UNMANNED AERIAL VEHICLES (UAVS) AND ENVIRONMENTAL MONITORING: FEASIBILITY OF MONITORING WATER QUALITY PARAMETERS IN REMOTE LOCATIONS USING UAVS LUISA I. FELICIANO-CRUZ UNIVERSITY OF PUERTO RICO, MAYAGÜEZ CAMPUS DEPARTMENT OF CIVIL ENGINEERING AND SURVEYING

PROBLEM • How can water quality be assessed in remote locations that are not readily accessible?

PROBLEM • If water quality monitoring is inadequate, it poses a risk to public health and safety. • According to UNICEF, the people living in rural areas in the developing countries are the most vulnerable.

PROBLEM

PROPOSED SOLUTION • Current water quality monitoring methods of in-situ and satellite remote sensing can be prohibitively expensive, or not at all possible to perform in remote or inaccessible locations. • This proposal takes advantage of the recent proliferation, availability and affordability of UAVs to overcome this limitation, providing a novel solution to this problem. • This research will provide a means to monitor, document and process optical and spectral water quality data to help determine and forecast water quality parameters in a water network, using the island of Puerto Rico as a Test Site.

BACKGROUND • Traditional water quality monitoring comprises a series of observations, measurements and samples to be collected and analyzed. The most common parameters that affect water quality can be physical, chemical or biological, in nature. • Physical properties of water quality include temperature, turbidity, flow, sediments, and erosion; • Chemical characteristics involve parameters, such as pH, dissolved oxygen, nutrients, oils, metals, pesticides and other pollutants. • Biological indicators, on the other hand, include algae and phytoplankton.

WQ MONITORING: IN SITU • Monitoring is commonly done in-situ and samples are then transported to a laboratory for analysis. • This methodology has several drawbacks. It requires time-consuming field work, highly trained personnel and specialized equipment that can be expensive. • Furthermore, the data obtained through this method is of the point-source type, which implies that it might not be representative of the entire water source of interest because most vary with time, location and extension. • This leads to coarse sampling that is insufficient for capturing fluctuations in water quality.

WQ MONITORING: SATELLITE REMOTE SENSING • Remote sensing refers to the use of aerial sensor technologies to detect and classify objects on Earth by means of propagated signals (e.g. electromagnetic radiation) without making physical contact with the object. • It may be split into active remote sensing (when a signal is first emitted from aircraft or satellites) or passive (e.g. sunlight) when information is merely recorded.

WQ MONITORING: SATELLITE REMOTE SENSING

Total Organic Carbon (TOC) Dissolved Organic Carbon (DOC) Chemical Oxygen Demand (COD) Biochemical Oxygen Demand (BOD) Nitrates (NO3) Nitrites (NO2) 190-350 nm Turbidity (NTU) Chlorophyll opticalnanofilter.com

Total Suspended Solids (TSS) 780-900 nm

WQ MONITORING: SATELLITE REMOTE SENSING • The human eye sees visible light in three bands (red, green, and blue). • Spectral sensors divide the spectrum into many more bands and can be extended beyond the visible to obtain the spectrum for each pixel in the image of a scene with the purpose of finding objects or identifying materials. • Hyperspectral or multispectral sensors are used and differ in the number of bands and how narrow the bands are. Multispectral imagery generally refers to 3 to 10 bands that are represented in pixels, whereas hyperspectral imagery can have hundreds or thousands of bands.

WQ MONITORING: SATELLITE REMOTE SENSING • Airborne and satellite imaging spectrometers can have up to several hundred bands with a spectral resolution of 10 nanometers (nm) or narrower. Compare this to broadband multispectral sensors such as Landsat 8 OLI, which has nine spectral bands and a spectral resolution of 106 nm.

Landsat 8 OLI (9 bands)

http://www.exelisvis.com/docs/HyperspectralAnalysisTutorial.html

AVIRIS (224 bands)

WQ MONITORING: SATELLITE REMOTE SENSING • The spectral resolution required for a specific sensor depends on the spectral characteristics of the material you are trying to identify. • Each material exhibits a unique spectral signature across the electromagnetic spectrum. Factors that influence a material's spectrum include composition, structure, grain size, viewing geometry, and mixture. • The high spectral resolution from a hyperspectral sensor allows you to identify materials, whereas multispectral sensors only allow you to discriminate between materials.

