DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

Thesis submitted to the Graduate School, Faculty of Geography, Gadjah Mada University in partial fulfillment of the requirements for the degree of Master of Science in Geo-information for Spatial Planning and Risk Management.

UGM By: Budi Hadi Narendra 19537/PS/MGISPRM/06 17511

Supervisors: 1. Dr. H.A. Sudibyakto, M.S. 2. Prof. Dr. V.G.(Victor) Jetten

GADJAH MADA UNIVERSITY INTERNASIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION 2008

THESIS DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

By: Budi Hadi Narendra 19537/PS/MGISPRM/06 17511 Has been approved in Yogyakarta On: February 11st, 2008 By Team of Supervisors: Chairman:

External Examiner:

Dr. Junun Sartohadi, M.Sc.

Dr. Pramono Hadi, M.Sc.

Supervisor 1:

Supervisor 2:

Dr. H.A. Sudibyakto, M.S.

Prof.Dr. V.G.(Victor) Jetten

Certified by: Program Director of Geo-Information for Spatial Planning and Risk Management, Graduate School Faculty of Geography, Gadjah Mada University

Dr. H.A. Sudibyakto, M.S.

Disclaimer This document describes work undertaken as part of a program of study at the Double Degree International Program of Geo-Information for Spatial Planning and Risk Management, a Joint Program of ITC the Netherlands and UGM, Indonesia. All views and opinions expressed therein remain the sole responsibility of the author, and do not necessarily represent those of the institute. Budi Hadi Narendra

Abstract Drought is one of slow onset natural hazards that has the greatest impact and affect in many sectors, include agricultural. To cope or manage the drought, people must be familiar with drought characteristics happen in the area. The characteristics should provide information about drought vulnerability showed in a map, spatial and temporal aspect of the drought, as well as water deficit volume during drought events. This research tries to explore drought characteristics in agriculture area of Gesing sub watershed based on meteorological and soil characteristics. The first approach is using rainfall data and Standardized Precipitation Index (SPI), and the second is defined by soil moisture drought modeling using PC Raster. Finally, the correlation between these methods was analyzed to know the differences. The research result reveals that annual rainfall characteristics can describe drought occurrence at that year. The drought years classified using rainfall anomaly by MGA are significantly correlated with droughts based on SPI 12-month time scale in December, as well as SPI 1-month time scale has a high and significant correlation with monthly rainfall deficiency. Soil moisture modeling generated using PCRaster can describe drought characteristics based on soil moisture deficit with flexible time scale. PCRaster output provides information about when, where, and how much water deficit occur in each time step. In daily time scale, soil moisture is closely linked with rainfall for time lag of one day. In monthly time scale, Drought information provided by SPI is less suitable in assessing agricultural drought compared modeling in PCRaster. Using SPI, a drought can be identified by showing negative value of SPI one month time scale. Keywords: drought, Standardized Precipitation Index (SPI), soil moisture modeling

