspended obstacles which come in the path of the transmitted ultrasonic signal. The data was collected using sensor Reason‟s TC2122 dual frequency survey echo sounder transducer which works on two resonant frequencies 33 kHz and 200 kHz and Reson's Navisound 415 hydrographic single beam echo sounder. General assumption is that the noise present is white Gaussian noise but the underwater noise does not full fill the classical white noise assumption [3] and hence Non-white noise is assumed. To reduce noise from the given data and to estimate approximate depth, two techniques are applied- denoising based on Stationary Wavelet Transform and Savitzky-Golay filter. This paper is organized as follows:-Section 2 deals with methods, limitations, wavelet transforms, Savitzky-Golay filter, section 3 & Section 4 deal with results & conclusion respectively. 2. MATERIAL AND METHODS Volume of the reservoir measurement requires two important aspects namely; getting the position coordinates accurate and the third dimension (i.e. depth). The advent of latest GPS technology allows us to get the position to accuracy in the range of centimeters. However the depth estimation depends on the method and the various nonlinear properties it encounters.

1.1 Measurement of reservoir volume: Ultrasound wave is basically cyclic sound pressure whose frequency ranges from 15 kHz to 200 kHz as discussed by Sabuj Das Gupta (2012). The depth measurement is quite sensitive to variations of the sound velocity profile. The sound velocity profile is affected by factors such as, variation of one degree

HYDRO 2014 International

2.1 Limitations of Existing techniques Echo-sounders are basically designed to operate in standard frequency. However the medium characteristics it is used is not same always. Also the characteristics of the bottom surface are

MANIT Bhopal

Page 214

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

not of same characteristics. This results in errors in terms of unacceptable depth readings. This could not be corrected beyond a limit as the echo reflection differs from different objects found below the water. CWPRS and many states are using a particular type of Echo sounder provided under hydrology project. However the data is collected it comes with various spiky non Gaussian noises, which could not been eliminated fully by the filters and the interpretation techniques provided in the software supports these system. As explained by M. Selva Balan, et all (2013) for large reservoir the preplanning is essential, which is possible with the image processing techniques applied on an satellite imagery as the one shown in fig 1 below.

signals, orthogonality and biorthogonality as per Michel Misiti et al (2000).

Fig 1. Contour extraction of the reservoir for pre survey planning

As per Meyer M. Kreidl et al (2002) there are a number of wavelets that can be used for noise removal: Haar, Daubechies, Symlet, Coiflet, Biorthogonal, Reverse Biorthogonal to name few. All of them are wavelets with filter having either orthogonality or biorthogonality. The HARR wavelets are performs the mathematical operations of averaging and finding difference on the decomposed values of signal. Daubechies wavelet are defined same as HARR, has balanced frequency responses but nonlinear phase responses. Symlet wavelet comprises of a symmetrical wavelet. Coiflet is the member of a family of wavelets having zero moments in the support of the functions and also in the scaling function. Biorthogonal wavelets are extension of orthogonal wavelet families to resolve the problem of incompatibility between the symmetry and perfect reconstruction. As per Michel Misiti et.al (2000) Meyer wavelet is an infinitely derivable orthogonal wavelet without compact support. In order to use the wavelet transform effectively the details of the particular application should be taken into account and the appropriate wavelet should be chosen. S.Kumari et. Al (2012) explained that they are chosen based on their shape and their ability to analyze signal in particular application. The performance of wavelet based denoising depends on wavelet decomposition structure.

As detailed by M.Selva Balan et al (2013), in normal conditions, the raw data collected by a survey boat generates lots of noise, which is very difficult to be removed by any manual methods. And hence two new filters were developed namely Wavelet and Savitzky-Golay.

For selecting particular type of wavelet, performance comparison of some known wavelet families was done and their effect on the given signal was observed. In present case, as explained earlier smoothness of the surface is the basic criteria for depth estimation, so accordingly one wavelet from each wavelet family was selected. These are shown in Table 1. Table 1.Wavelet selected from respective wavelet family.

2.2 WAVELET TRANSFORM Wavelet transforms have become one of the most important and powerful tool for signal denoising as shown by SJS Tsai, (2002). Discrete Stationary Wavelet Transform is undecimated versions of discrete wavelet transform which is used for signal denoising and pattern recognition as shown by Chu-Kueitu et al, (2004). The main idea is to average several detailed coefficients which are obtained by decomposition of the input signal as explained by V. Matz et al, (2005).Signal denoising using wavelet consists of three steps of decomposition, thresholding of the coefficients and reconstruction. Decomposition of signal is done over an orthogonal wavelet basis using the discrete transform. Thresholding is used to select a part of the coefficients and using the threshold coefficients the signal is reconstructed. The reproduced signal is the denoised signal. Wavelet transforms make use of different basis functions to decompose the signal. These basis functions can be differentiated by scaling and shifting parameters. The properties of wavelet play a key role in the selection of a wavelet for a particular application. The main properties of wavelet include speed of convergence which quantifies the localization of the wavelet in time and frequency, symmetry for avoiding dephasing, regularity to obtain reconstructed smooth and regular

HYDRO 2014 International

Wavelet Family Haar Daubechies Symlet Coiflet Meyer Biorthogonal Reverse biorthogonal

Selected wavelet Haar db8 sym5 coif5 Dmey bior2.2 rbior2.2

The detailed and approximation coefficients are obtained using signal decomposition. Further decomposition of approximation coefficients up to specified level is done. The maximum decomposition level depends on number of data points contained in a data set. Present depth analysis 5 decomposition levels were found to be appropriate. K.Mathan Raj et. al (2011) shown a thresholding of data in wavelet domain to smooth out or to remove some of the coefficients of wavelet transform of measured sub-signal introduced due to noise or obstacles in water bodies. Two commonly used types of thresholding are hard and soft thresholding. In hard thresholding if any coefficient (x) less than

MANIT Bhopal

Page 215

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

threshold value(t) then it is set to zero otherwise it remains unchanged.

(1) Soft thresholding [10][11] is similar to hard thresholding with a little difference i.e. no coefficient remains unchanged instead it is shrunken by threshold value(t). The present analysis is done using soft thresholding technique.

(2) 2.3. SAVITZKY-GOLAY FILTER The Savitzky-Golay filter is a particular type of low-pass filter. Sophocles J. Orfanidis (2012) shows that it is well-adapted for data smoothing. It is also referred to as least-squares or Polynomial Smoothing filter. Rather than having their properties defined in the Fourier domain, and then translated to the time domain, Savitzky-Golay filters derive directly from a particular formulation of the data smoothing problem in the time domain as shown by Filip Wasilewski. Ronald W. Schafer (2011) shows that these filters are of type-I FIR low pass filters with nominal pass band gain of unity. Savitzky and Golay proposed the method of data smoothing based on local least-squares polynomial approximation. Polynomial smoothing is the process which replaces the noisy samples by the values that lie on the smooth polynomial curves drawn between the noisy samples. Sophocles J. Orfanidis (2012) has shown that for every polynomial order, the coefficients must be determined optimally such that the corresponding polynomial curve best fits the given data. Instead of applying averaging filter it is better to perform least squares fit of a small set of consecutive data points to a polynomial. Savitzky A., and Golay, M.J.E. (1964) proved that Least-squares fit technique is used to choose the polynomial coefficients such that they give minimum mean square error. The output smoothed value is taken at the center of the window to replace the original data. Fig 2 below shows the plots of raw data as well as S-Golay filter processed data.

Figure 2. Plots of Raw depth and data interpolated by S Golay Filter In Savitzky-Golay filter, the odd-indexed coefficients of the impulse response design polynomial are all zero. The nominal normalized cut off (3 dB down) frequency depends on both the implicit polynomial order and the length of the impulse response. The impulse response of filter is symmetric, so the frequency response is purely real. These filters have very flat frequency response in their pass bands and fair attenuation characteristics in their stop band regions. As per Ronald W. Schafer, (July 2011) following are the constraints on polynomial fitting; - The number of data points must be strictly greater than the number of undetermined coefficients to achieve smoothing by the Savitzky-Golay process. - If the order of the polynomial is too large, the solution will be of no value. Generalize algorithm is as follows: Consider frame size odd, and polynomial.

or

filter length N whered is order

is of

Ifx is noisy signal with noisy samples , n = 0,1,.......,L-1 and it is supposed to be replaced by its smoothed output version y which contains , n = 0,1,.......,L-1 then input vector hasn =L input points and x = is replaced byN dimensional one, havingM points on each side ofx. (3) There are 3 cases, for calculating the output result. These cases are explained in [16]. Smoothed output y is calculated as

HYDRO 2014 International

MANIT Bhopal

Page 216

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

(4) The Savitzky-Golay elements of matrixB.

filter

coefficients

are

the

(5) (6) Where

,

, whe

re-d [12][13]. 3. RESULTS AND ANALYSIS

From Table 3 it can be seen that as the order of polynomial increases, PSNR value also increases. So PSNR is directly proportional to order of polynomial for Savitzky-Golay filter. Computational complexity is less for higher order. (Processer used-Intel core i5) Table 4. Values of PSNR by varying frame size and with fixed order for Savitzky-Golay filter

As shown by S.Kumari et. al. (2012) the peak signal to noise ratio represents the measure of peak error. It is given as,

File

Or4_31

Or4_33

Or4_41

Or4_49

File1

44.0280

43.8143

43.0032

42.6637

File2

49.4808

49.1640

48.2272

47.8468

(7)

File3

44.4367

44.0604

43.4561

43.1529

Where

File4

44.7824

44.6722

44.1377

43.7461

File5

40.8566

40.6501

40.0252

39.9840

File6

37.2930

37.0813

37.0102

36.8819

File7 Avg.Tim e (sec)

41.1404

41.1594

40.9853

40.5424

2.41

2.55

2.53

2.56

(8) MSE is Mean Square Error with I = original value O= output value and R= maximum input value Generally PSNR should be greater than 30dB in order to reduce noise effectively. For comparing results of Savitzky-Golay filter, another parameter used is Time Constraints which is time required for execution of program. Table 2. Values of PSNR for different types of wavelets.

From Table 4 it can be seen that as the lesser the frame size, more is the PSNR. So PSNR is inversely proportional to frame size for Savitzky-Golay filter. Computational complexity is less for smaller frame size. (Processer used-Intel core i5) The volume of reservoir is determined using the area and depth at different locations in the bed. All the data related to these locations is collected to provide the complete profile of the reservoir and then boundary is applied for determining the volume in Surfer11 software. Actual volume of the reservoir calculated by design equation: 15475058 cubic meter Table 5. Values of Volume of reservoir without denoising and with denoising of signal. Without denoising

The results presented in Table 2 show PSNR values for different wavelets. It can be seen that Haar wavelet is giving better result than other wavelets in this case.

Volume in cubic meter Error Percentage error

15448266 26792 0.173

Denoised with Haar wavelet 15472741 2317 0.015

Denoised with Savitzky Golay 15475539 481 0.003

3D plots of depth data are obtained using surfer11 are shown in figures 3to 8 below on two different data sets collected from reservoirs:

Table 3. Values of PSNR by varying order and with fixed frame size for Savitzky-Golay filter.

HYDRO 2014 International

MANIT Bhopal

Page 217

International Journal of Engineering Research Issue Special3

Figure 3 : Original signal for right arm of lake

Figure 4 : Signal processed using Haar wavelet

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 5 : Signal processed using Savitzky-Golay Filter

denoised echo yields a smooth profile (ie less peaks) the reservoir volume become more realistic. The error percentage is reduced to 0 .003 for the signal denoised with Savitzy-Golay. The analysis has been done on a large volume data where percentage error for the signal without denoising and with denoising is small. However for a reservoir with less volume this much error will be a considerable amount that will affect the planning for the water discharge. In case of low frequency reflections (which represent depth with sediment) the variations due to noise are huge which will give erroneous sediment volume, which in turn affects the reservoir planning the dredging process. With this filters the accuracy of sediment volume will be considerably reduced. REFERENCES

Figure 6 : Reservoir bed plotted from RAW data

Figure 7 : Reservoir bed denoised with Haar wavelet

Figure 8 : Reservoir bed denoised with Savitzky-Golay

The 3D profiles shows that wavelet and Savitzky-Golay filters have smoothened the noisy data and hence improves the accuracy of sedimentation volume calculations. Fig 7 shows the capacity loss calculated with one survey using two frequencies. Li ve Vol ume Pl ot (Ch 1 i n Mcum) Ori gi nal Vol ume pl ot Area (Ch 2 i n Mcum)

Volume Plot 350 300

Volume (Mcum)

250 200 150 100 50 0 -10

-5

0

5

10

15

20

-50

Water Level (meters)

25

30

Fig7. Final plot showing the loss in capacity based on single survey done with two different frequencies 35

40

4. CONCLUSION The analysis of ultrasonic depth data received through sediment and water using two techniques: HARR wavelet Transform and Savitzky-Golay filter. It is found that out of all wavelet transforms, HARR wavelet is most suitable for noise reduction in ultrasonic signal based on high PSNR value. In SavitzkyGolay Filter analysis, higher order of polynomial with lesser frame size increases the PSNR. The results from surfer plots show that the HARR wavelet with decomposition level up to 5 and Savitzky-Golay filter with order 4 and frame size 31 can be effectively used for smoothing the data obtained which can lead to estimation of depth with minimum error using empirical formula designed for a particular application.

i. Arnaud Jarrot, Cornel Ioana, Andr´e Quinquis, (2005)"Denoising Underwater Signals Propagating Through Multi–path Channels", Oceans Europe (Volume:1) pp.501-506. ii.Bernhard Wieland, (October 2009) "Speech Signal Noise Reduction with Wavelets", pp.55-56. iii. Chu-Kueitu, Yan-Yao Jang, (2004)"Development of Noise Reduction Algorithm for Underwater Signals", Underwater Technology, International Symposium on, pp.175-179. iv. Golden Software, Surfer Manual online ver 12. v. International hydrographic Bureau, (2005)"Manual on hydrography", M-13, pp.126. vi. K.Mathan Raj, S.Sakthivel Murugan, V. Natarajan, S.Radha, (2011)"Denoising Algorithm using Wavelet for Underwater Signal Affected by Wind Driven Ambient Noise", Recent Trends in Information Technology (ICRTIT), pp.943-946. vii. Md. Abdul Awal, Sheikh Shanawaz Mostafa and Mohiuddin Ahmad, (2011)"Performance Analysis of Savitzky-GolaySmoothing Filter Using ECG Signal", IJCIT, VOLUME 01 ISSUE 02, pp.24-29. viii. M. Kreidl, P. Houfek, (2002)"Reducing Ultrasounic Signal Noise by Algorithms based on Wavelet Thresholding", Acts Polytechnica Vol. 42, pp.6065. ix. Michel Misiti, Yves Misiti, Georges Oppenheim, Jean-Michel Poggi, "Wavelets and their Applications", ISTE 2000. x. M. Selva Balan, Dr. Arnab Das, Madhur Khandelwal, Piyush Chaoudhari, ―A Review of Various Technologies for Depth Measurement in Estimating Reservoir Sedimentaion‖, IJERT, Vol. 2, Issue 10, Oct 2013, pp.223-228. xi. M. Selva Balan, Sedimentation survey using dual frequency echo sounder, Two days work shop on ―Reservoir Sedimentation‖ by Beuro of Indian Standards (BIS) , January 2013. xii. Ronald W. Schafer, (July 2011)"What is a Savitzky-Golay filter?", IEEE SIGNAL PROCESSING MAGAZINE, pp.111-115. xiii. Savitzky A., and Golay, M.J.E. (1964)"Analytical Chemistry", Volume 36, pp.1627-1639. xiv. Sabuj Das Gupta, Islam Md. Shahinur, Akond Anisul Haque, Amin Ruhul, Sudip Majumder,(October 2012)"Design and Implementation of Water Depth Measurement and Object Detection Model Using Ultrasonic Signal System",International Journal of Engineering Research and Development, Volume 4, Issue 3, pp.62-69. xv. SJS Tsai, (2002)"Chapter 4 Wavelet Transform and Denoising". xvi. Sophocles J. Orfanidis, (2010)"Introduction To Signal Processing", Pearson Education, Inc., pp.427-451. xvii. S.Kumari, R.Vijay, (January 2012)"Effect of Symlet Filter Order on Denoising of Still Images", Advanced Computing :An International Journal(ACIJ).Vol.3.No.1, pp.137-143. xviii. V. Matz and J. Kerka, "DIGITAL SIGNAL PROCESSING OF ULTRASONIC SIGNALS" 2005, pp.3 xix. wavelets.pybytes.com by Filip Wasilewski. xx. William H. Press, Brian P. Flannery, Saul A. Teukolsky, William T. Vetterling, (1988-1992)"Numerical Recipes in C:The Art of Scientific Computing", Cambridge University Press, pp.650-651

The reflected echo of sensor without denoising when plotted yields a profile consisting of a number of peaks. Since the

HYDRO 2014 International

MANIT Bhopal

Page 218

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Simulation Study On Performance Of Household Rainwater Harvesting Systems

overview of the materials and methods and the results of the study are discussed in the subsequent sections.