SENSOR QUOTES HYPERSPECTRAL   CAMERA   QUOTES maker

HySpex

model OCI-­‐U AV-­‐ 1000 OCI-­‐U AV-­‐ 2000 Mjolnir-­‐1024

Surface  Optics

710-­‐GX

BaySpec BaySpec

price 26,850 32,200 73,600 25,000

webpage http://www.bayspec.com/spectroscopy/oci-­‐ uav-­‐hyperspectral-­‐camera/ http://www.bayspec.com/spectroscopy/oci-­‐ uav-­‐hyperspectral-­‐camera/ http://www.hyspex.no/products/mjolnir.php http://surfaceoptics.com/products/hyperspe ctral-­‐imaging/710-­‐gx/

MULTISPECTRAL   CAMERA   QUOTES maker Tetracam

model ADC  Lite

price 3,295

Tetracam

ADC  Micro

2,795

webpage http://www.tetracam.com/adc_lite.html http://www.tetracam.com/Products-­‐ ADC_Micro.htm

WQ MONITORING: SATELLITE REMOTE SENSING • There are limitations as to the use of satellite remote sensing in certain areas due to their temporal resolution. • This refers to the length of time it takes for a satellite to complete one entire orbit cycle. It also refers to how often a sensor obtains imagery of a particular area, which can vary from several days to several weeks. • Satellite images are only useful if the satellite’s fixed orbit and resolution by chance, suits the investigation. • Local atmospheric conditions, like cloud cover or pollution, could render satellite data acquisitions unusable, causing a gap in the information collected. • The further away the sensor is from the area being measured, the more difficult it is for it to record position and spectral information accurately.

WQ MONITORING: UAVS?? • The use of UAVs addresses the limitations of other methods by providing real-time, data-ondemand. Drawbacks? Not yet known! • Optical sensors will be onboard a DJI Matrice 100 UAV. It is a state-of-the-art quadcopter rotarywing UAV, that features: multiple expansion bays to allow additional computing power, cameras, communication devices and sensors to be connected; dual batteries to achieve 40 minute flights; carbon-fiber construction for weight reduction and rigidity; maximum payload weight of 1.2 kg; vibration-dampening material to eliminate feedback from the motors, making everything on the craft stable and vibration free; remote controller that includes Lightbridge, integrated GPS, 2 km range and DJI Pilot app support; a revolutionary guidance sensor kit that allows obstacle avoidance; SDK support (software development kit) that makes it programmable and upgradeable. • The petition for exemption under Section 333, requesting FAA authorization to operate a UAV for civil (non-governmental) purposes other than for recreation or hobby is being drafted.

UAV QUOTES maker DJI Horizon   Hobby

model Phantom   3 Chroma

price

payload  

1,259

none

1,299

not   tested   for   payload

flight   time

comments

23   min

photography   use,   non-­‐removable   camera-­‐ gimbal

30   min

use   of   third-­‐party   hardware   not   recommended

in   stock in   stock in   stock

Walkera

TALI  H 500

1,459

not   specified*

25   min

photography   use

3DR

IRIS+

1,600

400   g

16   min   (max   payload)

photography,   video   use photography   use,   upgradeable

3DR

SOLO

1,900

420   g

20   min   (max   payload)

Walkera DJI

Voyager   3 Inspire   1

1,999 2,899

not   specified* 465   g

20-­‐25   min 18   min

photography   use photography   use

DJI

Matrice   100

3,000

1200   g

20   min   (max   payload)

developer   kit,   programmable

DJI

S1000

9,899

5300   g

15   min   (max   payload)

filmmaking

Precision   Hawk Microdrones Sensefly

Lancaster md4-­‐1000 eBee

25,000 30,000 *

1000   g 1200   g *

30-­‐60   min 88   min *

agriculture,   mapping agriculture,   mapping,   inspection agriculture,   mapping,   GIS,  surveying

*   awaiting   response   from   maker

availability ships   in   5-­‐7   days

preorder,   ships   in   6-­‐ 8   weeks in   stock ships   in   1-­‐2   days ships   in   15-­‐20   days ships   in   7-­‐10   days not   specified ships   in   4-­‐8   weeks *

webpage http://www.dji.com/prod uct/p hant om-­‐3 http://www.horizon hob by.com/st orefr ont s/s tore s/chr oma-­‐fami ly http://www.walkera.com/en/g ood s.php? id= 2291 https://store.3drobotics.com/ pro ducts /ir is https://store.3drobotics.com/ t/ sol o http://www.walkera.com/en/g ood s.php? id= 2294 http://www.dji.com/prod uct/ ins pir e-­‐1 http://dev.dji.com/en/prod ucts/f lying-­‐ platf orms /matr ice-­‐100 http://www.dji.com/prod uct/ sprea di ng-­‐w ings-­‐ s1000 http://www.precision hawk.com/ http://www.microdrones.com /en /pro ducts /md4-­‐ 1000 /at-­‐a-­‐gla nce/ https://www.sensefly.com/a pp licati on s/ surveyi ng.html