i

Acknowledgements Firstly, I will start by saying Alhamdulillaahi rabbil ´aalamiin to Allah SWT for the Will, Guidance, and Permission such that I can finish my MSc study. My sincere thanks are for my supervisors Dr. Sudibyakto, M.S. and Prof. Dr. Victor Jetten for their valuable support, kind advises comments and particularly their enormous encouragements. I would like to thank all the UGM Geography Faculty and ITC members who directly or indirectly helped me in the successful completion of my degree, essentially for Geo-Information for SPRM management. I am also thankful to the Meteorological and Geophysical Agency (BMG) of Jawa Tengah and Water Resource Agency (BPSDA) Probolo for providing me the rainfall and meteorological data. I am appreciative of my employer, Forestry Department for providing me to pursue higher studies. Thanks to the Bappenas (National Development Planning Agency) for the scholarship program since in EAP course until finishing the study. Also for the NEC (Netherlands Education Centre) in Jakarta who gives STUNED fellowship therefore I have opportunity to study and stay abroad as well as get great experiences in some Europe countries. Warm thanks go to my entire classmates who gave me corporation and a lot of happiness during the class, exam, “wiskul”, and holiday in UGM and ITC. Their support, help, love, care and friendship were valuable and unforgettable. I would like to give my special thanks to Arif, Maya, and Estu as my fieldmates during thesis fieldworks in Purworejo. As nice friends, they had been very helpful, cooperative and very fruitful discussion during fieldwork. I will always remember the hospitality provided by Arif’s family during stay at their house. Gratefully thank to Mas Rahman, Mone, Nugroho, Rino, Mas Safrudin, and Rudi for their help in software and modeling, as well as to Ebta, Pak Hosen, Wulan, Muktaf, Defi, Dody, Bu Lily, Firda, Utia, Anna for their warm and friendly discussing during the entire study. Not only to my friends during the study but also I thankful to all of my friends and collages in Yogyakarta and Enschede who has accompany me during spend my time in sport, traveling, and holiday. You all have made my live pleasurable and unforgettable. Finally, I would like to express my heartfelt gratitude to my family, my wife and beautiful daughter, as well as my parents, sisters and brothers for their eternal encouragement which led to successful completion my study. I cannot express my thankfulness to them in words, I can say that it’s only because of their love, support and blessings that I gained the strength to complete this study.

Budi Hadi Narendra Yogyakarta, Indonesia January, 2008

ii

List of Content

1. 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 2. 2.1. 2.2. 2.3. 2.4. 3. 3.1. 3.1.1. 3.1.2. 3.1.3. 3.1.4. 3.2. 3.3. 3.4. 3.4.1. 3.4.2. 3.4.3. 3.4.4. 3.5. 4. 4.1. 4.2. 4.3. 4.4. 5.

Abstract ………………………….…………………………………………………………… Acknowledgement ………...……………………………………………………………… List of Figures ………………………………………………………….……………………. List of Tables ………….……………………….………………………………………..... List of Appendices …..…………...………………………………………………………. Introduction ..................................................................................... Background...................................................................................... Problem Statement ........................................................................... Research Objectives ........................................................................ Research Questions …....................................................................... Overview of Research Methodology ………………………………………………. Structure of the Thesis ………………………………………………………………… Literature Review ............................................................................ Drought Definitions ........................................................................... Drought Monitoring and Drought Characteristics .…….………………………. Standardized Precipitation Index (SPI) .……..……………………………………. Soil Moisture Drought ……………………………………………………………………. Methodology ………………………………………………………………………………… Study area …………………………………………………………………………………... Geography …………………………………………………………………………………… Climate ………………………………………………………….…………………………….. Soil ……………………………………………………………………………………………... Agriculture Area ………………………………………………………………………….… Rainfall Analysis ……………………………………………………………………………. SPI Calculation ……………………………………………………………………………... Soil Moisture Drought Modeling ……………………………………………………... The Use of PC Raster …………………………………….…………………………….… Determining Potential Evapotranspiration …..……………….…………………... Soil and Crop Characteristics …..…….…………………….………………………... Preparing PC Raster Inputs …………………………….………………………………. Correlation Analysis ……….……………………………………………………………… Result and Discussion ………….………………………………………………………... Rainfall Characteristics …..…………..………………….…………………………….. SPI Analysis ………………………………………………….……………………………... Soil Moisture Drought Modeling …………………………………………………….… Correlation of Soil Moisture drought and SPI .….…………………………….. Conclusion and recommendation ……………………..……………………………… References …………………………………………………………………………………… Appendices …...……………………………………….…….……………………………...

iii

i ii iv v vi 1 1 1 2 2 2 3 4 4 5 6 7 10 10 10 11 12 12 13 14 13 14 15 20 22 23 24 24 29 36 42 50 51 54