P.G. Jairaj1 P. Athulya2 Professor in Civil Engineering, College of Engineering, Trivandrum-695016, Kerala, India 2 Former M.Tech Student, College of Engineering, Trivandrum695016, Kerala, India Email: [email protected], [email protected]

2. MATERIAL AND METHODS:

1

ABSTRACT Water shortage has become a serious problem all over the world due to rapid urbanization and climatic changes. To cope with such situation small onsite Rainwater Harvesting (RWH) Systems can act as alternate water supply source in rural as well as urban areas. But the efficiency of these RWH systems is largely affected by the distribution pattern of rainfall as well water demands. This paper investigates the performance of Rooftop household rainwater collection systems located at various geographic regions in Kerala state, India considering the variation in demand and rainfall. The operation of Rooftop household rainwater collection systems was simulated using Standard Operating Policy (SOP), and the performance was evaluated by three indicators namely; Reliability, Resilience and Vulnerability. From the simulation study, it is revealed that while designing the rainwater collection systems, sufficient care is to be given to the spatial and temporal distribution pattern of rainfall. Keywords: Rainwater Harvesting System, Standard Operating Policy, Demand, Capacity, Performance evaluation

In the present study the performance of household roof top rainwater collection systems was described by its ability to satisfy the demand without failure. Using the actual rainfall data at the locations, the runoff from the catch surface was worked out on a daily basis. This runoff collected in the rainwater collection tank was used for satisfying the various household demands. A simulation model using Standard Operating Policy (SOP) was made use of for the computation of yield from the system and the evaluation of the system performance was carried out using the indicators: Reliability, Resilience and Vulnerability as follows. 2.1 Simulation Model: A typical flat rooftop household rainwater harvesting system having a collection tank capacity of was considered for carrying out the simulation study. The yield from the rainwater harvesting system was drawn according to the water demand. Simulation of the operation of the system was carried out using Standard Operating Policy (SOP) given by Equations (1) to (3). In simulation whenever the demand is not satisfied associated failure occurs, computed in terms of deficit volume defined by Equation (4).

(1)

1. INTRODUCTION: Due to anthropogenic activities the surface water systems are getting dried up, ground water is depleting and water bodies are getting polluted. Moreover the water resources are being depleted faster than it can be replenished. The need of rainwater harvesting (RWH) has been felt to meet the ever increasing demand for water and reduce the large volume of surface runoff. Among the RWH procedures the roof top harvesting using collection tanks is a widely used one. For a given roof top area the efficiency of the RWH system greatly depends on the variability in the rainfall and the demand and in turn is associated with the capacity of collection tank. An efficient rainwater harvesting system shall be able to accommodate the runoff coming from the catchment surface area so as to satisfy the demand with maximum reliability. This requires proper sizing of rainwater harvesting systems, so as to have the maximum efficiency. This paper focuses on the performance analysis of household rooftop rainwater collection systems located at various geographical areas of Kerala state, India, by analysing the performance indices: Reliability, Resilience and Vulnerability of the system subject to the restrictions imposed by capacity of the collection tank, demand to be met and the magnitude of available rainfall. A brief

HYDRO 2014 International

(2)

(3)

(4) where Yt is the yield from the collection system at period t (m3); Qt inflow to the collection tank in period t (m3); Dt is the demand during the period t (m3); St is the storage in the time period t and Spillt the spill occurring (m3) if any when the collection tank is full and Smax the maximum design capacity of the collection tank. Det represents the deficit occurring (m3) in period t. Performance of the system was evaluated in terms of Periodbased Reliability (R), Resilience (Res) and Vulnerability (Vul). Period based reliability estimation evaluates the system reliability on the basis of the number of time periods of non-

MANIT Bhopal

Page 219

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

failure of system to meet the water demand to the total number of periods in the study. The term Resilience is used as a measure of how fast the system is likely to return to satisfactory state, once the system has entered an unsatisfactory state. This definition of resilience (Res) is equal to the inverse of the mean value of time the system spends in an unsatisfactory state and computed using Equation (6) (Kjeldesn et al., 2005). Vulnerability was calculated as the mean deficit incurred during the period of study indicated by Equation 7 (Kjeldesn et al., 2005).

applied in Equations (5), (6) and (7) to obtain the period-based reliability, resilience and vulnerability of the system. 3. RESULTS AND DISCUSSION: The study focuses on the analysis of performance of the rainwater collection systems located in various geographical locations of Kerala state. Simulation models were developed to analyse the performance of the system. Performance analysis was carried out using the indicators Reliability, Resilience and Vulnerability subjected to the restrictions imposed by available rainfall, water demand and storage capacity of the collection tank.

(5) 3.1 Performance of RWH for Average Rainfall:

(6)

(7) where NT and Nfailure are the total number of periods in the study and number of periods in which failure occurs, d(j) represents the duration of jth failure event, v(j) is the deficit occurred during jth failure event and M is the number of failure events. 2.2 Analysis of problem: The performance analysis of rainwater collection systems in the areal extent of Kerala state located at: Trivandrum, Kollam in Southern region; Kottayam, Chittur, Cochin in Central region; Calicut, Kannur in the Northern region. The analysis was carried out on a yearly basis (June to May). The daily rainfall data for the period 1982 to 2011 for the IMD stations at Trivandrum, Kollam, Kottayam, Cochin, Chittur, Calicut and Kannur were made use of in the study. The details pertaining to the study are given in Athulya (2013). The daily yield from the rainwater collection system depends on water demand to be met, and was computed on the basis of daily per capita demand. As per IS 1172, the per-capita demand for the household systems in India is 135 lpcd. In the study, the variation in daily percapaita water demand was considered in the interval 30 lpcd to 135 lpcd, to incorporate the variation in demand values. A five user flat roof terrace house of effective catchment area of 100 m2 with coefficient of runoff of 0.75 was adopted for computation of runoff that can be harvested in the study. The temporal variation of rainfall was also incorporated by evaluating the system performance for average rainfall situation as well as rainfall values taken at different probability levels. For the cases studied, the total deficit and the number of period for which the system failed to satisfy the demand were worked out from the simulation results for different combinations of collection tank capacities and daily demand. This in turn is

HYDRO 2014 International

The simulation study of the operation of the roof top rainwater collection system at the different locations for average rainfall were carried out; yielded the reliability, resilience and vulnerability values for the specific demand and capacity of the system considered. The variation in reliability against capacity for specific demand values are tabulated to obtain the tradeoff as in Table 1. From the table it can be seen that for the RWH located in Southern Kerala the magnitude of rainfall limits the reliability of the system, while in the case of Northern Kerala the capacity of the collection tank limits the reliability of the system. For average rainfall situation resilience and vulnerability indices were also calculated for the proposed rainwater harvesting stations located in the study area; and the set of representative values obtained for station Kannur are given in Table 2. From the table, it can be observed that, resilience of the system increases with increase in capacity showing that the duration of time in which system spends in unsatisfactory state decreases in general. But the increase in resilience is found to be not uniform as in the case of reliability with capacity. Similarly even though vulnerability of the system decreases with increase in capacity it is found to be not directly related to the capacity of the collection tank. The vulnerability and resilience estimates generally exhibit a non-monotonic behavior, i.e. the estimates, for a specified demand, do not vary monotonically as the capacity increases. So it can be inferred that vulnerability and resilience indices describe the system performance once the failure has occurred, whereas the reliability index describes the overall efficiency of the system. So for further analysis in the present study the only reliability index was taken into account. 3.2 Performance of RWH system for variation in rainfall The system performance indicator reliability of household rooftop rainwater collection system with capacity of the collection tank was analyzed for probability levels of rainfall for the stations. The tradeoffs were generated between the reliability of the system and capacity of the collection tank for different demands and rainfall values taken at different probability levels and are tabulated in Table 3. From the table it can be seen that the performance RWH systems of the stations located in Southern region is poor even for 50 % probability level of rainfall, when compared to the

MANIT Bhopal

Page 220

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Table 1: Reliability of the rainwater collection system for average daily rainfall

Table 2: Resilience and vulnerability values obtained for station Kannur

Table 3: Reliability of the RWH system for different probability levels of rainfall

stations in Northern Kerala. It can be observed that the rainwater collection systems located at Kannur is found to be most reliable at all probability levels of rainfall. Also rainwater collection systems located at Trivandrum is found to be least reliable compared to the other stations, since the reliability obtained even at 50% probability of rainfall is less than 50% except for 30 lpcd demand, and for higher probability levels of rainfall, the system reliability obtained is less than 50% for all cases considered. From the study it can be seen that the uncertainty associated with rainfall values affects the performance of the system. 4. CONCLUSIONS The focus of the present study was to analyse the spatial and temporal variation in the performance of household rainwater collection systems incorporating the variability in rainfall and demand values. The performance analysis was carried out for the RWH systems located in different regions of Kerala state. The specific conclusions from the study are as follows: On analysing the performance of RWH for average rainfall situation it seen that RWH collection systems located in Northern region of Kerala are found to be more reliable compared to the Southern and Central regions since they are able to satisfy the complete household demands with

HYDRO 2014 International

MANIT Bhopal

Page 221

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 the need of the hour. In India, the Right to water has been protected as a fundamental human right by the Indian Supreme Court as part of the Right to Life guaranteed under Article 21of the Indian constitution. India with majority of population dwelling in rural areas faces the problem of acute shortage of potable water in some rural area. The present paper addresses such issues in one such rural area called Nawli village in the Mewat district of Haryana with community participation. The area has the problem of saline water which is unfit for drinking as well as other domestic uses. So on-ground water recharge measures were taken up with community participation. Rainwater harvesting is the oldest technology to provide water for human needs. It has been observed through our desktop research that small communities are increasingly accepting rainwater harvesting and its augmentation as a possible solution to meet their water needs. So the community-based water resource management practices can be the most suitable option which not only will help the community develop and meet their essentials but also give them a sense of accomplishment. Also ArcGIS tool came handy in dealing with the diverse geomorphic features of the area and demarcating streams and watersheds, which further helped in augmenting the possibility of maximum recharge of water. Keywords: ArcGIS, community participation, water resource management, transect walk. 1. INTRODUCTION: 1.1 Background to the study

5. ACKNOWLEDGEMENT The authors would like to express their gratitude to the Indian Meteorological Department (IMD) Trivandrum for having provided the data required for the study. REFERENCES: i. Athulya (2013) Modelling the performance of rainwater collection systems. Thesis submitted to University of Kerala. ii. Kjeldsen T.R. and Rosbjerg D (2005) Choice of reliability, resilience and vulnerability estimators for risk assessment of water resources systems. Hydrological Sciences Journal 49:755-767.

COMMUNITY-BASED WATER RESOURCE MANAGEMENT, STUDY AREA NAWLI VILLAGE, MEWAT DISTRICT, HARYANA Amit Kumar Dogra1 Nitin Singh2 1 41-R, Sec-IV, D.I.Z. Area, Gole Market, Delhi-110001, India 2 C-21,Nehru Nagar, Bhopal-462003 Email: [email protected] ABSTRACT The water regime in India is of major concern with the speedy and uncontrolled usage of ground water. Proper management and judicious planning for these underground water resources is

HYDRO 2014 International

Water is the key to development and sustenance of all communities. Under conditions of increasing stress on this essential renewable but scarce natural resource, effective and efficient management of water is a burning issue. Historically, drinking water supply in the rural areas in India has been outside the government‟s sphere of influence. Community-managed open wells, private wells, ponds and small-scale irrigation reservoirs have often been the main traditional sources of rural drinking water. The first government-installed rural water supply schemes were implemented in the 1950s as part of the Government policy to provide basic drinking water supply facilities to the rural population. (NRDWP, 2009-2012). Out of the 121 crores around 85 crores live in rural areas (Cenus, 2011) out of which around 37.7 million are affected by waterborne diseases annually, 1.5 million children are estimated to die of diarrhea alone and 73 million working days are lost due to waterborne disease each year. (WHO, 2012). Groundwater is the major source to meet the domestic, irrigation and industrial demands. Groundwater occurs in a wide range of rock types and usually requires little or no treatment; therefore, it is often the cheapest and simplest water supply option. Unlike surface sources, community participation in groundwater management is seldom seen and the government authorities have not given any attention to this so far. Since majority of the groundwater structures are owned by the farmers, community participation in groundwater use can help reducing overexploitation by their need.

MANIT Bhopal

Page 222

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

1.2 Statement of problem The present study aims at resolving the water crisis for the Nawli village of Mewat district, Haryana. The available water in this area became brackish in nature over a period of time due to over extraction and un-management irrigation practices. The community‟s knowledge is valuable in deciding what improvements are needed in specific local context. Also the „modernization programmes‟ can be more effective and also much less expensive than the current approach which hardly involves the user community. So the involvement of preexisting local experience and institutions will also facilitate proper management of the „revamped‟ water resources systems. The aim of the paper was to set up the plan to meet the water requirement of the people of Nawli village in the Mewat district of Haryana by community participation. Following objectives were defined to be addressed based on the research. 1.

2. 3.

Identification of water related issues through Participatory Rural Appraisal techniques such as Transect walks in the area and its direct and indirect effect on the community. To assess the feasibility and viability of rainwater harvesting in the given area (by studying best-practices) Giving possible solutions according to the research studies and analysis

Figure 1. Location Map of study area

Table. 1: Demographic profile of the area (Source: Census, 2011)

1.3 Study area Nawli is a Village in Ferozepur Jhirka Tehsil in Mewat District of Haryana State, India. The village is connected via SH-13 to different parts of Haryana. District headquarters is Mewat which is 5 km from the village. The village lies in the semi-arid region and receives an annual average rainfall of over 500mm but less than 650mm. (Authority, 2005). Total geographical area of the village is 21.7 sq. km. The location of Nawli is a Village in Haryana State of India is shown in Figure 1 and demographic profile of the area is given in Table 1.

2 MATERIAL AND METHODOLOGY Proposal Outline A checklist was made to procure data regarding the study of the area such as rainfall data, post and pre-monsoon data, type of soil, topography etc. Also the drainage pattern and contouranalysis was done with the help of Arc-GIS tool. And micro

HYDRO 2014 International

MANIT Bhopal

Page 223

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

catchments were defined and maps were made accordingly for building-up of check-dams as such. The water related issues were enlisted through transect-walks, organized with the help of locals. A transect walk is a PRA (Participatory Rural Appraisal) technique which is a systematic walk along a defined path (transect) across the community/project area together with the local people to explore the existing situation conditions by observing, asking, listening, looking and producing a transect diagram (Mohan, 2003). The village was divided into two parts, one in the form of upper settlement which was somehow away from the Aravalis and another in the form of lower settlement which was on the foothills of Aravalis. For the upper settlement two trails were identified to look upon the issues. Also for the lower settlement one trail was identified as such. Mapping of the existing situation was done and the also the water demand for the whole population was calculated. In the given upper settlement the majority of houses are pucca in nature. So rooftop rainwater harvesting is being proposed for the area which will meet to the drinking water demands. 3. RESULTS AND DISCUSSION The proposed measures for the area are check dams on the foothills. Most of the water demands of lower settlements will be met by these check dams. Also roof-top harvesting. Most of the houses were found to be pucca in nature in upper settlement and therefore roof-top harvesting is possible. The various calculations involved in the process are detailed out separately. 3.1 Calculations. Calculation for amount of rainfall that can be harvested: No. of households. :393 (source: primary survey) No. of pucca households. :314 Average household dimensions :30 sq. mts. Average annual rainfall :594 mm or 0.594 m Runoff coefficient considered :0.85 for roof catchments ; tiles and corrugated metal sheets Maximum amount of rainfall that can be harvested from the rooftop (CPWD, 2009): = Area of catchment (A) * average annual rainfall (R) * Runoff coefficient (C) Annual average water harvesting potential from 9 sq. mts. Roof = A*R*C = 30 * 0.594 * 0.85

HYDRO 2014 International

= 15.04 cu. m. (15,040 litres) Calculation for Drinking water requirements Daily per capita drinking water requirement (drinking and cooking only) is 10 litres. The water tank has to be designed for the dry period, i.e the period between the two consecutive rainy seasons. With a monsoon extending over three months, the dry season is of roughly 265 days. Drinking water requirement for a single household (dry season) = 265 * 8 * 10 = 21,200 litres Check dam specifications Depth of flow. six inches when drainage area is less than 5 acres. Average storage capacity. 39551 cubic meters. The center of the check dam must be at least 6 in (152 mm) lower than the outeredges. The drainage pattern of the aravalis is being prepared by contour analysis and by defining these drains check dams were made. The proposed number of structures are 6 in the whole village. With 5 of the check dams near the lower settlement. The maximum catchment area is around 10 acres and the minimum catchment is 7 acres for different micro-watersheds. Total no. of proposed check dams. 06 Average storage capacity of Check dams= 39551 cu. mts Total water retaining capacity = 237306 cu. mts. Or 237.30 million litres 3.2 Check Dam Location A check dam is a small, temporary or permanent dam constructed across a drainage ditch, swale or channel to lower the speed of concentrated flows for a certain design range of storm event. (NRCS manual) The contour analysis was done for the area and by the same drainage map was prepared. The major drain lines which dissect the lines were prepared using the flow direction tool in ArcGIS. The water can be restored in these structures for a whole of the dry season with the given amount of rain. With the use of suitable filter material water fit for household purposes can be obtained which will be free from any pollutants. Bio-sand filters can be used which costs between rupees 75,000 to 90,000 and cleans 2500-3000 L/d. Also the O&M costs are very low, around 300-500 (Rs.) (IRRAD, 2012) The check dams are proposed on the ends of these drains. Six check dams are proposed for the area, with five of them near lower settlement. The maximum catchment area is around 10 acres and the minimum catchment is 7 acres for different micro-watersheds. Firstly the stream flows were delineated by the help of flow accumulation tool in the ArcGIS. Than these minor streams were overlapped with the given map of Nawli village.

MANIT Bhopal

Page 224

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 4. Major catchments 3.3 Rooftop Rainwater Harvesting Figure 2. Major streams Rooftop rainwater harvesting is the technique through which rainwater is captured from the roof catchments and stored in reservoirs. Harvested rainwater can be stored in sub-surface ground water reservoir by adopting artificial recharge techniques to meet the household needs through storage in tanks. (C.G.W.B., 2007) In saline areas, rainwater provides good quality water and when recharged to groundwater, it reduces salinity and also helps in maintaining balance between the freshwater interfaces.