TEST SITE GOLDEN CREEK. MAYAGÜEZ, PR (MONITORING STATION LOCATION)

• The Golden Creek (18°12'49.48"N, 67° 8'29.68"W) that runs along the University of Puerto Rico Mayagüez Campus at Mayagüez, PR, has been selected as a test site to conduct a preliminary assessment. • It is a tributary of the Yagüez River, which flows into the Mayagüez Bay. • A portable monitoring station will be located at the site. The data will be used for calibration purposes. Source: http://gis.jp.pr.gov/geolocalizador/

METHODOLOGY 1) collecting remotely sensed UAV and in-situ data simultaneously;

2) analyzing data from all available sensors and in-situ measurements, through big data analytics.

3) using image processing and pattern recognition techniques to analyze the images and obtain spectral signatures for the water constituents, using several software packages available on campus for this purpose, such as: MATLAB, ENVI and Geomatica;

4) conducting laboratory analysis on in-situ water samples;

METHODOLOGY 5) using statistical methods to try and correlate the remotely sensed data with the water quality parameter concentrations measured in the laboratory; 6) verifying if the resulting model can predict water quality parameter concentrations, within acceptable error levels; 7) proving the feasibility of monitoring water quality parameters in remote locations using unmanned aerial vehicles (UAVs); 8) reporting findings; 9) extending the study to other river systems in Puerto Rico.

METHODOLOGY

INTELLECTUAL MERIT • The main goal of the proposed research is to determine water quality parameter concentrations on river systems based on remotely sensed data using UAVs. • Although, other researchers have conducted similar studies, they have mostly analyzed coastal waters and have relied on satellite-borne remote sensors for water quality monitoring. Satellite remote sensing, although effective, has limitations in terms of temporal resolution. • The use of UAVs addresses this limitation by providing real-time, data-on-demand. There are a handful of published works that have made use of UAVs for water quality monitoring, as discussed in the literature review section. It must be pointed out that none of the papers reviewed regarding water quality monitoring using remote sensing technology onboard UAVs was able to correlate remotely sensed data with in-situ measurements to predict or determine water quality parameters. • The investigations have been mostly qualitative, rather than quantitative, in nature.

BROADER IMPACTS • The proposed activity will result in an increased awareness of how anthropogenic activity is compromising the planet’s water resources. • The technological developments accessible, via workshops and publications, to a wide range of users who can further develop and exploit the technology into new products, processes, applications, materials and/or services. • Due to the interdisciplinary nature of the proposed work, it provides the opportunity to merge the knowledge of different background areas, such as Electrical Engineering and Environmental Engineering. • Local governmental and environmental agencies could also benefit from, or further develop, the results of this investigation to: reinforce environment protection policies and programs, control pollution, and contribute to the development of water quality standards and guidelines. This effort will aid in the protection and restoration of water quality in remote locations, contributing to public health and wellbeing.

BROADER IMPACTS • In terms of socioeconomic aspects, the use of unmanned systems for water quality monitoring is appealing because it tends to be less expensive and time consuming compared to satellite-borne, manned solutions or in-situ field measuring. • There are, however, some policy related aspects that need to be addressed regarding UAV use. The Federal Aviation Agency (FAA) is currently working on UAV regulations which are expected to be in place sometime during the fall of 2015. • Meanwhile, certain states restrict UAV use to non-commercial purposes, impose low ceilings, and forbid drone use in and around areas of heavy air traffic and/or landmarks. These regulations are currently being revised in order to achieve a more comprehensive UAV integration into the national air space. The issues of protecting the privacy of citizens and the addressing of safety concerns are also being taken into consideration by the FAA.