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

Conceptual framework of research ………………………………………….…… Sequence of drought occurrence and impacts …………….…………….….. Whole area of Gesing sub watershed …………………………………….…..…. Rainfall and meteorological stations surround study area ………………… Average monthly rainfall from 3 rainfall station for 27 year record …….. Soil type map of study area ………………………………………………………… Crop type on agriculture area ……………………………………………………… Soil sampling points …………………………………………………………………… Trend line of annual rainfall …..……………………………………………………… Average monthly rainfall 1980 – 2006 from each station …………….….. Trend line of dry month start ….…………………………………………………... Trend line of dry month number averaged from 3 stations …..………….. Trend line of rainy season start ….……………………………..................... Different drought frequencies showed by different SPI time scales ….. SPI 12 on December and average annual rainfall …………………….….… SPI 6 on April and 6-month moving average rainfall ……………….……. Rainfall zone map ………………………………………………………………….…... Crop type map ……………………………………………………………………….….. DEM map ……………………………………………………………………………….…. Ksat map …………………………………………………………………………….……. Porosity map ……………………………………………………………………….……. Field capacity map ……………………………………………………………….……. Wilting point map ………………………………………………………………….…… Correlogram of soil moisture 2002 ………………………………………........... Correlogram of soil moisture 2006 ………………………….…………........... Lag cross-correlation between rainfall and soil moisture ……………....... Drought occur in paddy field ………………………………………………………… PCRaster output maps compared with SPI maps ……………………….…..

iv

3 5 10 11 11 12 13 20 24 25 27 28 29 31 32 34 36 37 37 38 38 39 39 40 41 41 42 48

List of Tables 2-1 2-2 3-1 4-1 4-2 4-3 4-4 4-5 4-6 4-7 4-8

Classification of SPI values ……………………………………………………………… Water content characteristics of various soil texture classes ……..…….…. General soil description in Bogowonto watershed ………………….............. Drought years based on MGA classification …………………………………….… Statistic descriptive of monthly rainfall of the 27 year record ……............. Anova of dry month number …………………………………………………..…….… Correlation of dry month start, dry month number and annual rainfall …... Drought category for each SPI time scales …………………………................ Annual rainfall characterized by SPI 12 and MGA classification …………... SPI 1 and 3-month time scales for 2002 and 2006 ………………….............. Coefficient correlation of drought based on MGA class with monthly rainfall, SPI 1, and SPI 3 …………………………………………………................. 4-9 Meteorological (non rainfall) data for 2002 and 2006 ……………………….. 4-10 Drought occurrences based on rainfall, SPI, and soil moisture deficit …... 4-11 Correlation of soil moisture deficit with SPI ……………………….................

v

6 8 12 25 26 27 28 29 33 35 35 35 43 44

List of Appendices Appendix Appendix Appendix Appendix

1 2 3 4

Appendix Appendix Appendix Appendix

5 6 7 8

: Start of dry period and average dry month number ……………..… : Correlation between dry month number and annual rainfall…..… : Start of rainy season in each station ………………………………….… : Monthly rainfall, normality, and SPI values in 2002 and 2006 for each station …………………………………………………………………..… : Result of soil texture analysis ……………………………………………... : Soil characteristics each sampling point …………………………….… : PCRaster script for the generation of soil moisture deficit ………… : Correlation of soil moisture and soil characteristics …………………

vi

54 54 55 56 57 58 61 65

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

1. Introduction 1.1. Background Drought differs from other natural hazards, because drought is a slow onset natural hazard but it can have the greatest impact and affect large number of people. Drought must be considered as a relative, rather than an absolute condition. It can occur in both high and low rainfall areas. Commonly the main cause of a drought is the lack of precipitation for extensive period of time. Through hydrological process, this precipitation lack will make lower in soil moisture, ground water, and streamflow. As slow onset event, drought effects may accumulate over time and may remain for several years. The impact of drought results from the shortage of water or the unbalance between supply and demand for water as one of environmental aspect, until agricultural, economic and social aspect. Gesing sub watershed is a part of Bogowonto watershed located in central Java. This area often faces drought disaster; even the same areas received enough total rainfall in a year. Almost every year, mass media reports drought occur in this area causing crop failure and water scarcity for domestic purposes. Increasing drought preparedness is important to minimize drought disaster effects. As the slow onset disaster, drought allows a warning time between the first indications, usually several months, to the point where the population will be affected. To cope or manage the drought, people must be familiar with drought characteristics happen in the area therefore it is important to investigate the drought. The drought characteristics should provide information about drought vulnerability showed in a map, spatial and temporal aspect of the drought, as well as water deficit volume during drought events.