Figure 5. elements of typical water harvesting system Source: (Environment, 2011) Figure 3. location of check dams 4. CONCLUSIONS The total water demand for various utilities will be met over a course of time; with the proposed interventions. But again some limitations will remain for against the drinking water needs

HYDRO 2014 International

MANIT Bhopal

Page 225

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

which can be met by the water procured through check dams. Also the recharge practices primarily depend on the rainfall pattern of the season overall. And if in any season this pattern swerves than a major offset will be created. Overall the domestic water needs including the drinking water need of the people will be sufficed with the proposed interventions. The quality of water will also be restored with the recharge of open wells. The water demand will be satisfactorily met with the proper community sensitization about the issues and management practices. The community needs to be engaged with some self-help groups for enhancing the knowledge base of the group. The community participation can play an important role to figure out the problems faced by the people. Planning for the people, with the people concept was pursued. The transect walks carried out to sight the generic problems of the people fulfilled the purpose wholly in the case of Nawli. As community knows its problems more than anyone else do. So by means of community-participation the problems of the locals can be sufficed with proper planning and technical skill-sets. Also the people participation in construction works (for checkdams) can be modeled by practices such as „shramdaan‘. Thus the community-based management of water resources can be useful tool to solve the problems of worst hit areas and benefit them with their own managed resources. REFERENCES i. ii. iii. iv. v. vi. vii. viii. ix.

Authority, M. D. (2005). Nawli. Firozepur Jhirkha. C.G.W.B. (2007). Manual on Artificial Recharge of Groundwater. New Delhi: Ministry of Water Resources. Cenus. (2011). census.org.in. Retrieved from census.org.in. CPWD, G. (2009). Rainwater Harvesting and Conservation. New Delhi: Consultancy Services Organisation. IRRAD. (2012). IRRAD compendium . Gurgaon: IRRAD. Manual, N. P. (n.d.). Check Dam: overview. Mohan, R. K. (2003). A Process for Participatory Rural Appraisal. 12. NRDWP, P. F. (2009-2012). Policy Framework. New Delhi: Ministry of Rural Development. WHO. (2012). South-Asia Healthcard.

HYDRO 2014 International

MANIT Bhopal

Page 226

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

SIMPLE MODEL TO ESTIMATE SOIL WATER RETENTION LIMITS OF CHATTISGARH STATE N.G.Pandey1, B. Chakravorty1, Sanjay Kumar2 & P. Mani1 1 Scientists, CFMS, NIH, WALMI Complex, Phulwari Sharif, Patna – 801 505, India 2 Scientist, National Institute of Hydrology, Jalvigyan Bhawan, Roorkee-801505, India ABSTRACT To know available water holding capacity of soil, knowledge of soil water retention limits is essential. This is useful in irrigation scheduling, water balance simulation and land use planning. The measured soil water limits: upper limit at field capacity (-33 kPa pressure) and lower limit at permanent wilting point (-1500 kPa pressure) was available for 67 established soil series of Chattisgarh state. In the present paper, a simple generic power equation has been developed to estimate the soil water limits based on soil survey data such as texture and bulk density. Linear regression was used to estimate the gravimetric soil water limits from sand and clay percentages. The volumetric soil water limits is estimated by multiplying gravimetric soil water limits with bulk density. The predictions were adjusted for coarse fragments and organic carbon present in the soil matrix. The standard error (SE) of the measured water content at field capacity (Wfc) and at permanent wilting point (Wpwp) are 0.98 and 0.54 respectively whereas the SE of the estimated water contents by the developed models are less, 0.89 and 0.51. The percentage error between the estimated and measured W fc and Wpwp found to be 10.1 and 5.8 respectively Comparison made through graphical representation of error bars also shows satisfactory result at (i) ±1.96 (SE) and (ii) ±10% error criteria. Coefficient of variation (CV) also indicates improvements. The goodness of fit (R2) value between the estimated Wfc with measured Wfc, is 0.85, which shows model estimation is reasonable. Similarly, for values of W pwp by model estimation R2 is 0.81. Paired „t‟ test for comparison shows that model is estimating well at 95% confidence level. Key words: gravimetric, volumetric, water content, field capacity, permanent wilting point INTRODUCTION Knowledge of the capacity of soil water reservoir is useful in irrigation scheduling, watershed management, water balance simulation, and land use planning. Field measurements of soil water retention limits are time consuming and tedious. Alternatively soil water retention limits can be estimated from available soil survey data. The drained upper limit is the highest water content of a soil after thorough wetting and draining until drainage becomes practically negligible. The upper limit corresponds closely to water content at field capacity (FC). The FC is a subjective term which depends on the type of soil. For sandy soils the upper limit reaches at suction pressure of 10 kPa whereas in clayey soil it is of 33 kPa. Thus the field capacity

HYDRO 2014 International

ranges from 10-33 kPa suction pressure as per soil type. The lower limit is the lowest water content of a soil after plants stop extracting water as a result of water deficit and finally the plant wilts. The lower limit corresponds closely to water content of 1500 kPa suction pressure, known as permanent wilting point (PWP). Different models used to estimate water retention properties of soil have been developed by Brooks and Corey (1964), van Genuchten (1980), Campbell (1985), Arya and Paries (1981) among others. The input requirement of these models is large and is either to be measured or estimated. The objective of the paper is to develop simple, generic equations to estimate the water retention limits of the soil water reservoir (water retained in between FC and PWP) based on percentage of sand and clay. There are other soil properties that can also be related to estimate the soil water limits. But the presence of more number of predictors (independent variables) makes the model more data specific and thus its generality is lost (Ritchie et al., 1999). The upper and lower limits can be measured in the laboratory using pressure plate apparatus on small soil cores removed from the field. As field measured soil water retention limits data are generally not available, Rawls and Brakensiek (1982, 1985) estimated soil water retention at specific matric potential based on the laboratory measurement. In the present paper, published soil survey data of Chattisgarh state have been used. A simple power relation has been developed using laboratory measured FC and PWP as a function of soil texture and bulk density data. Ritchie et al. (1999) used the database of field measured soil water limits of 401 soil samples from 15 states of US and developed a simple texture based empirical relation (Eq. 1) to estimate field capacity.



W fc  0.186 ( s / c) 0.141  fc  W fc ( b )



(1)

where, Wfc and θfc are the gravimetric and volumetric water content at field capacity (expressed in fraction), s is percentage of sand, c is percentage of clay, and γb is the bulk density of soil. The gravimetric water content was taken to develop the relation to avoid the error associated with measurements of bulk density. To generalize equation (1) he suggested some corrections to water content at field capacity due to the presence of coarse fragments and high organic carbon content. For verification and adaptation of this empirical equation, field measured values of water content for Indian soil condition is required which is not available. MATERIALS AND METHODS The soil survey database of Chattisgarh state, described by Tamgadge et al. (2002) was used. The field work was conducted by National Bureau of Soil Survey and Land Use Planning (NBSS&LUP), Nagpur on 1:250,000 scale maps marked with grids and strips. In total there were 1430 grids (each 10kmX10 km) and total numbers of 6650 soil samples were analyzed by laboratory methods. Random observations of 240 nos. strips

MANIT Bhopal

Page 227

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

(master profile each covering an area of 50 km2) were made. From the analyses of soils of Chattisgarh state, 67 soil series were established in which the laboratory measured water retention data at FC and PWP were available. Each established soil series has depth wise soil layers characterization. The database used in this paper contains 230 values of soil water limits with other corresponding soil attributes such as texture, bulk density, coarse fragments and organic carbon content. The volumetric water content data were converted to gravimetric water content dividing by bulk density. This was made to make the model sensitive to bulk density and also to avoid the error associated with the measurement of bulk density. The general form of the power equation for gravimetric and conversion to volumetric water content corresponding to FC and PWP are:



W fc  a ( s / c) b  fc  W fc ( b )



(2a);



W pwp  a ( s / c) b  pwp  W pwp ( b )



(2b)

Where, s is sand and c is clay in percentage, a and b are fitting parameters of the power equation and γb is the bulk density. This is not the only possible form of the equation. Polynomial or exponential functions of sand and/or clay will not be satisfactory unless it is used with large number of terms. It was assumed that clay content is directly and sand content is inversely proportional to Wfc and Wpwp. By using the ratio sand/clay, we used two pieces of information and at the same time eliminated one parameter (Ritchie et al., 1999). We also assumed non zero values of sand and clay contents. The parameters can be estimated with linear regression on the logarithmic transformation of Eqs.(2a, 2b). 3(a) log W  log a  b log ( s / c) fc

log W pwp  log a  b log ( s / c)

3(b)

where log(a) is the intercept and b is the slope. A regression equation was developed using the above database in which the water content at FC and PWP was equated with texture. Soil matrix comprises with coarse fragments and organic carbon content affects water retention. Accordingly, the following correction factors were incorporated to generalize the relation. (i)

Brandy (1996) has reported the ranges of organic matter content of some inorganic soils viz Alfisol (0.8-6.5%), Vertisols (1.5-3.0%), Ultisols (1.5-4%), Aridisol (0.21.7%). Organic matter content in soil is directly proportional to percentage of clay content (Brandy (1996). Soils with high percentage of clay protects the organic matter from degradation and thus due to presence of more organic matter in clay enriched soils, water holding capacity increases. Hollis et al. (1977), Rawls and Brakensiek (1982) have found the following correction factors due to presence of organic carbon (OC).

 fc oc  fc 0.01*OC

(5a );

 pwp oc  pwp  0.005*OC (5

b)

Where, θfc oc and θpwp oc are the volumetric water contents at FC and PWP respectively adjusted with organic carbon, OC expressed in percentage. In this paper relation on gravimetric basis has been developed in which the correction factor is divided by bulk density. Out of 230 datasets, 220 sets were used to develop the empirical relation and the remaining 10 sets selected from the database at random were used to validate the model result. RESULTS AND ANALYSIS 220 water retention data measured at FC were plotted with the ratio of sand/clay. The data were fitted to the power equation (2a) and the parameters „a‟ and „b‟ were estimated through trend analysis. The estimated value of a=11.188 and b=-0.2538 with R2 = 0.549 (Fig. 1). Similarly, another set of 220 measured water retention data at PWP were plotted. The data was fitted to the power equation (2b) and the parameters a=5.5097 and b= 0.2538 were estimated with R2 = 0.4594 (Fig. 2). In the developed equations (6a, 6b) for Chattisgarh state, the gravimetric water content Wfc and Wpwp are expressed in percentage.





W fc  11 .188 ( s / c) 0.2538 (6a  fc  W fc ( b ) );



W pwp  5.5097 ( s / c) 0.2915 (6  pwp  W pwp ( b ) b)



Coarse fragments (>2 mm in size) present in the soil matrix affect in reducing porosity and thus it also affect the soil water retention limits. Necessary correction in water retention at FC and PWP was done multiplying Wfc and Wpwp with (1-C/100) respectively, where C is the percentage of coarse fragments by weight (Rawls et.al.1992).

W fc cf W fc (1C /100)(4a) Wpwp cf Wpwp (1C /100)(4b ;

(ii)

)

The model was adjusted for high organic carbon content. Organic matter is 1.74 times of organic carbon (Barua 1998). Maximum organic matter accumulation takes place in the top 20 cm of the soil and it decreases with depth.

HYDRO 2014 International

MANIT Bhopal

Page 228

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 1: Scatter plot of measured Wfc and fitted model parameters

The standard error (SE) of the mean is the standard deviation divided by the square root of the sample size. From the descriptive statistics (Table 2) it was found that the standard error of the measured water contents Wfc and Wpwp are 0.98 and 0.54 respectively. In the model estimation, the standard error has reduced to 0.89 and 0.51. The percentage error between the estimated and measured Wfc and Wpwp founds to be 10.1 and 5.8 respectively. Graphical representation of error bars (Figs. 3 and 4) was used for comparison of estimated model values with measured Wfc and Wpwp. Fig. 3 shows that if ±1.96 times standard error is added to the estimated value of the model, it comes within the range of measured values, thus the model estimation is within 95% confidence level. Similarly, Fig. 4 shows that the estimated error of the model is well within ±10% variation.

Figure 2: Scatter plot of measured Wpwp and fitted model parameters Each measured water content data was converted to gravimetric water content by dividing with bulk density. Before fitting, necessary corrections in the original dataset due to presence of coarse fragments and organic carbon contents in the soil matrix were made using Eqs. (4a, 4b) and (5a, 5b). Hence the empirical model (Eqs. 6a, 6b) so developed is made generic and simple. Out of the 230 datasets, 220 sets were used to develop the empirical relation and the remaining 10 sets, selected from the database at random were used to validate the model.

Coefficient of variation (CV) of estimated values also showed some improvements over measured values. Analyses of skewness revealed that the positively skewed distribution has been improved by the model estimation but with flatter curve (kurtosis). The goodness of fit criteria was used for comparing estimated Wfc with measured Wfc. The R2 value of 0.85 shows model estimation is appropriate. Similarly for W pwp , R2 value is 0.81 indicating reasonable model estimation.

Table 1: Estimated soil water limits of the fitted model

Table 2: Descriptive statistics and paired t-test

Sand=2.0-0.05 mm; Silt=0.05-0.002 mm; Clay 0.85 is excellent for a hydrological model, values between o.65 – 0.85 are very good, o.5 – 0.65 are good, 0.20 – 0.50 are poor and < 0.20 are very poor. 4 RESULTS AND DISCUSSION The Kadalundi river basin mostly occupied by forest, plantation, and grassland followed by agriculture, urbanized area and water body throughout the study period (Figure 4; Table 2). The LULC changes for Kadalundi River basin from 1988 to 2000 were shown in figure 4 and over all amounts of change (%) and percentage growth in LULC is shown in Table 2. On the basis of amount of change detection and percentage growth the LULC map throughout the study period (1988 and 2000) indicates that most significant change occurs in urban, forest, grassland and agriculture area out of seven classes. Agriculture, urban, plantation and water area increases however forest and grassland decreases but there is no significant change occurs in bare area of this basin. In Kadalundi basin, LULC change from 1988 to 2000, proportional extent of urban are increase by 3.31% to 5.80%, agriculture by 6.92% to 9.49% and water body by 4.17% to 5.28% and plantation 19.73% to 28.83%and decrease of forest 36.46% to 30.83% and grassland 29.21 to 19.57%.

Figure 4: Land use /land cover classification of 1988 and 2000 of Kadalundi River Basin. The area of agriculture and plantation increased (by 28.56 km2 and 101.15 km2 respectively (Table 2)) due to population growth and most of grassland and forest parts converted into agriculture and plantation area. In addition, agriculture and urban area expanded near the bed of river (slope 0-5% mostly) which accelerate the removal of grassland and forest near these areas. On the basis of percentage growth analysis there is negative percentage growth found in forest (-15.36%) and grassland (33%) but all other classes shows positive growth (Table 2). The percentage growth of urbanization increased by 75.23 % which is higher to all other LULC classes throughout the study period.

MANIT Bhopal

Page 236

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

On the other hand no change occurs in bare land area. The increase of urbanization is due to increase of population growth and conversion of agriculture to urban and in the study area. Table 2. Area (Km2) and overall amount of change (%) in LULC of study area over the period of 1988-2000.

Figure 6(a) Precipitation, observed discharge, simulated discharge of Kadalundi River basin during the period of 1988 to 2000 with their trend impose by using 2000 LULC and (b) correlation between observed and calibrated/validated discharge. The simulation and calibration/validation results of LULC change impacts on surface runoff at the sub-basinal scale have been estimated using SWAT model and how the runoff at the outlet of the River basin changes for the rainfall by 1988 to 2000. The comparison between simulated and observed monthly stream-flow values in the periods of January 1988 to December 2000 based on LULC 1988 is shows in the figure 5a, similarly simulated monthly stream flow for the period of 1988 to 2000 for 2000 LULC given in figure 6a. Further the results and calibration from January 1988 to December 1995 and validation for January 1996 to December 2000. There is a good correlation found between calibrated/validated and observed value. R2 values for the calibration and observed are 0.896 and validation is 0.870 for LULC 1988 (Figure 5a) while overall R2 values is 0.843(Figure 5b) whereas R2 values for the calibration and observed are 0.896 and validation is 0.860 for LULC 2000 (Figure 6a) while overall R2 values is 0.842(Figure 6b) for monthly stream flow. These results suggest that model performance are good.

Figure 5(a) Precipitation, observed discharge, simulated discharge of Kadalundi River basin during the period of 1988 to 2000 with their trend imposed by using 1988 LULC and (b) correlation between observed and calibrated/validated discharge.