IGERT INTEGRATION • Since much of the nation’s water infrastructure, both natural and man-made, is being compromised mostly due to anthropogenic activity, it can be considered as an aging civil infrastructure under the Integrative Graduate Education and Research Traineeship (IGERT) Intelligent Diagnostics for Aging Civil Infrastructure program. The research being proposed supports the IGERT Intelligent Diagnostic strategic goal by: • the use of sophisticated sensing equipment for data collection, such as: optical or multi & hyperspectral sensors; • the use of computer modeling tools like MATLAB, Geomatica and ENVI for signal & image processing and pattern recognition; • performing big data analytics due to the large data sets that will be generated, as well as the data fusion needed for multi-sensor data merging

IGERT INTEGRATION • The IGERT traineeship accomplishments under the proposed research will be: • to integrate interdisciplinary backgrounds in Electrical Engineering (such as: remote sensing, pattern recognition, image processing and instrumentation) and Environmental Engineering (water resources, water quality, environmental protection and hydrology); • to broaden knowledge through interdisciplinary short courses, workshops, seminars and field work; • to communicate interdisciplinary research findings through the IGERT Fellows’ Retreat; • to develop the ability to write proposals and interdisciplinary scientific publications.

REFERENCES •

AL-WASSAI, F. A. & KALYANKAR, N.V., “MAJOR LIMITATIONS OF SATELLITE IMAGES”, JOURNAL OF GLOBAL RESEARCH • IN COMPUTER SCIENCE, VOL. 4, NO. 5, PP. 51-59, 2013.

•

ANDERSON, K. & GASTON, K. J., “LIGHTWEIGHT UNMANNED AERIAL VEHICLES WILL REVOLUTIONIZE SPATIAL ECOLOGY,” FRONTIERS IN ECOLOGY AND THE ENVIRONMENT, VOL. 11, NO. 3, PP. 138–146, 2013.

•

CHEN, C., LIU, F. & TANG, S., "ESTIMATION OF HEAVY METAL CONCENTRATION IN THE PEARL RIVER ESTUARINE WATERS FROM REMOTE SENSING DATA," GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2012 IEEE INTERNATIONAL, PP. 2575-2578, JULY 22-27, 2012.

LIM, H. S., MATJAFRI, M. Z., ABDULLAH, K. & WONG, C. J., "TOTAL SUSPENDED SOLIDS (TSS) MAPPING USING ALOS IMAGERY OVER PENANG ISLAND, MALAYSIA," COMPUTER GRAPHICS, IMAGING AND VISUALIZATION, 2009. CGIV '09. SIXTH INTERNATIONAL CONFERENCE ON, PP. 503-509, AUGUST 11-14, 2009.

•

MERZAND, T. & KENDOUL, F.,“DEPENDABLE LOW-ALTITUDE OBSTACLE AVOIDANCE FOR ROBOTIC HELICOPTERS OPERATING IN RURAL AREAS,” JOURNAL OF FIELD ROBOTICS, 2013.

CLEAN WATER ACT OF 1972, 33 U.S.C. § 1251 ET SEQ. (2002). RETRIEVED FROM HTTP://EPW.SENATE.GOV/WATER.PDF

•

ORE, J. P., ELBAUM, S., BURGIN, A., ZHAO, B. & DETWEILER, C., “AUTONOMOUS AERIAL WATER SAMPLING”, GRANT REPORT, RETRIEVED FROM: HTTP://CSE.UNL.EDU/~JORE/PDF/AUTONOMOUS_AERIAL_WATER_SAMPLING.PDF

DJI MATRICE 100 USER GUIDE. RETRIEVED FROM: HTTP://DOWNLOAD.DJIINNOVATIONS.COM/DOWNLOADS/DEV/MATRICE/EN/M100_USER_MANUAL_V1.0_EN.PDF

•

RUIZ-VERDÚ, A., KOPONEN, S., HEEGE, T., DOERFFER, R., BROCKMANN, C., KALLIO, K. & PYHÄLAHTI, T., “DEVELOPMENT OF MERIS LAKE WATER ALGORITHMS: VALIDATION RESULTS FROM EUROPE”, PROCEEDINGS OF THE 2ND MERIS/AATSR WORKSHOP, ITALY, SEPTEMBER 22-26, 2008.