1.2. Problem Statement In the lower part of Gesing sub watershed, agriculture is a primary economic sector. The water deficit is often the most limiting factor for crop production. Both long term and short term drought has severe impacts on agriculture. Even though most area in the watershed receives enough rainfall in a year, but in dry season some areas start to become drought. In drought mitigation actions and programs, it is important to understand the drought characteristics through drought analysis. It consists of reliable information as a main factor in the decision making process. A drought analysis based only on rainfall data is often done because in many areas the rainfall data are more available than other meteorological or remote sensing data. Unfortunately this analysis does not directly provide soil moisture deficit during a drought which is very important for agricultural plant growth and yield. Spatial soil moisture modeling is expected to provide the drought characteristics more efficient and applicable in agricultural sector. The information is essential for a broad group of users within the geo-informatics society who are interested in monitoring, mitigation and management of drought. 1

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

1.3. Research Objectives The general objective of the research is to explore drought characteristics in agriculture area of Gesing sub watershed based on meteorological and soil characteristics. Specific objectives of the research are the followings: - To analyze drought characteristics using rainfall data and Standardized Precipitation Index (SPI) - To analyze drought characteristics by defining soil moisture drought model using PC Raster - To analyze the correlation between drought characteristics performed by SPI and soil moisture drought model

1.4. Research Questions 1. How far rainfall data and SPI can explain drought characteristics? 2. How do we transform a spatial soil moisture model to drought characteristics? 3. How is the correlation between drought characterized by SPI and soil moisture model?

1.5. Overview of Research Methodology The following figure shows briefly summarized schematic work flow of the various steps that are undertaken to achieve the research objectives.

2

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

Temperature, relative humidity, windspeed, sun duration, radiation

Initial soil moisture, porosity, saturated hydraulic conductivity, organic matter, texture

Crop factor: type, rotation, coefitient

Percolation, wilting point, soil moisture

Potential evapotranspiration

Daily rainfall

Infiltration

Monthly rainfall Actual evapotranspiration Time series analysis

SPI analysis

Soil moisture Calibration

Actual soil water content

Spatial Soil moisture

Soil moisture drought map

Correlation

Meteorological drought map

Figure 1-1: Conceptual framework of research

1.6. Structure of the Thesis This thesis contains six chapters. The first chapter highlights the background, objectives and research questions. Chapter two provides with a literature review of the concepts of drought, monitoring, and uses of SPI and soil moisture model. The third chapter details the study area and methodology considered in order to achieve the research objective. Chapter four presents the results obtained and discusses them. The last chapter draws conclusion of this study and gives recommendation for further research.