HYDRO 2014 International

On the basis of the results (observed and calibrated/validated) presented in this study, LULC affect river discharge throughout the study period. Various other parameters also can changes discharge on basin scale for example like climate change. Noorazuan et al. (2003) detected that urbam extent and changes in urban related LULC cloud affect river stream flow behavior by increasing the surface runoff in the Langat river basin, Malaysia, South –east Asia. Shi et al. (2007) studied the effect of LULC change on surface runoff in the Shenzhen region, China and concluded that urbanization led to increases in the maximum flood discharge. The results shows that a small but increase in discharge continuously by average 1.37m3/sec from 1988 LULC to 2000 LULC and average increase of 3.68 % from 1988 to 2000. The small change in discharge is increased due to increase of urbanization near river bed and plantation (managed forest) in the study area. 5 CONCLUSIONS Uses satellite images collected from remote sensing for studying LULC classification, change detection and percentage growth are helpful for hydrological impact study. The LULC change has a major role in the change of surface runoff. Increase in surface runoff was observed for the Kadalundi River Basin for the LULC change from 1988 to 2000. A slight increase observed in long term discharge (12 years) with change in LULC. It may be due to increase in urban area and decrease in forest area and also location of LULC change. The drastic change in LULC may affect other environmental stress through generating more sediment yield and erosion that were usually directly related to runoff volume and velocity. Thus, urbanization and deforestation is the strongest contributor for surface runoff which can be considered as a major environmental stress controlling the hydrological parameters such as runoff, water yield, and ET, for Kadalundi River Basin. The approach used in this study estimates contributions of changes for LULCs to surface runoff which provide quantitative information for decision makers to make better choices for land and water resource. Lastly, present approach provides a solid example of integrating hydrologic

MANIT Bhopal

Page 237

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

modeling (using remotely sensed digital LULCs) to understand the potential impact of landscape change on the water resource, a vital ecosystem of the Western Ghats region in India. The present approach can be used to different river basin, where time-sequenced digital Landcover is available, to predict the change in surface runoff to LULC changes. REFERENCES i. Arnold, J.G., Srinivasan, R., Muttiah, R.S., Williams, J.R., 1998. Large area hydrologic modeling and assessment – Part 1: model development. Journal of American Water Resource Association 34 (1), 73–89. ii. Hanriksen, H., Troldborg, L., Nyegaard, P., Sonnenborg, T., Refsgaard, J. and Madsen, B., (2003).Methodology for construction, calibration and validation of a national hydrological model for Denmark. Journal of Hydrology, 280, 52-71. iii. Liem T. Tran, Robert V. O‘Neill. 2013. Detecting the effects of land use/land cover on mean annual streamflow in the Upper Mississippi River Basin, USA. Journal of Hydrology 499. 82–90. iv. Lu D, Weng Q. 2007. A survey of image classification methods and techniques for improving classification performance. International Journal of Remote Sensing 28: 823–870. v. Neitsch, S.L., Arnold, J.G., Kiniry, J.R., Williams, J.R., 2005. Soil and Water Assessment Tool. Theoretical Documentation. Version 2005, USDA-ARS, Temple, TX, USA. 494 pp. vi. Noorazuan MH, Rainis R, Juahir H, Sharifuddin, Zain M, Jaafar N. 2003. GIS application in evaluating land use-land cover change and its impact on hydrological regime in Langat River basin, Malaysia. Proceeding of the 2nd Annual Asian Conference of Map Asia, October 2003, Kuala Lumpur, Malaysia. vii. Petchprayoon Pakorn, Peter D. Blanken, Chaiwat Ekkawatpanit, and Khalid Hussein. 2010 Hydrological impacts of land use/land cover change in a large river basin in central–northern Thailand. Int. J. Climatol. 30: 1917–1930. viii. Shi PJ, Yuan Y, Zheng J, Wang Jing-Ai GY, Qiu GY. 2007. The effect of land use/cover change on surface runoff in Shenzhen region, China. Catena 69: 31–35. ix. Tang, L., Yang, D., Hu, H., Gao, B., 2011. Detecting the effect of land-use change on streamflow, sediment and nutrient losses by distributed hydrological simulation. J. Hydrol. 409, 172–182. x. The NCEP/NCAR 40-year reanalysis project, Bull. Amer. Meteor. Soc., 77, 437-470, 1996. xi. USDA, S.C.S., 1972. National Engineering Handbook Section 4 Hydrology. xii. Wenming Nie, Yongping Yuan, William Kepner, Maliha S. Nash, Michael Jackson, Caroline Erickson. (2011) Assessing impacts of Landuse and Landcover changes on hydrology for the upper San Pedro watershed. Journal of Hydrology 407 105–114.

Impact Of Land-Use Land-Cover Changes On Runoff Generation In A Bangalore Urban Catchment R. L. Gouri1 V. V. Srinivas2 Research Scholar, Department of Civil Engineering, Indian Institute of Science, Bangalore-560 012, India 2 Associate Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore 560012, India Email:[email protected] 1

ABSTRACT The process of urbanization is often rapid and disorganized in developing countries like India. Owing to this, existing natural

HYDRO 2014 International

drainage system gets affected, as storm sewer network might replace only a part of the natural system. Further, changes in land use /land cover (LULC) associated with urbanization of catchments has implications on generated runoff in terms of increase in peak discharge, runoff volume and velocity. Increase in runoff velocity causes reduction in time to peak discharge, resulting in flash floods. In addition, encroachment into storm sewers limits their capacity to convey runoff, causing more devastation during floods due to increase in stage for a given discharge. The present study is aimed at assessing LULC change in an urban catchment located in Bangalore (North) and evaluating its effect on runoff generation. Remote sensing satellite images corresponding to the years 1996, 2002, 2006 and 2012 have been analyzed to identify changes in LULC. Impact of the changes on runoff is investigated by inputting design hyetographs to Storm Water Management Model (SWMM) developed for the urban catchment and analyzing the runoff generated by the model. From the foregoing analysis it was observed that there is a significant increase in built-up area and corresponding reduction in agricultural area. This has resulted in increased runoff in most of the sub-catchments of the study area. Vulnerable reaches in existing storm water drainage network of the studied urban catchment are identified. Keywords:Land use/Land cover(LULC), generation, Storm runoff, Bangalore

SWMM,

Runoff

1. INTRODUCTION: Most of the cities in India are undergoing rapid development in recent decades, and many rural localities are undergoing transformation to urban hotspots. These developments have associated land use/land cover (LULC) change that effects hydrological response (e.g., evapotranspiration, infiltration and runoff) from catchments,. The effects are often evident in the form of increase in runoff peaks, volume and velocity in drain network. There is a need to account for LULC change information in studies related to urban watershed management or development, and to model such changes for use in efficient design of storm water conveyance structure. Often, in most of the urban localities, the existing storm water drains are in dilapidated stage owing to improper maintenance or inadequate design, which causes flooding and consequent damage to property. Most studies in the past have assessed effect/impact of LULC change on (i) water balance components in agricultural watersheds (Schilling, et al. 1998); and (ii) runoff generation in river basins (Niehoff, et al., 2002; Hundecha&Bardossy, 2004; Tang, et al., 2005; Rongrong&Guishan, 2007; Liu et al., 2011). They have concluded that reduction of vegetation cover leads to increase in runoff and vice versa, which is the case in urban catchments. There is a stressing need for studies on impact assessment of LULC change on runoff generation in urban catchments in India, as most cities are witnessing rapid development. However, hardly there have been any attempts to study effect of land use change. Recently Nagarajan and Poongothai (2011) assessed effect of LULC change in Manimuktha sub-watershed of Vellar basin, Tamilnadu, India.

MANIT Bhopal

Page 238

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The current study is motivated to assess effects of LULC change on runoff generation in an urban catchment in Bangalore, India. The LULC changes were examined based on remote sensing products usingGeographic Information System (GIS) tools, and effect of LULC change on runoff generated from the catchment was assessed using Storm Water Management Model (SWMM). Significant change in LULC was observed especially during recent years and it was reflected in the generated runoff volume. The remainder of the paper is structured as follows. A case study is presented on storm water distribution system in Yelahanka zone of Bangalore city in the following section. Subsequently conclusions drawn based on the study are provided. 2 CASE STUDY: 2.1 Study area and Data description: The study area considered for assessing LULC change on hydrological response from catchments is located in northern zone (Yelahanka) of Bangalore, India, (Figure 1). Length of storm water drain network in the study area is about 18 km. The network is divided into 19 conduits that are shown in Figure 1. Area of watershed contributing flow to the network is about 70 km2. The watershed is divided into 34 sub-watersheds (Figure1).

Figure 1: Study area For use in the study, daily rainfall records over the period 19882010 corresponding to two rain gauges located at Hebbal and Gandhi KrishiVignana Kendra (GKVK) Agricultural College were collected. Locations of the gauges are shown in Figure1.Design hyetographs were constructed based on Intensity-duration-frequency (IDF) curves developed for the study area. As the available records at Hebbal and GKVK

HYDRO 2014 International

agricultural college were at daily scale, they had to be disaggregated to hourly scale for construction of design hyetographs corresponding to sub-daily durations. The disaggregation was carried out using k-nearest neighbor method (Anandhi et al. 2012). For this purpose, relationships between daily and hourly rainfall records corresponding to a rain gauge located at Indian Institute of Science (IISc) campus were considered. The gauge at IISc had 10-min scale rainfall records over the period 2003-2012, and they were found to be reasonably well correlated at daily scale with records corresponding to gauges located at Hebbal and agricultural college. For each of the stations IDF curves were developed using the derived hourly rainfall by frequency analysis using Extreme value type-I distribution. Subsequently those curves were considered as the basis to construct design hyetographs corresponding to various sub-daily durations and return periods for the locations Hebbal and agricultural college by using alternate block method (Chow et al 1988). Herein, it is to be mentioned that records corresponding to gauge at IISc were short, and hence those were not considered for frequency analysis to construct IDF curves. For the study area LULC maps were prepared corresponding to the years 1996, 2002, 2006 and 2012. Among those, maps corresponding to the years 1996 and 2002 were based on satellite imageries obtained from IRS-IC-LISS-III at 1:50,000 scale, whereas the maps for the years 2006 and 2012 were prepared based on thematic maps available for the study area fromBhuvan-Thematic Services National Remote Sensing Centre (NRSC), ISRO, Hyderabad, India, (http://bhuvan.nrsc.gov.in) at 1:50,000 scale. A level II classification was carried out to classify land use in the study area corresponding to each of the LULC maps. This resulted in identification of 12 different classes of land use and land cover. Investigations were carried out to detect changes in the LULC classes for locations in the study area over the years 1996, 2002, 2006 and 2012. For this purpose, transition matrix indicating LULC change between various map pairs was determined using IDRISI software that has land-use change modeler for land-use change analysis and prediction. The considered map pairs included 1996-2002, 2002-2006, 2006-2012, and 1996-2012. The transition matrices were analyzed to draw inferences on LULC change. Hydrologic Soil Groups (HSG) were identified for the study area based on soil map obtained from National Bureau of Soil Survey and Land Use Planning (NBSS&LUP, Prasad et al., 1998). Each of the LULC maps was merged with HSG map and Weighted Curve Number (WCN) corresponding to each of the 34 subwatersheds was computed following the procedure described in Chow et al. (1988). The foregoing analysis was carried out using ARC-GIS 9.1 software. Discharge from the entire watershed, which contributes flow to the network, was estimated corresponding to various rainfall durations (e.g., 1h, 3h) and return periods (e.g., 2-year, 5-year) using Storm Water Management Model (SWMM) (Rossman, 2010). Inputs to SWMM included design hyetograph for the duration of interest, land-use/land-cover and soil details. Further, estimation of initial abstractions in the model was based on Natural Resources

MANIT Bhopal

Page 239

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Conservation Service (NRCS) curve number method. Streamflow generated by SWMM was considered as the basis to identify vulnerable conduits in the network. 3 RESULTS: The LULC maps prepared corresponding to the years 1996, 2002, 2006 and 2012 were analyzed for detection of changes in various land-use classes in each of the sub-watersheds. It was observed that LULC has changed during the entire period and the changes were significant during 2006-2012. Figure 2 presents LULC map of the study area prepared for the year 2012. Maps corresponding to the remaining years are not shown due to lack of space. Comparison of LULC classes across the years 1996, 2002, 2006 and 2012 corresponding to 12 level II classification is shown in Figure 3. It can be observed from the figure that between the years 1996 and 2012 crop land has decreased from approximately 25 km2 to 13 km2. Similar reduction is also evident in areas corresponding to forest, forest plantations, lakes, grassland and fallow lands. In contrast, builtup urban area has significantly increased (from 18 km2 to 33 km2) and a relatively lower increment is evident in built-up rural area.

HYDRO 2014 International

MANIT Bhopal

Page 240

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 other sub-watersheds. This increase in WCN can be attributed to transition from crop land to urban built-up area, which was observed from LULC maps. Changes in LULC were relatively minimal for the remaining sub-watersheds. Interesting LULC changes in sub-watershednumbered 14 were insignificant. It is to be noted that the WCN has increased significantly between the years 2006 to 2012 when compared with the earlier years. .

1996

2002

2006

2012

Curve number

100

90

80

70

Figure 2: LULC map of the study area for the year 2012

0

5

10

15

20

25

30

35

Subwatershed number 35

Figure 4: Weighted Curve Number corresponding to the years 1996, 2002, 2006 and 2012.

1996 2002 2006 2012

Area (km2)

30 25 20 15 10 5 Mining

Barrenland

Fallowland

Grassland

Wetland

Lake

Rural

Urban

Forest Plantation

Forest

Crop Plantation

Crop Land

0

Figure 3: Change in LULC The LULC maps corresponding to the years 1996, 2002, 2006 and 2012 were merged with the HSG map to derive WCN corresponding to each of the 34 sub-watersheds in the study area. Figure 4 depicts WCN for 34 sub-watersheds in the study area corresponding to each of the years. On comparing the WCN across the years 1996 to 2012, it can be observed that WCN for all the sub-watersheds have increased. The sub-watersheds numbered 16, 18, 22 to 26 and 32 (Figure 1) have shown significant change (>10) in the value of WCN as compared to

HYDRO 2014 International

Experiments were carried out to examine effect of LULC change on water balance components such as evaporation, infiltration and runoff. Each experiment involved providing a design hyetograph as input to SWMM along with LULC information corresponding to each of of the four years (1996, 2002, 2006 and 2012), and analyzing output from the model to quantify changesin each of the water balance components due to LULC change. To quantify effect of design hyetograph on the results, hyetographs corresponding to various durations and return periods were developed, and one of those hyetographs was considered at a time for performing an experiment. Changes in water balance componentsdue to LULC change was quantified corresponding to each of the hyetographs. One of the typical results obtained from these experiments is presented in Figure 5. Comparison of runoff generated for different years shows increase in runoff and consequentreduction in infiltration. Reductionin infiltration and increase in runoffcould be attributed to increase in imperiousness owing to increase in built-up area. On the other hand, change in evaporation was found to be insignificant, possibly due to insignificant changes to factors (e.g., temperature, humidity, wind speed) effecting the component. For brevity, the distribution of catchment response to a design hyetograph corresponding to 3-h duration and 5-year return period is shown in Figure 5.

MANIT Bhopal

Page 241

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Evaporation

Infiltration

2002

2006

2012

50

40 30 20

10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Sub-watershed number

Figure 6: Volume of runoff generated corresponding to 1h and 3h duration rainfall of 2year and 5year return period for each of the 34 sub-watershed.

Runoff

Duration 3-hr of 2-year return period

100

1996 2002 2006 2012

80

80 70

60

60 Runoff (103 m3)

Water balance components (mm)

120

1996 60

Runoff (103 m3)

Figure 6 shows effect of LULC change on runoff generated from each sub-watershed in the study area by SWMM model corresponding to a few typical design hyetographs. It can be observed from the figure that rainfall with higher return period generatesmore runoff, as expected. Comparison across the years 1996, 2002, 2006 and 2012 showed that the volume of runoff has increased consistently from the year1996 to the year 2012 across all sub-watersheds. The increase is significant during 2006-2012 for sub-watersheds numbered 16, 18, 22 and 26 across all hypothetical rain events (design hyetographs). These sub-watersheds were earlier found to have marked increase in their WCN (Figures4 and 6). Similarly, no significant changes were noted in runoff from sub-watershed numbered 14 for which changes in LULC were previously found to be insignificant.(Figures 4 and 6).

40 20

50 40 30 20 10

0

0

LULC of 1996

LULC of 2002

LULC of 2006

LULC of 2012

Figure 5: Water balance componentsquantified (in depth units) corresponding to 3h duration 5-year return period design hyetograph.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Sub-watershed number

Duration 3-hr of 5-year return period 1996

2002

2006

2012

90

80 70

1996

2002

2006

Runoff (103 m3)

Duration 1-hr of 2-year return period 2012

60

Runoff (103 m3)

50

60 50 40 30 20

40

10

30

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Sub-watershed number

20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Sub-watershed number

Duration 1-hr of 5-year return period

HYDRO 2014 International

Figure 6 (continued…): Volume of runoff generated corresponding to 1h and 3h duration rainfall of 2year and 5year return period for each of the 34 sub-watershed Storm water conduits numbered 11, 12, 13, 14, 17, 18 and 19 (shown in Figure 1) were identified to be vulnerable reaches in the storm water drainage network being studied, as they were found to be inadequate to convey discharge corresponding to various design hyeographsand LULC past scenarios generated by SWMM. The identified vulnerable conduits are located immediately to the downstream of watersheds numbered 16, 18,

MANIT Bhopal

Page 242

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

22 to 26 and 32 where marked increase in imperviousness was noted in the recent years. 4

SUMMARY AND CONCLUSIONS:

Changes to LULC in sub-watersheds contributing flow to storm water distribution system in Yelahanka zone of Bangalore city was examined. The changes were found to be significant for subwatersheds numbered 16, 18, 22 to 26 and 32, and insignificant for sub-watershed numbered 14.Transition from crop land to urban built-up area was evident, especially during 2006-2012, indicating increase in imperviousness (curve number). Investigations on changes in water balance components due to LULC change over the study area indicated reduction in infiltration and increase in runoff. Increase in volume of runoff was found to be significant for the watersheds for which increase in imperviousness (curve number) was high. Vulnerable reaches in the storm water drainage network were also identified, and the vulnerability was attributed to LULC changes in sub-watersheds that are located immediately to the upstream of those conduits. REFERENCES i. Anandhi A, Srinivas VV, Kumar DN, Nanjundiah RS (2012) Daily relative humidity projections in an Indian river basin for IPCC SRES scenarios. TheorApplClimatol 108(1-2): 85-104 ii. Chow VT, Maidment DR, Mays LW (1988) Applied hydrology. McGraw-Hill, NY iii. Eastman, J.R., 2009. IDRISI Taiga (Worcester, MA: Clark University). iv. ESRI 2011. ArcGIS Desktop: Environmental Systems Research Institute.

Release

10.