•

•

•

•

•

•

•

ERICKSON, A. J., WEISS, P. T. & GULLIVER, J. S., “WATER SAMPLING METHODS,” IN OPTIMIZING STORMWATER TREATMENT PRACTICES. SPRINGER NEW YORK, JAN. 2013, PP. 163–192. • FRICKER, P., (2005, NOVEMBER), “THE BENEFITS OF AN AIRBORNE DIGITAL SENSOR: AN ADVANCED SYSTEM FOR HIGHRESOLUTION WEB-BASED MULTI-SPECTRAL IMAGERY”, PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, PP. 1243-1244. • GUANTER, L., RUIZ-VERDÚ, A., ODERMATT, D., GIARDINO, C., SIMIS, S., ESTELLÉS, V., HEEGE, T., DOMÍNGUEZ-GÓM EZ, J. A. & MORENO, J. (2010). ATMOSPHERIC CORRECTION OF ENVISAT/ MERIS DATA OVER INLAND WATERS: VALIDATION FOR EUROPEAN LAKES, REMOTE SENSING OF ENVIRONMENT, 114, PP. 467-480. • HADJIMITSIS, D. G. & CLAYTON, C., “FIELD SPECTROSCOPY FOR ASSISTING WATER QUALITY MONITORING AND ASSESSMENT IN WATER TREATMENT RESERVOIRS USING ATMOSPHERIC CORRECTED SATELLITE REMOTELY SENSED IMAGERY”, REMOTE SENS. 3, PP. 362-377, 2011. •

•

HADJIMITSIS, D. & CLAYTON, C., (2011, JULY), “SPECTROSCOPY-ASSISTED SATELLITE WATER QUALITY MONITORING”, SPIE NEWSROOM.

•

KENDOUL, F.,“SURVEY OF ADVANCES IN GUIDANCE, NAVIGATION, AND CONTROL OF UNMANNED ROTORCRAFT SYSTEMS,” JOURNAL OF FIELD ROBOTICS, VOL. 29, NO. 2, PP. 315–378, 2012.

•

LI, J., ZHANG, B., ZHANG, X. & GAO, L., "PRELIMINARY STUDY ON THE POTENTIAL OF SHORT-WAVE INFRARED REMOTE SENSING DATA ON INLAND WATER QUALITY MONITORING," GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, IGARSS '05. PROCEEDINGS. 2005 IEEE INTERNATIONAL , PP. 4541-4544, JULY 25-29, 2005.

LIM, H. S., MATJAFRI, M. Z. & ABDULLAH, K., “ALGORITHM FOR TURBIDITY MAPPING USING DIGITAL CAMERA IMAGES FROM A LOW-ALTITUDE LIGHT AIRCRAFT,” IN PROC. 2ND IEEE INT. CONF. COMPUT. SCI. INF. TECHNOL. (ICCSIT), PP. 200–204, 2009.

SCHAEFFER, B. A., SCHAEFFER, K. G., KEITH, D., LUNETTA, R. S., CONMY, R. & GOULD, R. W., “BARRIERS TO ADOPTING SATELLITE REMOTE SENSING FOR WATER QUALITY MANAGEMENT”, INTERNATIONAL JOURNAL OF REMOTE SENSING, VOL. 34, NO. 21, PP. 7534-7544, 2013. SHEELA, A. M., LETHA, J., JOSEPH, S., RAMACHANDRAN, K. K. & SANALKUMAR, S. P., “TROPHIC STATE INDEX OF A LAKE SYSTEM USING IRS (P6-LISS III) SATELLITE IMAGERY”, ENVIRONMENTAL MONITORING ASSESSMENT, 177, PP. 575-592, 2011. UNICEF, PROGRESS ON DRINKING WATER AND SANITATION 2012 UPDATE, 2012. RETRIEVED FROM: HTTP://WWW.UNICEF.ORG/MEDIA/FILES/JMPREPORT2012.PDF XU, J. P., LI, F., ZHANG, B., GU, X. F., & YU, T., “REMOTE CHLOROPHYLL-A RETRIEVAL IN CASE-II WATERS USING AN IMPROVED MODEL AND IRS-P6 SATELLITE DATA”, INTERNATIONAL JOURNAL OF REMOTE SENSING, 31, PP. 4609-4623, 2010.

•

YANG, C., “A HIGH-RESOLUTION AIRBORNE FOUR-CAMERA IMAGING SYSTEM FOR AGRICULTURAL REMOTE SENSING”, COMPUTERS AND ELECTRONICS IN AGRICULTURE 88, PP. 13–24, 2012.

•

ZANG, W., LIN, J., WANG, Y. & TAO, H, “INVESTIGATING SMALL-SCALE WATER POLLUTION WITH UAV REMOTE SENSING TECHNOLOGY,” IN WORLD AUTOMATION CONGRESS (WAC), 2012, JUNE, PP. 1–4.

QUESTIONS??

This work was supported in part by the IGERT: Intelligent Diagnostics for Aging Civil Infrastructure, under the IGERT Program of the National Science Foundation (Award Number DGE-0654176)