3

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

2. Literature Review 2.1. Drought Definition According to Wilhite & Glanz (1985), drought can be explained as conceptual (encyclopedia type) or operational definitions. The latter define drought characteristics using a variable based on the discipline of interest, for instance precipitation in meteorological drought, streamflow in hydrological drought, or soil moisture in agricultural drought. Originally, drought is caused by lack of precipitation mentioned as meteorological drought. After that, drought transmits through hydrological system into soil moisture drought. In an agriculture area, this drought can decrease agricultural products and in a forest area it will increase the trigger of forest fire. Further more, the drought can develop into groundwater and streamflow drought, both known as hydrological drought. In this step, there is not enough surface water, even groundwater to supply drinking water, irrigation, industrial, or hydropower. Other authors also make drought terms based on the consequences such as socio-economical drought (Wilhite and Glantz, 1985; Tallaksen and Van Lanen, 2004). Meteorological drought is usually defined by a precipitation deficiency threshold over a predetermined period of time. The threshold chosen, such as 75 percent of normal precipitation, and duration period, for example, six months, will vary by location according to user needs or applications. Agricultural drought is defined more commonly by the availability of soil water to support crop and forage growth than by the departure of normal precipitation over some specified period of time. There is no direct relationship between precipitation and infiltration into the soil. Infiltration rates vary, depend on antecedent moisture conditions, slope, soil type and the intensity of the precipitation event. Soil characteristics also differ: some soils have a high water-holding capacity while others do not. The latter are more prone to agricultural drought. Hydrological drought is even further removed from the precipitation deficiency since it is normally defined by the departure of surface and subsurface water supplies from some average condition at various points in time. Similar to agricultural drought, there is no direct relationship between precipitation amount and the status of surface and subsurface water supplies in lakes, reservoirs, aquifers and streams because these hydrological system components are used for multiple and competing purposes, such as irrigation, recreation, tourism, flood control, transportation, hydroelectric power production, domestic water supply, protection of endangered species and environmental and ecosystem management and preservation. There is also a considerable time lag between departures of precipitation and the point at which these deficiencies become evident in surface and subsurface components of the hydrologic system.

4

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

Socio-economic drought differs markedly from the other types of drought because it reflects the relationship between the supply and demand for some commodities or economic good, such as water, livestock forage or hydroelectric power, which depend on precipitation. Supply varies annually as a function of precipitation or water availability. Demand also fluctuates and is often associated with a positive trend as a result of increasing population, development or other factors.

Figure 2-1: Sequence of drought occurrence and impacts (NDMC, 2007)

2.2. Drought Monitoring and Drought Characteristics Drought monitoring is done in order to identify climate and water supply trends. It can detect and predict the occurrence and severity of drought. This information is very important to reduce the drought impact if it can be distributed in right time and format, as well as followed preparedness plans. An effective drought monitoring must integrate precipitation characteristics and other climatic parameters with water information such as stream flow, groundwater levels, reservoir and lake levels, and soil moisture into a comprehensive assessment of current and future drought. There are three distinguishing features in a drought occurrence: intensity, duration and spatial coverage. Intensity refers to the amount of the precipitation shortfall. It is generally measured by the departure from normal of a climatic parameter such as precipitation, an indicator such as the reservoir level or an index such as SPI (World Meteorological Organization, 2006). Different types of drought require different drought indicators. In the agricultural drought monitoring, the most suitable indicators needed are the factors that are 5

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

responsive to soil moisture status as a result of soil water balance process. Soil moisture deficit has critical relation with crop water requirements and it will be important in assessing the impact of drought on crops. Another essential characteristic of drought is its duration. Droughts can develop quickly in some climatic regimes, but usually require a minimum of two to three months to become established. The magnitude of drought impacts is closely related to the timing of the onset of the precipitation shortage, its intensity and the duration of the event. There are many tools to identify drought characteristics. The choice depends on hydroclimatology of the region, the type of drought, the vulnerability of society, the purpose of the study and the available data. The lack of a standard definition, making this choice is subjective (Hisdal et al., 2004).

2.3. Standardized Precipitation Index (SPI) The Standardized Precipitation Index (SPI) is a tool developed in 1993 by Tom McKee, Nolan Doesken and John Kleist in Colorado Climate Centre with the main purpose to defining and monitoring drought. Compared with PDSI (Palmer drought severity index), SPI is a more simple tool because it just based on rainfall data and less calculation effort. Basically the SPI is the number of standard deviations that the monthly rainfall data would deviate from the long-term mean. Firstly, a transformation is applied to make rainfall data follow a normal distribution (McKee et al., 1993). Hayes et al. (1999) used the SPI to monitor the 1996 drought in the United States of America. They show how the SPI usefully can detect the start of the drought, its spatial extension and temporal progression. They also show that the onset of the drought could have been detected one month in advance of the Palmer Drought Severity Index (PSDI). The SPI can be computed for different time scales, can provide early warning of drought and help assess drought severity, and is less complex than the Palmer index. Table 2-1: Classification of SPI values (McKee et al., 1993) SPI Values 2.0 and more

extremely wet

1.5 to 1.99

very wet

1.0 to 1.49

moderately wet

-.99 to .99

near normal

-1.0 to -1.49

moderately dry

-1.5 to -1.99

severely dry

-2 and less

extremely dry

SPI was formulated to calculate rainfall deficit in multiple time scales. The time 6