Redlands,

CA:

v. Hundecha, Y., &Bárdossy, A. (2004). Modeling of the effect of land use changes on the runoff generation of a river basin through parameter regionalization of a watershed model. Journal of Hydrology, 292(1), 281-295. vi. Liu, Z., Yao, Z., Huang, H., Wu, S., & Liu, G. (2012). Land use and climate changes and their impacts on runoff in the yarlungzangbo river basin, China. Land Degradation & Development. vii. Nagarajan, N and PoongothaiS (2011) Trend in Land Use/Land Cover Change Detection by RS and GIS Application. International Journal ofEngineering and Technology 3 (4): 263-269 viii. Niehoff, D., Fritsch, U., &Bronstert, A. (2002). Land-use impacts on storm-runoff generation: scenarios of land-use change and simulation of hydrological response in a meso-scale catchment in SW-Germany. Journal of Hydrology, 267(1), 80-93. ix. Rongrong W, Guishan Y, (2007), Influence of land use/cover change on storm runoff—A case study of Xitiaoxi River Basin in upstream of Taihu Lake Watershed, J Chinese Geographical Science, 17(4): 349-356 x. Rossman AL (2010) Storm Water Management Model, Version 5: User‘s Manual, EPA/600/R-05/040. Environ Res Lab, EPA, and Athens, Georgia. xi. Schilling, K. E., Jha, M. K., Zhang, Y. K., Gassman, P. W., &Wolter, C. F. (2008). Impact of land use and land cover change on the water balance of a large agricultural watershed: Historical effects and future directions. Water Resources Research, 44(7):1-12 xii. Prasad CRS, Reddy RS, Seghal J, Velayutham M, 1998. Soils of Karnataka for optimising land use. NBSS Publ. 47b (Soils of India series), National Bureau of Soil Survey and Land Use Planning, Nagpur, India. xiii. Tang, Z., Engel, B. A., Pijanowski, B. C., & Lim, K. J. (2005). Forecasting land use change and its environmental impact at a watershed scale. Journal of environmental management, 76(1), 35-45.

HYDRO 2014 International

Change Detection In Land Use/Land Cover Using Remote Sensing And Gis – A Case Study For Ur Basin In Tikamgarh District S. Karwariya1* S. Goyal2 V. C. Goyal3 T. Thomas4 1: Sr. Project Officer, NIH-TIFAC-DST Project, NIH-Regional Centre, Bhopal. 2: Sr. Scientist & Head MPRA Division, M.P.Council of Science & Technology, Bhopal. 3: Scientist-F, National Institute of Hydrology, Jalvigyan Bhavan, Roorkee. 4: Scientist-D, National Institute of Hydrology, Regional Centre, WALMI Campus, Bhopal Email*: [email protected] ABSTRACT : Land use and land cover is an important component to understand basin level land status as it shows the present as well as past status of the earth‟s surface. The land use/land cover pattern of a region is an outcome of natural and socio-economic factors and its utilization gives information about the human livelihood and development. Land use and land cover are two separate terminologies which are often used interchangeably (Dimyati et al 1994). Land cover is a basic parameter which evaluates the content of earth surface as an important factor that affects the condition and functioning of the ecosystem whereas the land cover is a biophysical state of the earth surface, which can be used to estimate the interaction of biodiversity with the surrounding environment. Land use and land cover is dynamic in nature and provides a comprehensive understanding of the interaction and relationship of anthropogenic activities with the environment (Prakasam, 2010). As land is becoming a scarce resource due to immense agricultural and demographic pressure, therefore the information on land use/land cover and possibilities for their optimal use is essential for the selection, planning and implementation of land use schemes to meet the increasing demands for basic human needs and welfare. Bundelkhand region in Central India is in the limelight due to continuous droughts leading to large scale out migration, degraded forests and lands, high soil erosion, reduction in soil productivity and crop yields and has one of the lowest socioeconomic indicators in the country. The Ur river basin, a major tributary of River Dhasan with a catchment area of 990.37 sq. km has been selected for the study. The land use change detection has been performed based on the analysis of the digital data of LISS III & IRS-P6 with a resolution of 23.5m pertaining to 2003-04 and 2011-12. It has been observed that there has been a considerable change in the land use pattern with an increase of 55.02 % in the area under kharrif crop whereas the area under double crop has increased drastically by 136.17%. The increase in area under kharrif crop and double crop has been at the expense of land with scrubs which had a decrease of 73.4%. Even though there is slight change in the acreage under forest.

MANIT Bhopal

Page 243

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Keywords: Land use, Land cover, Change Detection, LISS III & IRS-P6, DSS, Watershed, GIS.

of land that can be sustained and in short the whole population and socio-economic determinants.

1. INTRODUCTION

1.3 Study area

Land is the basic natural resource that provides habitat and sustenance for living organisms, and a major source of economic activities. The increasing population and economic activities are putting pressure on the available land resources. According to Census 2011 the population growth rate of India is 17.66. The population is increasing day by day so the land use/land cover areas also change. Land use relates to the human activity or economic function associated with a specific piece of Land (Lillesand et al. 2004). Examples of land use include agriculture, urban development, grazing, logging, and mining. In contrast, land cover relates to the composition and Characteristics of land surface elements (Cihlar 2000).

The study area represents the typical topography and geology of the Bundhelkhand region and moreover large number of tanks is alo present here. Ur river basin is situated in Tikamgarh district of Madhya Pradesh and lies on the Bundhelkhand Plateau between the Jamni, a tributary of the Betwa and the Dhasan rivers. It extends between latitudes 24°35‟0” N and 25°05‟0” N and between 78°50‟0” E and 79°10‟0” E longitudes. The total geographical area of basin is 990.37 sq. km. The total population is 3, 32, 998 based on the 2011census. The Ur River flows in a south to north direction. Ur river basin is bounded by Chhatarpur district to the east, Lalitpur district to the west; Jhansi district to the north and Sagar district to the south. The location map of the study area is given in Figure 1.

Historically humans have been modifying land to obtain the essentials for their survival, but the accelerated rate of exploitation has brought unprecedented changes in ecosystems and environmental processes at local, regional and global scales. The prime aim of these land use/land cover changes including land conversion from one type to another and land-use management is to satisfy mankind‟s immediate demands i.e., the need to provide food, water, and shelter (Arshad Amin and Shahab Fazal (June 2012) Land Transformation Analysis Using Remote Sensing and GIS Techniques , Journal of Geographic Information System, 2012, 4, 229-236).The land use/land cover pattern of a region is an outcome of natural and socio-economic factors and their utilization by man in time and space. Land is becoming scarce resource due to immense agricultural and demographic pressure. Hence, information on land use/land cover and possibilities for their optimal use is essential for the selection, planning and implementation of land use schemes to meet the increasing demands for basic human needs and welfare. This information also assists in monitoring the dynamics of land use resulting out of changing demands of increasing population. Land use and land cover change has become a central component in current strategies for managing natural resources and monitoring environmental changes. Watershed is the smallest sized hydrologic unit. It is an ideal base for broad level planning of most of the land and water development programme. Change detection in watersheds helped in enhancing the capacity of local governments to implement sound environmental management. This involves development of spatial and temporal database and analysis techniques.

The climate of study area is semi-arid and the year comprises of four seasons. The winter season from December to February followed by the hot season from March to mid-June; .rainy season from mid-June to September end and the transition season from October to November. After February the temperature rises progressively. May is generally the hottest month with mean daily maximum temperature at about 43 0C sometimes may even rise up to about 47 0C on individual days. The relative humidity is high during the monsoon season being generally above 70 percent. In the rest of the year the air is comparatively dry. The driest part of the year is summer season when the relative humidity is less than 20 percent. The topography of basin is undulating and comprises of very high hills along the ridge line with the elevation varying between 200 m to 400 m above mean sea level. The elevation gradually decreases from the southern part of the basin towards the north. Therefore the Ur river also flows in a north-east direction till its confluence with Dhasan river. 2. MATERIAL AND METHODS 2.1 Data Products In the present study multidate satellite data of 2003-04 and 2011-12 (LISS III & IRS-P6) were used.

1.2 Need for mapping Change Detection in Land Use/Land Cover The whole part of the earth like hills, different types of lands, water bodies, and built-up area comes under the land use land cover. Land has been going through tremendous transformations due to growth in agriculture, industrialization and urbanization. The changes in land use affect the ecosystem in terms of land cover, land quality and capability, weather and climate, quantity

HYDRO 2014 International

MANIT Bhopal

Page 244

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 approach i. e. remote sensing and GIS in conjunction with secondary data has been adopted. The remote sensing and GIS data were handled with the help of Erdas Imagine 9.3 and Arc GIS Desktop 9.3 respectively. The detailed methodology of the study is given below:

Figure 1: Location Map of Study Area 2.2 Ground truth verification: Entire study area was visited to get an acquaintance of different ground feature and cover type with respect to satellite data. The doubtful area on preliminary interpreted maps from satellite data were carefully verified in the field. After verification, these areas were reconciled on the maps and corrections were made to obtain final maps. Figure 2: Flow chart showing methodology of landuse/landcover change detection.

2.3 Visual analysis: Visual interpretation technique was used for the mapping of land use/land cover. Prior to interpretation of multidate satellite data, a reconnaissance survey of the study area was done to develop a classification scheme based on local knowledge and ancillary information. An interpretation key was also developed based on standard Photo-elements like tone, texture, size, shape, association, pattern, location etc. to identify and map different classes. With the help of interpretation key on screen preliminary interpretation of satellite data was done using ERDAS IMAGINE 9.3 software.

2.5 Overlay analysis, final map preparation and area statistics generation:

2.4 Methodology

Out of the total geographical area of 990.37 km2, agriculture constitutes 579.98 km2 in 2011-12, which was 399.69 km2 in 2003-04. Major cause of this unprecedented increase in area under agricultural use was transformation of wasteland into agriculture. Landuse map clearly shows that about 85- 90 % of wasteland area converted into agricultural land. The main reason of this land transformation is massive growth of population. It has been also observed that there has been a considerable change in the land use pattern with an increase of 55.02 % in the area under kharrif crop whereas the area under double crop has increased drastically by 136.17%. The increase in area under

The study is mainly based on primary sources of data, the data used for the preparation of Land use/land cover map is multidate multispectral satellite data of IRS LISS III & IRS-P6 of the year 2003-04 and 2011-12 on 1:50,000 scale. Similarly, the secondary data were used for validation of land use/land cover maps. A variety of remote sensing change detection methodologies have been developed and evaluated over the past twenty years (Rogan et al., 2002; Woodcock & Ozdogan, 2004; Healey et al., 2005). For this study an integrated geo-spatial

HYDRO 2014 International

Union method is applied for Overlay analysis using Arc Map 9.3 Software. The final maps were prepared after reconciliation of doubtful areas observed in preliminary maps. The final maps were prepared/ composed and area statistics was generated using Arc Map 9.3 Software. 3. RESULTS AND ANALYSIS

MANIT Bhopal

Page 245

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

kharrif crop and double crop has been at the expense of land with or without scrub/ waste land which had a decrease of 70.4%. The cropping pattern in the study area which had the practice of Rabi cultivation only, has now changed with more farmers adopting the soybean crop in the kharrif season along with wheat in the Rabi season. Even though there has been slight change in the acreage under forest, the area under fellow land has increased from 21.54 to 69.65 sq. km, a 223 % increase.

Figure 5: Statistics of land use/land cover during year 2003 & 2011. Table 1: Area statistics of land use/land cover distribution in Urr river basin during year 2003-04, and 2011-12

Table 2: Land use/land cover changes during year 2003-04, and 2011-12

Figure 3: Land use/land cover-2003-04 use/land cover-2011-12

Figure 4: Land

4. CONCLUSIONS

HYDRO 2014 International

MANIT Bhopal

Page 246

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The temporal change in land use /land cover was detected through preparation of land use/ land cover maps pertaining to 2003-04 and 2011-12 using multidate satellite data .The prepared maps were overlaid using GIS to obtain Change Detection maps to know the changes occurred in different land use classes during 2003-2011. The following conclusions are derived from this study: 1. Remotely sensed data especially satellite data can be effectively used in mapping as well as monitoring of temporal changes in land use/land cover of an area. 2. A considerable change in different land use/land cover categories was observed during 2003 to 2011 especially cropland and wastelands. Changes in forest sub-classes were also noticed. 3. The overall increase in built-up area was 17.5 % during 2003 to 2011. 4. Croplands also registered an overall increase of 45.3 % in area during 2003-2011.The increase in area was as a result of transformation of Land with or without Scrub/Waste land into cropland. 5. In case of forest, the land transformation in dense, open and scrub was also observed for the period 2003-2011. During this period dense forest converted into open forest and open forest into scrub forest. The study with the help of Geographic Information System (GIS) and Remote Sensing technique is very useful tool for Landuse/Landcover change detection /land transformation mapping. The measurement of land use/ land cover change is very useful for future management and planning at local and regional level. Finally, although human demands cannot be stopped, with proper management and planning it can be restricted and directed in a desirable and sustainable way, REFERENCES: i. Bansal, A., Karwariya S. and Goyal S. (2012) Change Detection in Land use / Land cover in Sewan Watershed Using Remote Sensing and GIS Technique. Int. Journal of Advances in Remote Sensing and GIS, Vol. 1, No. 2, 2012 ii. Dabbs, D.L. and Gentle, G.C. (1974). Landscape classification and plan succession Trends in Peace Athabasca Delta Can. Wildlife Service Rep. SER: 32-34. iii. District Statistical Handbooks of Sehore (2002 to 2007). iv. District resource map, Geological Survey of India, Central Region, Sehore. v. Gautam, N.C., and Narayan, L.R.A.(1983). Landsat MSS data for land use/land cover inventory and mapping–a case study of Andhra Pradesh, J. Indian Soc. Remote Sensing, 11(3): 15-28. vi. Hiese. (2001). From Application of Remote sensing and GIS for land use/land cover change analysis in a mountainous terrain. A case study of part of Kohima district, North- east India. vii. Jenson, J.R., (2002). Digital Image Processing. viii. Kamat, D.S., Gopalan, A.K.S, Majumdar, K.L., Ramakrishnan, R, Rao, U.R., Nag Bhushan, S.R., Thayalan, S., Krishnappa, P. and Sadashivaiah, A.S. (1985). Monitoring changes in ecology in the Kudremukh mining region, Int. Jour. Remote Sensing, 6:541-548. ix. Karwariya.Sateesh and Goyal Sandip., (2011). Land use and Land Cover mapping using digital classification technique in Tikamgarh district, Madhya Pradesh, India using Remote Sensing. x. Krishna Murthy, Y.V.N., Srinivasan, D.S., Balasubrahmania, R., Karale, R.L. Mahalanobis J.K., Roy, D.K. and Adak, N.K. (1992). Land use/ Land cover mapping of Ranganadi Catchment, Arunachal Pradesh, Natural

HYDRO 2014 International

resources Management–A new perspective (Ed. R.L. Karale) NNRMS Publication, Department of Space, Govt. of India, Bangalore, PP. 342-345. xi. Lillesand, J.M. and Kiefer, R.W., ―Remote Sensing and Image Interpretation‖. xii. Lodwick, G.D. (1981). A computer system for monitoring environmental changes in multitemporal Landsat kata, Can. Jour. Remote Sensing, 7:24-33. xiii. Rao Nageshwara, K., and Vaidyanathan, R. (1981). Land use capability studies from aerial photo interpretation – a case study from Krishna Delta, India. Geog. Rev. of India, 43(3). xiv. Sahai, Beldev. (1989). Remote Sensing of environment, Proc. Of the Application Seminar on Remote Sensing of Environment, Ministry of Environment and Forest, SAC (ISRO), Ahmedabad. xv. Savindra Singh, Types of drainage pattern and geomorphology. xvi. Shrivastav, P.N. and Guru, S.D., Gazetteer of India.

Multi Objective Optimization Of Cropping Pattern In A Canal Command Area Paritosh Srivastava1 and Raj Mohan Singh2 Resaerch Scholar, Civil Engineering Department, MNNIT Allahabad, and Assistant Professor at HCST Mathura 2 Associate Professor, Department of Civil Engineering, MNNIT Allahabad, India E-mail: [email protected] 1

ABSTRACT : Optimal cropping pattern depends on water availability and other constraints like crop area, soil properties, use of fertilizer, and local socio-economic conditions. Availability of water depends upon various hydrological and climatological factors like rainfall in area, aquifer properties of the area, existing canal network in the area etc. Use of surface and groundwater conjunctively not only improve production but also help in sustainable water utilization. Benefit and production (yield) are the two most important objectives for optimal cropping pattern. Single objectives considering benefit and production are formulated and solved for nine scenarios. Single objective separately optimize benefit or production only. Single objective maximization of benefit/production does not guarantee maximization of production/benefit. Multi objectives optimization of benefit and production simultaneously give optimal solution incorporating benefit and production. In present paper multi objectives optimization problem is formulated and solved using fuzzy programming approach (FPA) with Linear and non-linear membership functions Keywords: Optimization, constraints, groundwater, Cropping pattern, Aspiration level 1.

INTRODUCTION

Growth of population is impacting natural resources available for human consumptions and food security. Exponential growth of population leads to reduce the net agriculture area available for crop production. The population may touch the 9.25 billion mark by 2050 from the current 7 billion (United Nations 2008). Agricultural production needs to increase to provide food and fiber for the increasing global population (Davies and Simonovic

MANIT Bhopal

Page 247

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

2011; Singh and Panda 2012). One of the biggest challenges in the coming decades will be meeting the irrigation requirements and increasing of the food production especially in countries with limited water and land resources (FAO, 2002). Food production can be improved by providing additional area to agriculture or grow more crop per unit area. But due to twin problems of urbanization and possible natural environments disturbances, addition in agricultural area is becoming difficult day by day. In addition, the availability of water for irrigation will decrease because of expected increases in water demand for different competing sectors such as domestic, industrial, and hydroelectric generation. For instance, in India the irrigation allocation will probably decrease from the present level of 83% to approximately 69% by the year 2050 (Panjiar, 2010). Therefore, it is imperative to optimize the available land and water resources to achieve maximum returns. Various optimization techniques have been used to arrive at an optimal cropping pattern for optimal use of land and water resources for the maximization of net benefits and yield from irrigated agriculture. Linear programming one of the oldest optimization technique and it has been applied for the single objective approach with the objective function and the constraints as linear function of variables. Some research workers have modified the linear programming techniques to accommodate specific area based conditions (Arya and Hagan 1969; Heady et al 1973;Gulati and Murty 1979; Panda et al 1983; Kumar and Pathak 1989; Singh et al 2001; Mohamad and Said 2011). A command area is a complex system and a single objective approach gives only a partially efficient solution. Optimization of one objective dose not guarantee the optimal value for another objective(s). Therefore Multiobjective approach is imperative to get optimized solution for all the objective simultaneously. Some pioneer work discussing multiobjective application in their field optimal cropping pattern (Cohon and marks 1973, Morales et al. 1992; Raju and Kumar 2005;Regulwar and Gurav2010; Gore and Panda 2009; Mirajkar and Patel 2012).

Figure -1 Map of study area (a) Location map of India (b) Location map of Uttar Pradesh in India (c) Location map of soraon canal command in UP Table-1 Basic description of Soraon command area S.No 1 2 3 4 5 6

1.1 Study area description

7

The study area comprises of

Parameter

Value

Total command area (Ha) Cultural command area(Ha) Canal length (Command area) Wind speed max

23,089.93 11543 72.4 km 5.16 Km/hr 42.90 C

Mean maximum temperature Mean minimum temperature Number of rainy rays

9.300 C 49

Soraon canal command area 0

situated in Allahabad district with latitude 24 47‟ and 0

0

0

25 47‟North and longitude 81 09‟ and 81 21‟E. The annual average rainfall in the command area is 947 mm. The other detail of the command area is given in Table-1. Soraon command Area is bounded by two drain systems Raya and Basana respectively as shown in Figure 1. These two drain ultimately bounded by Kachar of Ganga river. Soraon rajwaha is the main canal in the study area feeded by Allahabad branch. Allahabad branch is part of Sarda Sahayak canal irrigation system which one is of the largest canal irrigation system in Utter Pradesh. Soil of the command area is mainly consist of silty loam and clay soil. Traditionally, the year is divided into two principal crop seasons, kharif (monsoon, July–October) and rabi (winter, November–April). Rice Wheat is the major crops in kharif and rabi season respectively. List of other major crops grow in command area is shown in Table 2.

HYDRO 2014 International

Table -2 Area, and production of major crops cultivated in the Allahabad

MANIT Bhopal

S.No

Crop

Season Kharif

Area (inHa)(% Area) 105361(67)

Production( Quintal) 228633

1

Paddy

2 3

Jowar Bajara

Kharif Kharif

5210 (3.2) 27862(17.5)

146 26116

4

Arhar

Kharif

18388(11.6)

22525

5 6

Wheat Barley

Rabi Rabi

215979(83) 7106(2.7)

583354 5945

7

Gram

Rabi

22489(8.6)

22805

8

Lentil

Rabi

9649(3.7)

8528

9

Pea

Rabi

4077(1.6)

5501

Page 248

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Source: Agricultural Department Report, Allahabad 2009-10

PDij = Production of the ith crop in jth seasons (Q/ha)

1.2 Data Acquisition

Multi-objective Based Planning

The data on crops, weather, groundwater, and canals were acquired from various Central and State Government departments and district administrations, located in and around the study area and from personal contact, such as Ground Water Cell, Irrigation Department, Office of tehsildar‟s, Department of Agriculture, and India Meteorological Department. Table -3 shows various data type and their sources Table -3 Data Type and their Sources

Max ZB = Σ j=1Σ i=1 {(PiYi j + PBi YiBi j )- Ci j }à (III) Max ZP = Σ j=1Σ i=1 =PDij à (IV) where Ã= Area in Ha 2.1 Constraints

S.No 1

3

Data Type Rain fall data Soraon Block (Allahabad) Canal Rostar Date /Irrigation Scheduling Canal Network Map

4

Soil Map

5

Groundwater Draft data Soraon block

2

2.

Sources IMD -Pune UP Irrigation Department UP Irrigation Department UP Agricultural Department Groundwater Department

METHODOLOGY

In single objective based planning an optimization model is formulated to achieve specified objective (maximization or minimization) under specified constraints. In present study maximization of benefit and production formulated and solved various soil and water availability constraints as discussed in section 2.1. Other than single objective Multiobjective including benefit and production is formulated and solved using fuzzy programming approach. Mathematically, objective function for maximization of benefit can be represented as: Max ZB = Σ j=1Σ i=1 {(PiYi j + PBi YiBi j )- Ci j } Ai j (i) where, ZB= Optimal benefit in rupees corresponding to the feasible solution Pi = Market price of the ith Crop (Rs/Kg) i=Crop index, i=1 to n, J=Crop season index; j=1 for Kharif & j=2 for Rabi Aij =Area allocated to ith crop grown in Jth seasons (ha) Cij = Cost of production of ith crop grown in Jth seasons (Rs/ha) Pi = Market price of the ith Crop (Rs/Kg) PBi= Market price of bi product of the ith Crop (Rs/Kg) Yij= Yield of the ith crop in jth seasons (Kg/ha) YiBij =Yield of the bi product ith crop in jth seasons of agricultural (Kg/ha) Maximization of Total Production/Yield Max ZP = Σ j=1Σ i=1 =PDij Ai j (II) ZP= Optimal Production in quintal corresponding to the feasible solution Aij =Area allocated to ith crop grown in jth seasons (ha)

HYDRO 2014 International

The previously stated Individual objectives and Multi-objective are subject to some constraints that are to be satisfied. The constraints which are taken into consideration are discussed below. 2.1.1 Area Constraints The area constraints are defined to account for the total available area for cultivation both in the kharif and the rabi. The crop rotation, the soil texture and the individual crop constraints as per available guidelines (Gore and Panda, 2009; Singh, 2012). Land availability constraints Land allocated to various crops under particular season should be less than or equal to the total cultivable area. Σ i=1Aij 10, the standard normal variate z is computed by using Equation (4) (Douglas et al., 2000). In a two-sided test for trend, H0 should thus be accepted if |z| ≤ zα/2 at the α level of significance. A positive value of „S‟ indicates an „upward trend‟; likewise, a negative value of „S‟ indicates „downward trend‟:

To explore the spatial distribution of the detected trends on annual and seasonal basis, the Z - value of Mann-Kendalltest was interpolated using ArcGIS 9.3 based on each stations slope value for the entire study period (1901-2010). The Spline method uses an interpolation method that estimates values using a mathematical function that minimizes overall surface curvature, resulting in a smooth surface that passes exactly through the input points. The Z-value maps of the study area for annual as well seasonal rainfall trends were prepared which are useful to determine the Z statistics of Mann-Kendall test at any point location in the study area. 3. RESULTS AND ANALYSIS 3.1 Mann-Kendell Test

(4) The Mann-Kendall test has two parameters that are of importance to trend detection. These parameters are the significance level that indicates the trends strength and the slope magnitude estimates that indicates the direction as well as the magnitude of the trend. The MK test checks the null hypothesis of no trend versus the alternative hypothesis of the existence of increasing or decreasing trend. 2.4 Magnitude of the Trend The magnitude of the trend in a time series was determined using a non-parametric method known as Sen's slope estimator method (Sen, 1968). This Method assumes a linear trend in the time series. In this method, the slope (T i) of all pairs are first calculated by

(5) Where Xjand Xk area data values at time j and k (j>k) respectively. The median of these N values of T i is Sen's estimator of slope which is calculated as

(6)

HYDRO 2014 International

The results for the non-parametric Mann-Kendall (MK) test applied to ascertain the significance of trends along with the Zvalues of the individual districts for monthly time series and seasonal and annual time series are shown in Table 3.1 and Table 3.2 respectively. The results indicated that, during the primary monsoon month i.e. July seven districts out of total nine districts area showing increasing rainfall at statistically significance level of more than 95% and four of them are supporting this trend at 99% significance level. This is followed by the pre-monsoon month May in which, five districts are confirming rising trend of rainfall with significance level of more than 95% and two districts out of this five are also supporting the trend at 99% significance level. The negative zvalues of January and December months‟ time series for all the districts are depicting the decreasing trend of rainfall but without any significant level. District Gurdaspur is supporting the rising trend of rainfall in all months expect six months of January, April, June, September, November and December. It is apparent from the results table 3.1 that the study area as a whole is supporting the rising trend Monsoon season and spring season with statistically significance level of more than 95%. Monsoon contributes around 80% of total annual rainfall, the annual rainfall series is also affirming the rising trend. The monsoon season series is having rising trend at 95% significance level for all the districts except Hoshiarpur. Five out of the nine districts have this significance level at 99%. Similarly in the spring season series, six out of nine districts are having statistically significant rising trend. Although, in the winter season series six districts are showing falling rainfall trend and three are showing rising trend, none of them are statistically significant. Table 3.1: Z-Value of MK Test for monthly series of rainfall data

MANIT Bhopal

Page 287

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

3.2 Magnitude of trend: The magnitude of the trend in the rainfall time series, as determined using the Sen Estimator, is given in Tables 3.3 and 3.4. The analysis of trends of rainfall variations by districts shows a large variability in the magnitude and direction of trend from one district to another. Monthly analysis of district rainfall indicated that the majority of the districts have very little or no change in non-monsoon months (Table 3.3) and the monsoon months witnessed increasing rainfall in the majority of districts.

The analysis of annual series is very similar to Monsoon series, as except Hoshiarpur districts all the districts are having statistically significant upward trends. Gurdaspur district is having highest significance level of more than 99.9% for annual series. It is evident from the Table 3.1., Table 3.2 and Fig 3.1 that there are rising trend of precipitation over the last century in most of the districts of north-west Punjab state. Table 3.2: Z-Value of MK Test for Annual and Seasonal series of rainfall

In order to get the study area‟s response the Z-values of annual and seasonal time series also have been interpolated using spline method. The Z-value maps of the study area for annual as well seasonal rainfall trends are presented in the Fig. 3.1. Z-value maps of the area are useful to determine the Z statistics of MannKendall test at any point location in the study area. The annual average rainfall of the lower Sutlej is supporting the rising trend with more than 95% significance level in most of the area. The analysis of the rainfall data by parametric and non-parametric statistical test and the magnitude of trends supports the rising trend over the study area. While analysing the results presented in the various tables and figures one can easily conclude that the only winter season is not supporting any trend but the monsoon and the annual rainfall in the region has been increased over the last century.

HYDRO 2014 International

MANIT Bhopal

Page 288

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014 rate of rise of rainfall over last century comes out be of an order of 1.42mm/year based on annual time series. Table 3.4: Sen Estimator of slope (mm/year) for Annual and seasonal rainfall.

4. CONCLUSIONS

Fig 3.1Mann-Kendall Z-Statistic for Annual, Monsoon, winter and spring rainfall trend Table 3.3Sen Estimator of slope (mm/year) for monthly rainfall.

On analyzing the estimates of the slope (mm/year) for annual and seasonal rainfall series of all the districts (Table 3.4), it is implicit that three districts experienced decreasing rainfall in the winter season. The maximum reduction was found for Kapurthala (-0.08 mm/year). The maximum increase out of 9 districts was experienced by Gurdaspur in rainfall (1.95 mm/year annually and 1.56 mm/year Monsoon season) followed by Rupnagar (1.72 mm/year annually and 1.31 mm/year Monsoon season) and Amritsar (1.53 mm/year annually and 1.31 mm/year Monsoon season). Considering the study area as a whole, the

HYDRO 2014 International

The understanding of spatial and temporal distribution and changing pattern of precipitation is most fundamental and primary input required for any kind of planning and management of water resources. The aim of this work was to study rainfall trend in Bist-doab region of Punjab and spatial distributionof the annual as well as seasonal and monthly precipitation series for the period of 1900-2010 were analyzed.Punjab state holds place of pride among the Indian States for its outstanding achievements in agricultural development productivity. Agricultural productivity is sensitive to global climatic changes and therefore the tracking of theses changes in terms of trend is essential to generate variable predictions about impact of change in climate and variability.Most frequently used non-parametric Mann-Kendall test and simple linear regression technique was used to identify the significant trends of rainfall in this study and the magnitude of the trends were ascertained by the wellestablished Sen‟ slope estimator method. Overall annual rainfall showed increasing trend at almost all the districts with certain significance level. Average rainfall of the study area was 546mm. The highest average annual rainfall is estimated at Gurdaspur district (661mm) and lowest at Moga district (379mm). While looking at the trend analysis part the maximum increase was 1.95 mm/year and the maximum decrease was 0.08 mm/year. Over the complete study area, the annual rainfall showed an increasing trend and the rate of rise was estimated 1.42 mm/year.Seasonal analysis showed that monsoonal rainfall increased statistically significant over all districts except Hoshiarpur district. Spring rainfall increased over 6 districts with statistically significance levels. Winter rainfall at five districts showed falling trend but these are not at significance levels. The annual rainfall trends fully indorsing to the monsoon trend, which implicit the dominance of the monsoon for rainwater arability in the region. The study indicated a clear pattern of rainfall trends in the study area with less variability, this could be due to homogeneous geographic and climatic conditions of the districts falling the study area.

REFERENCES: i. Borgaonkar HP and Pant GB (2001) Long-term climate variability over monsoon Asia as revealed by some proxy sources. Mausam 52: 9-22.

MANIT Bhopal

Page 289

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

ii. Douglas, E.M., Vogel, R.M., Kroll, C.N., 2000. Trends in Floods and low flows in the United States: Impacts of spatial correlations. Journal of Hydrology 240, 90-105 iii. Gill K.K., Kukal S.S., Sandhu S.S. and HarjeetBrar (2013). Spatial and temporal variation of extreme rainfall events in central Punjab. International Journal of Applied Engineering Research. ISSN 0973-4562, Volume 8, Number 15 (2013) pp. 1757-164 iv. Hirsch, R. M., J. R. Slack, and R. A. Smith (1982), Techniques of trend analysis for monthly water quality data. Water Resour. Res., 18(1), 107– 121 v. Goswami B N, Venugopal V, Sengupta D, Madhusoodanan M S and Xavier P K (2006), Increasing trend of extreme rain events over India in a warming environment, Science, 314, pp. 1442-1445. vi. Kendall MG and Stuart A (1961) The Advance Theory of Statistics: 2, Hafner Publishing Co, New York. vii. Kendall, M.G., 1962. Rank correlation methods, third ed. Hafner Publishing Company, New York. viii. Kumar V and Jain SK (2010). Trends in seasonal and annual rainfall and rainy days in Kashmir Valley in the last century. Quaternary International 212, 64-69. ix. Kumar V, Singh P and Jain SK (2005) Rainfall trends over Himachal Pradesh, Western Himalaya, India. Proc. Conf. Development of Hydro Power Projects – A Prospective Challenge, Shimla. x. Lanzante, J. R. (1996), Resistant, robust and non-parametric techniques for the analysis of climate data: theory and examples, including applications to historical radiosonde station data. Int. J. Climatol., 16: 1197– 1226. xi. Mann, H.B. (1945) Nonparametric tests against trend. Econometrica, 13, pp. 245–259 xii. Pant G B and Rupa Kumar K (1997) Climates of South Asia: Behaviour Studies in Climatology, John Wiley, pp. 126-127. xiii. Pant GB and Borgaonkar HP (1984) Climate of the Hill regions of Uttar Pradesh. Himalayan Research and Development 3, 13-20. xiv. Sen PK (1968) Estimates of the regression coefficient based on Kendall's tau. Journal of the American Statistical Association 63: 1379-1389. xv. Sharma KP, Moore B and Vorosmarty CJ (2000) Anthropogenic, climatic and hydrologic trends in the Kosi Basin, Himalaya. Climate Change 47, 141-165. xvi. Shreshtha AB, Wake CP, Dibb JE and Mayewski PA (2000) Precipitation fluctuations in the Nepal Himalaya and its vicinity and relationship with some large-scale climatological parameters. Int J Climatol 20: 317-327. xvii. Singh, P., Kumar, V., Thomas, T. & Arora, M. (2008a). Changes in rainfall and relative humidity in different river basins in the northwest and central India. Hydrol. Processes 22, 2982–2992. xviii. Singh, P., Kumar, V., Thomas, T. & Arora, M. (2008b). Basin-wide assessment of temperature trends in the northwest and central India. Hydrol. Sci. J. 53(2), 421–433 xix. Vijay Kumar, Sharad K. Jain & Yatveer Singh (2010). Analysis of long-term rainfall trends in India, Hydrological Sciences Journal, 55:4, 484-496

HYDRO 2014 International

MANIT Bhopal

Page 290

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Net Radiation Estimation From A Remotely Sensed Data Using Sebal Model M.V.S.S.Giridhar1 and P. Suneel2 Asst. Prof., Centre for Water Resources, Institute of Science and Technology, J N T U H, Hyderabad, email: [email protected] 2 P.G. Student, Centre for Water Resources, Institute of Science and Technology, J N T U H, Hyderabad

1

Abstract : The objective of the present study is to estimate the daily net radiation from the remote sensing images for Nagarjuna Sagar Left Bank Canal Command Area, Andhra Pradesh, India. Net radiation is one of the important parameter for further estimation of actual evapotranspiration. The study was conducted for the day of the imagery taken by the LANDSAT satellite on 22nd October 2011 with thermal and near infrared (NIR) bands. The study area lies between latitude 160 35‟06.34”N to 170 02‟40.77”N and longitude 790 16‟16.42”E to 800 03‟03.11”E. The command area map was digitized on the toposheet and processed for the estimation of net radiation with the help of ARCVIEW and ERDAS. The DN values of bands were used to estimate radiance, reflectance, Net radiation and soil heat flux. The thermal range of bands has been used to estimate temperature related components like the estimation of net radiation and sensible heat flux. After the estimation of all energy balance components the evaporative fraction was used to estimate the daily rate of actual evapotranspiration using SEBAL algorithm. The minimum and maximum net radiation estimated in the command area is 8.1183 W/m2 and 225.3886 W/m2 respectively on the day of pass. Further, from the net radiation and soil heat flux, the actual evapotranspiration was also calculated for the entire command area which will be useful for effective crop water management at a local level. Keywords: Nagarjuna Sagar, SEBAL, LANDSAT, radiation, actual evapotranspiration 1.0 INTRODUCTION: Land managers today are faced with multiple challenges. They must, as a result of expanding economic pressures, become advocates for the sustainable use of our natural resources while ensuring that the quality of the resource is maintained. The public and the courts are increasingly demanding objective and effective strategies for managing these exhaustible resources. Fresh water has become our most precious natural resource and the wise management of this resource is one of our greatest challenges. An understanding of the natural systems and the physical laws that govern each component of the hydrologic cycle is very important for the water resource manager. In particular, the evaporation processes from the various surfaces of the earth need to be understood in order to achieve a sustainable development of our water resources. Cropland irrigation is a major consumer of water in semiarid and arid regions and an efficient and reliable method for determining the consumptive use of water by crops is crucial for adequate water management. As fresh water becomes scarcer, competition for fresh water