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

scale shows drought impact caused by different water sources. Drought caused by soil moisture deficit was a respond of rainfall shortage in relatively short time scale, while groundwater, streamflow, and reservoir storage reflect longer rainfall anomalies. For several purposes, McKee design SPI for 3, 6, 12, 24, and 48-month time scales and then classified the drought class as shown in Table 2-1. The droughts occur if SPI reaches -1.0 value or less. Every drought event can be calculated for the duration, intensity, and magnitude (NDMC, 2007). There was a study focused on three analyses: relationship between NDVI and SPI at different time scales, response of NDVI to SPI during different time periods within a growing season, and regional characteristics of the NDVI SPI relationship. The result shows that the 3-month SPI time scale has the highest correlation to the NDVI, because the 3-month SPI is the best way for determining drought severity and duration (Ji and Peters, 2003).

2.4. Soil Moisture Drought Deficit of precipitation and high evapotranspiration are the important factors causing soil moisture drought which usually occur next step after meteorological drought. Water deficit actually is not only caused by insufficient water input into hydrological system but also on the rate of water losses through evapotranspiration, discharge from the area, or by various human activities. Evapotranspiration is also an important factor in drought because it causes water loss almost at the same time and place with precipitation occurrence. The potential evapotranspiration can be determined by interaction of meteorological factors such as temperature, humidity, wind speed, radiation and plant types. The actual evapotranspiration depends on catchment’s characteristics such as land use, soil, and water table (Tallaksen & van Lanen, 2004). How much water can be held by soil depends on several factors. The most important are soil texture, structure, and organic matter. In the soil itself, water is held around soil particles and organic matter, also in soil pores with different potential. When soil is saturated with water, it has no more water potential and there is available water free. With time, some of the water from saturated soil will drain to the underlying layers of soil (Bureau Land Management, 2003). Representative values for various soil textural classes are presented in Table 2-2.

7

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

Table 2-2: Water content characteristics of soil texture classes (USDA, 1955)

Medium sand Fine sand Sandy loam Fine sandy loam Loam Silt loam Clay loam Clay

Water per 30 cm of soil depth Field capacity Permanent Plant-available wilting point water capacity % by cm/30 % by cm/30 % by cm/30 weight cm weight cm weight cm 6.8 3.0 1.7 0.7 5.1 2.3 8.5 3.7 2.3 1.0 6.2 2.7 11.3 5.0 3.4 1.5 7.9 3.5 14.7 6.5 4.5 2.0 10.2 4.5 18.1 8.0 6.8 3.0 11.3 5.0 19.8 8.7 7.9 3.5 11.9 5.2 21.5 9.5 10.2 4.5 11.3 5.0 22.6 10.0 14.7 6.5 7.9 3.5