HYDRO 2014 International

intensifies and better irrigation management will be required to achieve greater efficiency in the use of this valuable resource. Out of all the components of the hydrological cycle, land surface evapotranspiration estimates are crucial for water balance studies. Also this is perhaps the most difficult component to estimate because of complex interactions between this and the components of the land-plant-atmosphere system. Land surface evapotranspiration is governed by the conditions of the lower part of the atmosphere, the presence and the properties of the vegetation layer and the sub surface soil moisture conditions. The condition of the lower part of the atmosphere depends on the supply of heat energy and the vapour pressure gradient, which, in turn, depend on meteorological factors such as temperature, wind speed, atmospheric pressure, and solar radiation. These factors also depend on other factors, such as geographical location, season, time of day, etc. Calculation of net radiation at any instance will generate actual evapotranspiration from each pixel to global level using remote sensing imagery. The net radiation flux at the surface (Rn) represents the actual radiant energy available at the surface. It is computed by subtracting all outgoing radiant fluxes from all incoming radiant fluxes. Net surface radiation = gains – losses. The amount of shortwave radiation (RS↓) that remains available at the surface is a function of the surface albedo (α). Surface albedo is a reflection coefficient defined as the ratio of the reflected radiant flux to the incident radiant flux over the solar spectrum. It is calculated using satellite image information on spectral radiance for each satellite band. The incoming shortwave radiation (R S↓) is computed using the solar constant, the solar incidence angle, a relative earth-sun distance, and a computed atmospheric transmissivity. The incoming longwave radiation (R L↓) is computed using a modified Stefan-Boltzmann equation with atmospheric transmissivity and a selected surface reference temperature. Outgoing longwave radiation (RL↑) is computed using the Stefan-Boltzmann equation with a calculated surface emissivity and surface temperature. Surface temperatures are computed from satellite image information on thermal radiance. The surface emissivity is the ratio of the actual radiation emitted by a surface to that emitted by a black body at the same surface temperature. In SEBAL, emissivity is computed as a function of a vegetation index. The final term in above Equation, (1-εo)RL↓, represents the fraction of incoming long wave radiation that is lost from the surface due to reflection. SEBAL requires a satellite image and some weather data. A land-use map for the area of interest is also helpful. This manual applies specifically to images produced from the Landsat satellites. Landsat Thematic Mapper (TM) bands 1-5 and 7 provide data for the visible and near infrared bands. The pixel size for these bands is 30m by 30m. TM band 6 provides data for longwave (thermal) radiation. The pixel size for this band is 60m by 60m for Landsat 7 and 120m by 120m for Landsat 5. We recommend using images from Landsat 7 since there is more error possible with recent Landsat 5 images due to sensor degradation. Some Landsat 5 satellite constants that are used in the computations are updated to 1986 and may be in error for 2002 without adjustment. SEBAL calculates ET through a series of

MANIT Bhopal

Page 291

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

computations that generate: net surface radiation, soil heat flux, and sensible heat flux to the air. By subtracting the soil heat flux and sensible heat flux from the net radiation at the surface we are left with a residual energy flux that is used for evapotranspiration (i.e. energy that is used to convert the liquid water into water vapour). In the SEBAL model, ET is computed from satellite images and weather data using the surface energy balance. Since the satellite image provides information for the overpass time only, SEBAL computes an instantaneous ET flux for the image time. The ET flux is calculated for each pixel of the image as a residual of the surface energy budget equation (1). ETact= Rn - G – H …1 where incoming components are positive and outgoing are counted as negative. The net radiation is the sum of the incoming and outgoing short and long wave components. In above Equation, the soil heat flux (G) and sensible heat flux (H) are subtracted from the net radiation flux at the surface (R n) to compute the residual energy available for evapotranspiration (λET). Soil heat flux is empirically calculated using vegetation indices, surface temperature, and surface albedo. Sensible heat flux is computed using wind speed observations, estimated surface roughness, and surface to air temperature differences. SEBAL uses an iterative process to correct for atmospheric instability due to the buoyancy effects of surface heating. Norman et al, (1995) developed two layer model of turbulent exchange that includes the view geometry and associated with directional radiometric surface temperature is developed and evaluated by comparison of model prediction with field measurement. Bastiaanssen et al, (1998) evaluated the major bottlenecks of existing algorithms to estimate the spatially distributed surface energy balance in composite terrain by means of remote sensing data are briefly summarized. Michael (2003) estimated of absolute surface temperature by satellite remote sensing. Land surface temperature is strongly influenced by the ability of the surface to emit radiation, i.e. surface emissivity. The objective of our study is to demonstrate the feasibility of Landsat TS product as a source for calculating spatial distribution of Ta to detect urbanization effect in Jakarta city (Hasti et al 2013). Gislain (2006) used the Surface Energy Balance Algorithm for Land (SEBAL) images was used to determine the actual evapotranspiration of acquisition day Landsat ETM+ images of 26/08/2000, 25/05/2001, 16/08/2002 and 31/05/2003.The development of Surface Energy Balance Algorithm for Land (SEBAL) by Bastiaanssen et al (1998a,b) built up the milestone for estimation of land surface turbulent energy fluxes. The relationships between visible and thermal infrared spectral radiances of areas with a sufficiently large hydrological contrast constitute the basis for the formulation of the SEBAL model. After its establishment, a lot of filed validations have been done in different area, especially in arid and semi-arid area. However, due to the difficulty to find exactly right pixels of dry and wet conditions in certain images, its application is limited in a certain degree. To solve related

HYDRO 2014 International

limitation of SEBAL, some correction have been made by Su (2002) to make it more practicability, who remedied a theoretical problem of SEBAL model and added a scheme to apply NWP fields with an Up-scaling and down-scaling approach (Su and Pelgrum et al., 1999). In another effort, Roerink et al (2000) developed a new method to derive the surface energy fluxes from remote sensing measurements, called the Simplified Surface Energy Balance Index (S-SEBI), which fits dry and wet cases present in the spatial radiometric data and showed reasonable success for application to semiarid areas. But after this approach a new model came into light that was SEBS model. The surface energy balance system (SEBS) is developed for the estimation of atmospheric turbulent fluxes and evaporative fractions using satellite earth observation data, in combination with meteorological information at proper scales (Su, 2002). The land surface parameters (albedo, emissivity, temperature, fractional vegetation cover and leaf area index) for the system are extracted from the reflectance and radiance measurement of the satellite. The other input used includes air pressure, temperature, humidity, and wind speed at a reference height. Energy balance models of ET require explicit characterisation of numerous physical parameters, many of which are difficult to determine globally and locally. SEBAL Bastiaanssen et al (1998a,b), SEBS (Su,2002), and RSEB (Kalma & Jump, 1990) estimate ET as a residual of the energy balance at the earth‟s surface, which contain biases from both the sensible heat flux and net radiation. The REBM model (Mc Vicar & Jupp, 1999, 2002) uses combined remote sensing data and meteorological data to calculate ET, while the triangle method (Gillies & Carlson, 1995; Nemani & Running, 1989: Nishidha & Nemani, 2003) uses the slope of surface temperature versus the NDVI to estimate the surface resistance to ET, and the dual-source model developed by Norman et al (1995) and Kustas and Norman (1999) uses multi-angular remote sensing. For all the above models, thermal remote sensing data, land surface temperature are the most important inputs. The objective of the present study is to calculate accurate land surface temperatures from the remote sensing imagery to give as a input to further estimate actual evapotranspiration at pixel by pixel level at locally and globally to feed as a input to climatic models for forecasting. 2.0 DESCRIPTION OF STUDY AREA The present study is to estimate the daily net radiation and finally to estimate actual evapotranspiration through remote sensing image for Nagarjuna Sagar Left Bank Canal Command area. The study was conducted for day 22 nd October 2001. The study area lies between latitude 16035‟06.34” N to 17002‟40.77”N and longitude 79016‟16.42”E to 80003‟03.11”E. The study area has been delineated using the command area map provided by the irrigation department. The command area map was georeferenced with respect to SOI toposheet. The boundary of the study area was digitized and cut-out. The present research work is aimed to apply the SEBAL models for estimation of net radiation and finally actual evapotranspiration in a semi arid region of India namely Nagarjuna Sagar left bank Canal (NSLC) Command area of Andhra Pradesh. SEBAL can compute ET for

MANIT Bhopal

Page 292

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

flat, agricultural areas with the most accuracy and confidence level. 3.0 METHODOLOGY Major part of the work has been done with ILWIS software, which is also used for compiling and analyzing the data. ILWIS has a very good GIS operation capability and is specially built for Land-Water analysis. With the added advantage of easy script writing facility, it can be said as the best software for this present kind of analysis. In order to be able to make the right decisions, access to different sorts of information is required. The data should be maintained and updated and should be used in the analysis to obtain useful information. In this process ILWIS can be an important tool. Figure 1 represents False Colour composite of NSLC command area which was digitized and cut out in ERDAS imagine 8.7. Command Area was delineated Using Arcview 9.1 Software with the help of SOI toposheets thereby after creating the boundary shape file we are suppose to import the boundary shape file to another software ERDAS IMAGINE 8.7 which is very flexible to create a subset of an image. We have to create subset by placing the boundary file above the band1, band2, band3, band4, band5, band6-1, band6-2, band7, band8 everything will be provided in individual format so that we can create subset for every band of the satellite imagery with the help of boundary file thereafter reprojection to be done for the satellite imagery so that UTM values will be converted into Lat-Long Values. The Landsat ETM+7 images dated October 22, 2001 were downloaded for the full scene of the path and row 143/48 from global land cover facility site. The raw images acquired had file formats as GEOTIFF. Pre-processing such as geometric, radiometric and atmospheric corrections which are a prerequisite for analysis of energy fluxes and land cover parameters were done. All the operation from importing the data to analysis of the data was carried out in the GIS & RS software ILWIS (Integrated Land Water Information System). All the data required for performing operations on landsat imagery were given in a meta data file namely Acquisition date, latitude, longitude, Path, row, UTM zone number etc. During product rendering image pixels are converted to units of absolute radiance using 32 bit floating point calculations. Pixel values are then scaled to byte values prior to media output. For relatively clear Landsat scenes, a reduction in between-scene variability can be achieved through a normalization for solar irradiance by converting spectral radiance, as calculated above, to planetary reflectance or albedo. This combined surface and atmospheric reflectance of the Earth is computed. The value of ESUNλ and„d‟ is provided. The reflectivity of a surface is defined as the ratio of the reflected radiation flux to the incident radiation flux. ETM+ Band 6 imagery can also be converted from spectral radiance (as described above) to a more physically useful variable. This is the effective at-satellite temperatures of the viewed Earthatmosphere system under an assumption of unity emissivity and using pre-launch calibration constants. The thermal band data (Band 6 on TM and ETM+) can be converted from at-sensor spectral radiance to effective at-sensor brightness temperature. The at-sensor brightness temperature assumes that the Earth‟s

HYDRO 2014 International

surface is a black body (i.e., spectral emissivity is 1), and includes atmospheric effects (absorption and emissions along path). The at-sensor temperature uses the prelaunch calibration constants given in Table. 3.1 SEBAL METHODOLOGY Surface Energy Balance Algorithm for Land (SEBAL) is a relatively new parameterization of surface heat fluxes based on spectral satellite measurements. SEBAL requires spatially distributed, visible, near-infrared and thermal infrared data, which can be taken from any suitable satellite sensor working in these wavelength. The SEBAL parameterization is an iterative and feedback-based numerical procedure that deduces the radiation, heat and evaporation fluxes. The key input data for SEBAL consists of spectral radiance in the visible, nearinfrared and thermal infrared part of the spectrum. SEBAL computes a complete radiation and energy balance along with the resistances for momentum, heat and water vapor transport for every individual pixel. The resistances are a function of state conditions such as soil water potential (and thus soil moisture), wind speed and air temperature and change from day-to-day. Satellite radiances will be converted first into land surface characteristics such as surface albedo, leaf area index, vegetation index and surface temperature. These land surface characteristics can be derived from different types of satellites. First, an instantaneous evapotranspiration is computed, that is subsequently scaled up to 24 hours and longer periods. In addition to satellite images, the SEBAL model requires weather data parameters, such as Wind speed, Humidity, Solar radiation and Air Temperature. There is no data on land cover, soil type or hydrological conditions required to apply SEBAL. The primary basis for the SEBAL model is the surface energy. The instantaneous ETact flux is calculated for each cell of the remote sensing image as a residual of the surface energy budget equation (1). It is important that the image used is for a totally clear sky. ET cannot be computed for cloud covered land surfaces, because even a thin layer of cloud can considerably drop the thermal band readings and cause large errors in calculation of sensible heat fluxes. Therefore, all satellite images should be thoroughly checked for the occurrence of cloud cover and if found, these areas should be masked out and dealt with individually. In order to detect clouds in an image, view the thermal band using a range of colors to differentiate temperature values. Clouds will then show up as uniquely colored areas in the image and can be flagged. Masking can be done following the processing of ET, but is an important final step. 3.1.1. Header file information TM images are generally created with an associated header file. The header file for the satellite image is a relatively small file that contains important information for the SEBAL process. The following information must be obtained from the header file for entry into SEBAL:  The satellite overpass date and time  The latitude and longitude of the center of the image  The sun elevation angle (β) at the overpass time

MANIT Bhopal

Page 293

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

 Gain and Bias levels for each band The satellite overpass time is expressed as Greenwich Mean Time (GMT) and must be converted to local time. For Landsat 7, the header file includes the gains and biases for bands 1-5 and 7. These parameters are used to convert digital numbers (DN‟s) in the original files into energy units. For band 6, the thermal band, both high and low gain images are supplied and the user must decide which one to use for SEBAL. We recommend using the low gain image, which yields slightly lower resolution, but is less likely to suffer from saturation. When the low gain image is selected, use the gain and bias information for low gain. If the header file is missing, the user must use the internet to find gain and bias data for the image. 3.3 Estimation of total daily net radiation It is the result of the energy balance between the incoming and outgoing long and shortwave radiation on the Earth' surface during one day. Positive fluxes indicate radiation reaching the surface and negative leaving it. The mathematical equation that expresses this balance, C1 is the conversion factor for surface Albedo from the image. It is the ratio between the average daily Albedo to the instantaneous Albedo as it is derived from the visible band image. A default value 1.1 can be taken.

stations not having a net longwave radiometer but information on standard daily averaged meteorological parameters, the exchange of long-wave radiation Lday between vegetation and soil on the one hand, and atmosphere and clouds on the other, can be represented by the radiation law. 4.0 RESULTS AND DISCUSSION The reflective and thermal bands of the ETM sensor were converted into reflectance and temperature maps using the calibration coefficients and solar zenith angle derived in the above section. Band 1, 2, 3, 4, 5 and 7 were converted into reflectance and band 6 was converted into temperature. The temperature is known as brightness temperature or radiant temperature because it is measured at the top of the atmosphere by the satellite. Temperature map was resampled to the reflective band resolution for all future analysis. The dry and wet pixels were identified using the criteria described in the methodology. Scatter plot between Surface Temperature and Albedo was made to find the dry pixel. Another scatter plot between surface temperature and NDVI was constructed to find the wet pixel. Estimated values were listed for the dry and wet pixel is given in Table 1. Table 1. Derived components of dry and wet pixel for the study region

3.3.1Daily terrestrial solar radiation (S↓day exo) The maximum instantaneous solar radiation outside the atmosphere, measured at an average Sun-Earth distance, and perpendicular to the solar rays is equal to 1367 watt/m2. The daily terrestrial solar radiation as defined as 24 hour average of the total energy reaching the top of the atmosphere at the point of consideration.

The solar radiation reaching the ground is a function of geometric and atmospheric factors such as date of the year, latitude, sunshine fraction and atmospheric gaseous components. Due to the highly temporal and spatial variation of the atmospheric components, the determination of the incoming shortwave radiation reaching the ground is usually done by means of atmospheric-solar models in combination with ground data collection. The incoming shortwave radiation or global radiation, 'S↓', is measured at ground stations by means of instruments called pyranometers. These instruments usually work in the entire visible broadband range (usually 0.305 - 2.4 µm). This range comprises almost 96% of the spectral solar irradiance.

The dT map is an important input for the determination of sensible heat. The value of dT varied from 3.157K to 14.753K for the study region. Since Sensible Heat Flux is an implicit function of components that contains it therefore an iterative process is run to derive the final value. The iteration process is carried with initial value of ψm (correction factor for momentum transport) and ψh (correction factor for heat transport) as zero then subsequently computing the maps of Friction Velocity (U*), Aerodynamic resistance for heat transport (Rah), Sensible Heat (SH), Monin Obukhov Length (L), x m, ψm, xh, ψh in sequential order. The value of ψm and ψh at the end of the first iteration is taken for the next iteration. This iterative process is run till a constant value of sensible heat is reached. Sensible heat flux ranged from 0.0066 to 141.0801 Wm-2 for the command area. Histogram of sensible heat flux shows a normal distribution of values.

3.3.3 Average daily net longwave radiation (Lday)

4.1 Net daily radiation

There is a significant exchange of radiant energy between the earth's surface and the atmosphere in form of radiation at longer wavelengths (3-100 μm). The average daily net longwave radiation is given by: The value of Lday can be determined with appropriate instruments (net longwave radiometer) and is the most accurate method since direct information is available. For

Daily net radiation is the integration of all the instantaneous values of net radiation for the whole day, there are separate set of formulas for its estimation.

3.3.2 Average daily incoming shortwave radiation (S↓day)

HYDRO 2014 International

4.1.1 Daily terrestrial solar radiation (S↓dayexo)

MANIT Bhopal

Page 294

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Day angle map was created using the Julian day value; this map is used to compute the eccentricity correction factor and the solar declination values. The sunrise hour angle map was developed using the latitude map and solar declination map. Using the above derived components the daily terrestrial solar radiation was computed in MJm-2day-1. The value of daily terrestrial solar radiation was nearly constant and ranged from 24.5959 M J m-2 day-1 to 24.7470MJm-2day-1 for the command area. Average daily incoming Shortwave radiation is estimated to be 166.0981 4.1.2 Average daily net longwave radiation (Lday) The average daily net longwave radiation was calculated using the meteorological data. Mean relative humidity for the satellite overpass day was recorded as 80% and the mean air temperature was taken as 180C. Using the meteorological data the average daily net longwave radiation has been estimated as -65.09 Wm-2. Also using the relation provided be De Bruin (1987), the net longwave radiation has been estimated as -64.163 Wm-2. 5.0 CONCLUSION The present Research work was undertaken to study the Surface energy balance algorithm for land (SEBAL) model on Nagarjuna Sagar left bank canal command area and to estimate the net radiation from the command area. The minimum and maximum radiation has been estimated as 8.1183 and 225.3886 w/m 2 respectively for the study area for the given date of pass.

iii. Gillies, R. R., & Carlson, T. N. (1995). Thermal remote sensing of surface soil water content with partial vegetation cower for incorporation into mesoscale prediction models. Journal of Applied Meteorology, 34, 745−756. iv. Hasti et al (2013) Air Temperature Estimation from Satellite Remote Sensing to Detect the Effect of Urbanization in Jakarta, Indonesia, Journal of Emerging Trends in Engineering and Applied Sciences (JETEAS) 4(6): 800-805 Kalma, J. D., & Jupp, D. L. B. (1990). Estimating evaporation from pasture using infrared thermometry: evaluation of a one-layer resistance model. Agricultural and Forest Meteorology, 51, 223−246. v. Kustas, W. P, & Norman, J. M. (1999). Evaluation of soil and vegetation heat flux predictions using a simple two-source model with radiometric temperatures for partial canopy cover. Agricultural and Forest Meteorology, 94, 13−29 vi. McVicar, T. R., & Jupp, D. L. B. (1999). Estimating one-time-of-day meteorological data from standard daily data as inputs to thermal remote sensing based energy balance models. Agriculture and Forest Meteorology, 96, 219−238. vii. McVicar, T. R., & Jupp, D. L. B. (2002). Using covariates to spatially interpolate moisture availability in the Murray–Darling Basin: A novel use of remotely sensed data. Remote Sensing of Environment, 79, 199−212. viii. Michael (2003) Estimation of Absolute surface temperature by satellite remote sensing, Thesis submitted to International Institute for Geoinformation Science and earth observation. Enschede, The Netherlands, pp:1-70. ix. Nemani, R. R., & Running, S. W. (1989). Estimation of regional surface resistance to evapotranspiration from NDVI and thermal infrared AVHRR data. Journal of Applied Meteorology, 28, 276−284. x. Nishida, K., Nemani, R., et al. (2003). Development of an evapotranspiration index from Aqua/MODIS for monitoring surface moisture status. IEEE Transactions on Geoscience and Remote Sensing, 41(2). xi. Norman, J. M., Kustas, W. B., & Humes, K. S. (1995). Source approach for estimating soil and vegetation energy fluxes in observations of directional radiometric surface temperature. Agricultural and Forest Meteorology, 77, 263−293. xii. Roerink, G.J., B. Su, and M. Menenti, 2000. S-SEBI - A simple remote sensing algorithm to estimate the surface energy balance, Physics Climate Earth Journal (B) 25(2):147-157. xiii. Su, Z. (2002). The surface energy balance system (SEBS) (for estimation of turbulent heat fluxes. Hydrology and Earth System Sciences, 6, 85−99.

Low Flow Analysis In Bina River Basin Of Madhya Pradesh V.K. Chandola1*, Sunil Kumar Yadav1, R.V. Galkate3, Palak Mehata4 1. Professor, Department of Farm Engineering, Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh, India – 221005, * Corresponding Author : Email: [email protected] 2. M.Tech. Student, Department of Farm Engineering, Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh, India – 221005 3. Scientist, National Institute of Hydrology, Ganga plains South Regional Center, WALMI Campus, Bhopal Madhya Pradesh, India - 462016. 4. Junior Research Fellow, Ganga plains South Regional Center, WALMI Campus, Madhya Pradesh, India 462016.

REFERENCES i. Bastiaanssen, W. G. M., Menenti, M., Feddes, R. A., & Holtslag, A. A. M. (1998). The Surface Energy Balance Algorithm for Land (SEBAL): Part 2 validation. Journal of Hydrology, 212–213, 213−229. ii. Bastiaanssen,W.G.M., Menenti, M., Feddes, R.A.&Holtslag, A.A.M. 1998. A remote sensingsurface energy balance algorithm for land (SEBAL): 1. Formulation. Journal of Hydrology 212–213: 198–212.

HYDRO 2014 International

ABSTRACT: The study was conducted in the sub catchment of Bina River, a tributary of Betwa River, in drought prone Bundelkhand region of Madhya Pradesh, India during 201112, to analyze the low flow events for assessment of drought. The daily stream flow data, recorded at Rahatgarh gauge

MANIT Bhopal

Page 295

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

discharge site of Bina River Basin, was analyzed for assessment of hydrological drought situation, drought frequency, duration, and severity of low flow using Flow Duration Curve Technique. The dependable flow at 75% probability of exceedance was considered as truncation level to obtain deficiency volume and its severity for each event of low flow condition. Stream flow drought severity was taken as the total deficit or cumulative deficient runoff volume below the truncation level during the period of the event of low flow condition.The river experienced 6 low flow events over the period of 9 years (1990-1998) indicating 1 to 2 low flow events every year. The low flow events in this basin usually begin during July to October and terminate during November to December. The severity of low flow varies from 2.88 to 287.37 MCM and duration of low flow events ranges from 10 to 22 days. The maximum severity of 287.37 MCM has been observed for 23 days during September, 1994. Key Words: Drought Severity; Flow Duration Curves; Hydrological Drought; Low Flow; Stream Flow; Truncation Level. 1.

using low flow analysis may be helpful in planning and development of water resources to meet the various water demands in the river basin. The low flow, according to the International Glossary of Hydrology (WMO,1974)is defined as the flow of water in a stream during prolonged dry weather. This paper deals with the assessment of hydrological drought situation using low flow analysis in Bina river basin using daily stream flow data of Rahatgarh Gauge-Discharge (G/D) site. 2. 2.1

MATERIALS AND METHODS Study Area

Bina river is an important tributary of Betwa River of Madhya Pradesh. It originates in Raisen district and enters in Rahatgarh block and traverse through Khurai and Bina blocks of Sagar district. The study area is located partly in Sagar, Vidisha and Raisen districts of Madhya Pradesh. The index map and drainage map of the

INTRODUCTION

Drought is an extended shortfall of precipitation that results in water supplies insufficient to meet the needs of humans and the environment and occur routinely as part of the natural hydrologic cycle (Wilhite and Smith, 2005). The prime cause of drought is occurrence of below normal precipitation, which is affected by various natural phenomenon. The occurrence of drought leads to reduction in river flow, consequent reduction in reservoir and tank levels and depletion of soil moisture and groundwater. The surface water deficits are reflected through low stream flows and reduced reservoir storages. The low stream flows are indicative of drought situations. When the stream flows are not sufficient enough to meet the required demand of water, it is considered that the drought has set in. The drought severity, frequency and duration can be studied by low flow analysis of the local streams. During the rainfall deficient condition the deviation from normal values is greater for stream flows than the rainfall.Due to uneven distribution of precipitation, catchment characteristics and predominant hydro meteorological factors in the watershed, all taken as the input, there is wide degree of variation in the runoff, taken as output of the hydrological system.

st y area is shown in Figure 1.

Bina river is one of the important river in Bundelkhand region of Madhya Pradesh. The Water Resources Department, Govt. of Madhya Pradesh has planned a “Bina Complex-Irrigation and Multipurpose Project (BC-IMP)” on Bina river to meet the domestic water demand of towns like Rahatgarh, Khurai and Bina and industrial water demand of Bina Refinery and proposed JP power project. Water will also be supplied to meet water demand of railways. Beside this, large quantity of water will be directly pumped from the river for irrigation.Thus the knowledge of the available flows in rivers during monsoon and lean period is vital in formulation of the yearly plan for water uses which include domestic and industrial water supply, besides planning for economic and ecological activities of a given region (Clausen and Pearson, 1995). The assessment of water scarcity situation

HYDRO 2014 International

MANIT Bhopal

ud

India

M

Page 296

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

Figure 2: Locations of raingauge stations and gauge-discharge site in Bina river 2.2.1

Map of Bina Basin up to Rahatgarh G/D Figure 1: Index Map of Bina River Basin up to Rahatgarh Gauge Discharge Site The geographical area of the Bina river basin up to Rahatgarh Gauge-Discharge site is about 1180 km2. Bina river basin is situated at 24⁰ 09' 36" to 24⁰42' 00" N latitude and 78⁰09' 06" to 78⁰23' 05" E longitude. The major part of the rainfall in the river basin is covered by four Raingauge stations namely; Rahatgarh, Jiasinagar, Begamganj and Gairatganj. The average annual rainfall of the study area is about 1196 mm and the mean minimum and maximum temperature of the region is 11.50⁰C and 40.9⁰Crespectively (Sunil Kumar Yadav, 2012). The study area falls under Vindhyan Region. The important rocks which are found in the area are sand stone, Quartzite sand stone, Lime stone and Deccan traps, called as Basalt. Basalt rocks overlie the Vindhyan sand stone. The land use and land cover in the area mainly comprises of agriculture, forests, settlements, barren land, etc. Agriculture is the main occupation in this region and irrigation requirements are met mainly through groundwater causing groundwater depletion (Sunil Kumar Yadav, 2012). The main crops grown in the area are paddy, oilseeds, wheat, gram and vegetable. 2.2

Methodology

The daily flow data of Rahatgarh G/D site of Bina river for the period of 9 years i.e. from 1990 to 1998,was collected from State Water Data Centre, Water Resources Department, Govt. of Madhya Pradesh, Bhopal. Daily flow data was analyzed to carry out low flow analysis using flow duration curves technique.

A Flow Duration Curve (FDC) is a simple and very useful method of displaying the complete range of river discharges from low flows to flood events. It is a relationship between any given discharge value and the percentage of time that this discharge is equaled or exceeded, or a relationship between magnitude and frequency of stream-flow discharges (Smakhtin, 2001). FDC technique is useful in water resources planning to evaluate dependable flows, river characteristics and its potential. In present study monthly FDC of Rahatgarh gauging site of Bina river were prepared using daily flow data for the period from 1990 to 1998 .Forpreparation of monthly FDC, the stream flow datawas grouped according to different months and was assigned a rank after arrangingdischarge data ofeach month group in descending order of magnitude.The probability of excedance of each discharge data was determined by following equation.

(3.1) Where, P= Probability of exceedance (%) M= Rank number of the discharge event N = Total number of discharge events Finally the monthly FDC were plotted by plotting the values of probability of exceedance against the corresponding discharge data. The flow duration curves can also be plotted using different time resolutions of streamflow data like daily, 10-days, weekly, monthly and annually. 2.2.2

HYDRO 2014 International

Flow duration curves

MANIT Bhopal

Estimation of truncation level

Page 297

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

The variable truncation approach is efficacious in depicting both the drought and wet events and therefore, in describing drought duration and severity (Pandey et. al., 2008). In the present study truncation level has been considered as the value of dependable flow at 75% probability of exceedance. The truncation levels for all 12 months were derived by drawing the monthly FDC using daily flow data of each month. 2.2.3

Low-flow analysis

Low flow is a term used in different meanings by different interest groups. It may be considered as the actual flow in a river occurring during the dry period of the season. However, the effect of changes in the total flow regime of a river on sustainable water yield and riparian ecology may be another aspect of this term. Low flow is a seasonal phenomenon, and an integral component of a flow regime of any river. Drought, on the other hand, is a natural event resulting from a less than normal precipitation for an extended period of time (Smakhtin, 2001). There are natural and anthropogenic factors, which influence the various aspects of the low-flow regime of the river. The natural factors include the soil distribution, infiltration characteristics and hydraulic characteristics; extent of the aquifers; rainfall frequency and amount, evapotranspiration rates from the basin; topography and climate. On the other hand, anthropogenic factors include ground water abstraction, subsurface drainage, deforestation, urbanization, Industrialization, construction of dams and subsequent regulation of a river flow regime. In present study, the low flow analysis was carried out with the help of FDC technique using daily discharge data of Bina river basin at Rahatgarh G/D site to identify the low flow events. The dependable flows at 75% probability of all 12 months were considered as a truncation level of respective months. Iftheriver flow on a particular day is lower than the truncation level, then it wasconsidered as deficitflow or low flow conditionwhereas if the river flow is higher than the truncation level, then it wasconsidered as surpluscondition or high flow.The prolonged deficit flow condition below its truncation level can be considered as the low flow event. Considering the size of the catchment, type of river and on stream demands the events of continuous deficit flow conditions for more than tendayswere considered as the lowflowevent. The analysis was carried out for the 9 years of daily flow data to identify the low flow events and their durations. Severity of each low flow event was taken as the total deficit or cumulative deficient runoff volume below the truncation level during the period of low flow event. 3.

during August whereas the minimum dependable flow of0.77 m3/s during July. The dependable flow at 75% probability was zero during nine months from October to June. From the analysis it can be concluded that Binariver is anintermittentriver having stream flow during monsoon season and 2-3 months thereafter. Figure 3: Flow Duration Curves for August at Rahatgarh

RESULTS AND DISCUSSION

Low flow analysis has been carried out in Bina river basin using daily stream flow dataof Rahatgarh G/D site. Flow Duration Curves technique was used to obtain dependable flow at 75% probability of occurrence which was considered as truncation level. The typical Flow Duration Curve for the August month is shown in Figure 3. The derived truncation levels for 12 months are given in Table 1. From Table 1, it can be seen that the river has maximum dependable flow of28.1 m3/s at 75% probability

HYDRO 2014 International

Table 1: Derived truncation level at 75% dependability levels at site Rahatgarh (m3/s)

The departure of daily stream flow from its truncation level demonstrating low flow conditions persisting for more than ten days periodis shown in Figure 4. From the analysis it can be seen that the Bina river had experienced 6 low flow events over the period of 9 years from 1990 to 1998 indicating 1 or 2 low flow events every year. Analysis was also carried out to obtain deficitflow volume and severity of low flow events as shown in Table 2. The low flow events in the basin usually began during July to September and terminated during November to December. The severity volume of low flow in the river varied from 2.88 to 287.37 MCM and duration of low flow rangedfrom 10 to 22 days. The maximum severity volume of 287.37 MCM was observed for22 days during September, 1994. The years 1991 and 1995 experienced twolow flow events each, which were highest in any year. In the year 1991 two low flow events of total 23 days experienced a total severity of 39.27 MCM. In

MANIT Bhopal

Page 298

International Journal of Engineering Research Issue Special3

ISSN:2319-6890)(online),2347-5013(print) 18-19, Dec. 2014

the year 1995 two low flow events of total 29 days experienced total severity volume of 95.69 MCM. In the year 1992 one low flow event of 16 days experienced total severity of 17.03 MCM. Therefore, on the basis of above analysis, it can be concluded that years 1991 and 1995 were the drought years of severe deficit runoff volume in Bina river at Rahatgarh. The information on frequency of occurrence of low flow and runoff volume deficit in the river is useful in improvement of existing backup practices and to undertake water resources management and development of river basin in systematic manner to meet the various water demands.

December. The truncation approach appears to be more effective in the investigation of drought characteristics of the river system. REFERENCES i. Clausen B and Pearson CP (1995) Regional Analysis of Annual Maximum Streamflow Drought. Journal of Hydrology, 173, 111-130. ii. Galkate RV, Thomas T, Pandey RP, Singh S and Jaiswal RK (2010) Drought Study in Chhindwara District of Madhya Pradesh, India. Third International Conference on Hydrology and Watershed Management (ICHWAM) JNTU, Hyderabad, India. iii. Pandey RP, Sharma KD, Mishra SK, Singh R. and Galkate RV, (2008) Assessment of Stream flow Drought Severity Using Ephemeral Stream flow Data. International Journal of Ecological Economics & Statistics (IJEES), Vol. 11, No. S08, 77-89. iv. Smakhtin VY (2001) Low flow hydrology: a review. Journal of Hydrology240, 147–186. v. Sunil Kumar Yadav (2012) Drought study and water availability assessment in Bina river basin. M.Tech. Thesis, Department of Farm Engineering, Institute of Agricultural Sciences, BHU, Varanasi, Uttar Pradesh, India. vi. WMO (1975) Drought and Agriculture, WMO Technical Note No. 138, Geneva.a

Figure 4: Low Flow events at Rahatgarh

Events 1 2 3 4 5 6 4.

Table 2: Duration and severity of low flow Periods Duration Severity Volume (MCM) 1-10 July, 1991 10 2.88 9-30 September, 1991 13 36.39 1-16 July, 1992 16 17.03 9-30 September, 1994 22 287.37 1-18 July, 1995 18 21.56 20-30 September 1995 11 74.13 CONCLUSIONS

Drought studies can provide an important input to water resources planners for water conservation and water management purposes. Flow Duration Curves technique helps to determine the probability of occurrence of particular flow at the site and dependable flow at various probability levels, which is helpful for planning of water resources projects and deriving truncation levels for specific purposes. Low flow analysis technique helps in assessing the hydrological drought frequency, its duration and severity in the river basin using daily stream flow data. The Bina river in Madhya Pradesh is of an intermittent nature which generally experiences 1 or 2 low flow condition every year. The low flow events in this basin usually begin during July to September and terminate during October to

HYDRO 2014 International

MANIT Bhopal

Page 299