Aggregation is closely related to biological activity and organic matter content in the soil. Organic matter acts as the “glue” to hold the framework of soil particles and pores together, and can build a stronger internal and superficial structure in the soil profile to a condition allowing easy entry of water and its storage in plant-available form. Organic matter in the form of mulch and leaf litter can also be a significant protection against surface sealing by raindrops. Under favorable conditions, the cloud droplets fall to the surface as precipitation (P). Over land areas, where P is greater than ET and the excess, called runoff (R) occur. Under certain circumstances a part of the excess water infiltrates to the deeper soil layers. Infiltration (I) is not easily determined, so for practical purposes it is better to consider a column which extends from the surface to a depth where significant vertical exchanges are already absent. In general, the form of the water balance, including also the net change in soil moisture content (ΔS) is given by: ΔS = P - Roff + Ron - ET - percolation For the calculation of the available soil moisture content, expressed in precipitation mm, the simplified form of the water balance equation was used. Because the study area is flat we can assume the runon and runoff are negligible. After the simplification for the calculation a reduced form of the equation was used. The upper one meter layer soil moisture content (SMC) in the next time unit (SMCi) will be expressed as a function of the previous soil water content: SMCi = SMCi-1 + P – ET In both cases, the time unit is one month. The maximum water a soil can contain is its porosity: all pores are full. Then quickly water drains through gravity until the soil reaches a moisture content called field capacity. This happens in few days so that water is not considered available to plants called as the wilting point (WP). After that he soil is too dry for plants to survive and only in an over can we extract more water. Thus the "plant available water" is the maximum value of available water content of the examined soil layer (AWC) can be expressed as AWC = FC – WP. 8

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

The two methods of soil moisture computation differ only in estimation of the ET term, but both concerning the potential evapotranspiration and also in the way of derivation of real evapotranspiration as a function of the potential one, as well (Sze, et al., 2005). The single-layer water balance model is a commonly used tool in New Zealand for soil moisture assessments, irrigation management, and pasture production studies. The modeling objective is generally to estimate the water status of particular soils, without the need for (or with a minimum of) field measurements, particularly during periods when growth-limiting water stress is likely. A further important application is the derivation of moisture deficit parameters for drought incidence and severity studies, such as comparing drought occurrence and risk between growing seasons, or between regions of the country (Porteous et al., 1994).

9

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

3. Methodology 3.1. Study Area 3.1.1. Geography The study area is located in Gesing sub watershed as a part of Bogowonto watershed located in south part of Central Java Province. In the north part, Gesing sub watershed is bordered by Mongo sub watershed, the south part by Ngasinan sub watershed, the west part by Bogowonto Hilir sub watershed, and the east part by Kulonprogo district. The total area of Gesing watershed is 49.63 km2. The sub districts (kecamatan) included in Gesing sub watershed are Bagelen, Bener, Kaligesing, Loano, and Purworejo.

389000

390000

391000

392000

396000

397000

N

Ngaran

study area Pandan Rejo

Kedung Gubah

9145000

9145000

9146000

9147000

Gunung Wangi

395000

Kali Gono

Brenggong Cangkrep Kidul

9144000

Tlogoguwo Kali Harjo Hulosobo

Semawung

9142000

9142000

Projection: UTM Datum: WGS 1984 Zone: 49S 390000

391000

392000

393000

river

0

2 Kilometers

road 394000

395000

396000

Figure 3-1: Whole area of Gesing sub watershed

10

397000

9141000

389000

Dono Rejo

Somongari

Kemanukan Piji

9143000

Pacekelan

9144000

Plipir Ganggeng

9146000

9143000

394000

9147000

9141000

393000

9148000

9148000

Based on PSBA – UGM (2004) and Balai PSDA Bogowonto, Gesing sub watershed is included in drought vulnerable area. The study area is focused on agricultural land, mainly located in lower part (downstream) of Gesing sub Watershed.

389000

390000

391000

392000

393000

394000

395000

396000

Kedungputri(86)

9146000

$ Keradenan(50) 87.5

G

es in

g

9143000

25.0

50.0

391000

392000

# Rainfall station

393000

394000

Land cover: Mix plantation Settlement Agriculture area 395000

396000

9141000

390000

1 Kilometers

Projection: UTM Datum: WGS 1984 Zone: 49S

Cengkawak (27.5)

$ Meteorological station Road River Contour line

87.5

0

#

389000

75.0 87.5 87.5

9142000

9142000

.5 37

87.5

9143000

riv er

50.0

7

6 0 5.

#

9144000

9144000

2.5

Kaligesing(100)

9145000

9145000

9146000

9147000

N

9147000

9141000

397000

#

9148000

9148000

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

397000

Figure 3-2: Rainfall and meteorological stations surround study area

3.1.2. Climate For 27 year record (1980-2006), the rainy season in 3 stations (Cengkawak, Kaligesing, and Kedungputri) are usually started in October until April, and the dry season is in May until September (see Figure 3-3). Annual rainfall has a variation from 1146 mm until 3855 mm. Based on temperature calculation the maximum temperature is 31.1°C and the minimum is 23.1°C. Relative humidity is between 80 – 90 %. 450

rainfall (mm)

400 350 300 250 200 150 100 50

Ju l A y S ugu ep te st m b O er ct ob N ov er em D be ec em r be r

Ja nu Fe ary br ua ry M ar ch A pr il M ay Ju ne

0

Figure 3-3: Average monthly rainfall from 3 rainfall stations for 27 year record 11

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

3.1.3. Soil Based on soil type distribution, Bogowonto watershed can be divided into 3 areas, as shown in Table 3-1. Table 3-1: General soil description in Bogowonto watershed Soil type % area Alluvial 31.9 Regosol 5.03 Latosol 63.07 Source: PSBA – UGM,

Productivity Low - high Low – high Medium - high 2004

Use Agriculture, settlement Agriculture, plantation Agriculture

389000

390000

391000

392000

393000

394000

395000

396000

397000

N

River Road Grey Alluvial Dark Brown Latosol Yellowish Red Latosol 0

389000

1 Kilometers

390000

391000

392000

393000

394000

395000

396000

9148000 9147000 9146000 9145000 9144000 9143000 9142000 9141000

9141000 9142000 9143000 9144000 9145000 9146000 9147000 9148000

Specific soil type in study area is revealed in figure 3-4. In this study, the soil physical properties as used in the PCRaster water balance model are obtained directly from soil samples.

397000

Figure 3-4: Soil type map of study area

3.1.4. Agriculture Area Agriculture area is dominated by paddy field with terrace system. There are 3 types of cropping pattern: paddy followed by paddy with area 324.7 hectare, paddy followed by groundnut with area 53.3 hectare, and paddy followed by tobacco with area 241.6 hectare as shown in Figure 3-5. The first crop type is always paddy started usually in November depends on rainy season start, and then followed by the second crop in March.

12

389000

390000

391000

393000

394000

395000

396000

397000

87.5

9147000

N Brenggong 5 62. 87.5

50.0

er riv es in g

87.5 Pacekelan 87.5

G

9144000

Plipir 7 5. 0

Proyeksi: UTM Datum: WGS 1984 Zone: 49S

Piji 389000

1 Kilometers

390000

391000

392000

393000

394000

Land cover: Mix plantation Settlement Paddy - Paddy Paddy - Groundnut Paddy - Tobacco 395000

396000

9141000

9141000

0

9142000

9142000

25.0

87.5

Semawung

9143000

Village border Road River Contour line

Kemanukan 5 0 .0

9144000

.5 37

Kali Harjo

9145000

Ganggeng

9145000

7

0 5.

9146000

9146000

Cangkrep Kidul

9147000

9143000

392000

9148000

9148000

DROUGHT MONITORING USING RAINFALL DATA AND SPATIAL SOIL MOISTURE MODELING

397000

Figure 3-5: Crop type on agriculture area

3.2. Rainfall Analysis Daily rainfall data for 27 year (1980-2006) were obtained from Meteorological and Geophysical Agency (MGA) and Water Resource Agency (BPSDA) Probolo. The data consist of daily rainfall from 3 stations located as shown in Figure 3-2. In these stations, rainfall data were collected using manual rain gauge (ombrometer). From daily rainfall, the data were tabulated into monthly data and then started to analyze. The rainfall analysis included rainfall difference among 3 stations, variability and frequency of dry months, changing of dry period start and duration, as well as start of rainy season. The data were prepared in Excel program whereas the statistical analysis was done by SPSS program. Dry month are classified refer to Oldeman classification who classified agroclimate for agricultural crop based on average number of wet month (P>200 mm/month), and average number of dry months (P