Remote Sensing, coupled with Geographic Information System (GIS), is a powerful tool for monitoring of water quality and water pollution. Satellite imageries have been used successfully in determination of various water quality parameters like Total Suspended Solids, turbidity, chlorophyll content, colour, temperature etc. by using the Visible, Reflected Infrared and occasionally Thermal Infrared bands of the Electro Magnetic Spectrum. Remote Sensing techniques have been used in sustainable management of water resources, whichinclude runoff and hydrological modeling, flood management, watershed management, drought management and management of Irrigation Command Areas. Satellite imageries obtained from Landsat Thematic Mapper (TM), Linear Imaging and Self Scanning (LISS) and Wide Field Sensor (WiFS) have been used extensively by various workers for calculation of drainage basin area, drainage density, Normalized Difference Vegetation Index (NDVI) and Leaf Water Content Index (LWCI). Application of Remote Sensing in flood management,drought management and Irrigation Command Area management were demonstrated in India by a number of researchers from time to time. Typical applications include mapping of active flood plains, long-range weather forecasting, estimation of crop acreage and crop production and preparation of “irrigability maps” through land use planning.
Keywords: Water Quality, Runoff and Hydrological Modelling, Watershed Management, Flood Management, Drought Management, Irrigation Command Area Management
Sustainable management of the available water resource is a challenging task for the new millennium. As stated by the World Water Council, “There is a water crisis today. But the crisis is not having too little water to satisfy our needs. It is crisis of managing water so badly that billions of people and the environment – suffer badly” (World Water Council, 2000). Remote Sensing techniques have been used effectively in integrated development and management of water resources of India (Balakrishnan, 1986).
Water has very low spectral reflectance in the visible part of the Electro Magnetic Region (EMR) whereas snow or ice has very high spectral reflectance in visible and near infrared (NIR) part of the EMR. Pure water absorbs nearly all incident energy in both the near infrared and middle infrared (MIR) wavelengths. The low reflectance of water in visible and NIR band has advantage in Remote Sensing as water becomes clearly distinguishablefrom either vegetation or soil cover throughout the reflective infrared portion.
Total Radiance (Rt) recorded by a Remote Sensing system over a water body is a function of the electromagnetic energy and is given by the equation:
Rt = Rp + Rs + Rv + Rbwhere, Rp = Atmospheric Path Radiance
Rs = Free-surface Layer Reflectance
Rv = Subsurface Volumetric Reflectance
Rb = Bottom Reflectance
In situ Spectroradiometer measurement of clear water with various levels of clayey and silty soil as suspended sediment shows that the reflectance peak shifts towards longer wavelengths as more suspended sediment is added to the water. Strong chlorophyll a absorption of blue light is observed between wavelengths of 400 and 500 nm and strong chlorophyll a absorption of red light is observed at approximately 675 nm (Lillesand andKiefer, 2000).
Application of visual and digital Remote Sensing techniques and integration of the remotely sensed data in specific layers through the Geographic Information System (GIS) are used by scientists in management of water resources and prediction of natural water related hazards like flood and drought. Visual Remote Sensing has been extensively used in detection of water pollution, lake eutrophication assessment and estimation of flood damage. The technique of visual image interpretation can be used in variety of ways to help monitor water quantity, quality and geographic distribution of water resources (Lillesand and Kiefer, 2000). In the present paper, various methods of application of Remote Sensing in water quality and water resources management are discussed.
The term water quality is generally used to describe whether or not water is suitable for a specific use and whether or not the surrounding environment is endangered by pollutants in the water. Till the middle of the twentieth century, modern economic development largely ignored considerations of water quality, with the result that an inverse relationship has been created between development and water quality. Water quality that is unacceptable to the biota is usually designated as polluted water. Water pollution can be categorizedinto (a) point source pollution in which effluents discharge through pipes and open channels from industrial and human habitat and (b) non-point source pollution (or diffuse pollution) which normally occurs from storm water runoff. Urban, industrial and rural areas are also part of the non-point source pollution in addition to open areas.
The sources of pollution are varied and are often unpredictable, both in time and in magnitude. Remote Sensing has an important role in water quality evaluation and formulation of management strategies, particularly in the case of non-point source pollution. The advantages of Remote Sensing like synoptic coverage, near real time data base generation and availability of multispectral, hyper spectral and multi temporal data, can be used effectively for water quality assessment and monitoring. Nevertheless, the use ofRemote Sensing is limited to surface measurements of turbidity, suspended sediment, chlorophyll, eutrophication and temperature, although experiments to determine various other water quality parameters have been carried out in the past.
REMOTE SENSING AND WATER QUALITY
Spectral properties of water vary with wavelength of incident radiation not only due to the molecular structure, but also due to impurities present in the water body. Hence, Remote Sensing for mapping or monitoring water quality becomes quite complex. The water surface behaves as a partially diffused and partially specular reflector. Specular reflection is uniform at all wavelengths, but absorption and back scatter produce distinctive spectral signature or spectral response pattern. Solar energy that is not specularly reflected is reflected downward into the water body. This refracted energy is either absorbed or get scattered. The remaining signal is indicative of water quality, which is volumetric reflectance or back scattered energy caused by the material in water. In case of deep water, where the bottom reflectance is negligible, the reflectance comes from the surface of water body. However, for shallow water, the total reflectance is a function of both surface and bottom reflectance.
Spectral signature (or Spectral Response Pattern) of water is dependent on a number of factors viz. time of the year, sun elevation angle, aerosol and molecular content of atmosphere, water vapour content of the atmosphere, specular reflection of skylight from water surface, roughness of water surface, water colour and content of dissolved coloured material in water, characteristics of water surface (film, foam, debris or floatingplant), reflectance and absorbance characteristics of suspended particles, multiple reflections and scattering of solar energy in water, depth of water and reflectance of bottom sediment, submerged or emerged vegetation and turbidity of water.
Physical characteristics of water quality can be ascertained through satellite imagery (Moore, 1977). Visible and infrared (reflected) parts of the EMR are most favourable for monitoring of water quality. Thermal Infrared (TIR) is also used occasionally for measuring water quality parameters but the method uses a direct measurement of emitted energy. Microwave region of the EMR is not very useful for water quality assessment due to its limited penetration depth. However, it is useful for detecting oil slicks, oil spills or other surface contamination.
In order to monitor water quality through Remote Sensing, the relationship between water quality parameters and spectral reflectance must be determined. As the reflectance changes with the modified value of water quality parameters, an empirical formula may be used. However, the formula may not be valid in multi temporal domain as the type of constituents in water may not remain constant. Sun elevation angle and atmospheric composition change with time and will affect the relationship between water quality parameter and spectral reflectance.
Multispectral mapping of seagrass has been done where bottom details in clear, calm ocean water have been mapped with penetration of about 20 m between the wavelengths of 0.48 to 0.60 μm using Landsat TM data (Smith and Jensen, 1998). A limitation of this method is that although blue wavelengths have the maximum penetration, extensive scattering occurs resulting in an “underwater haze”. Penetration of only a few metres was obtained using the red wavelength.
Application of Remote Sensing, in most cases, is limited to determination of only a few water quality indicators (Engman et al., 1991). A summary of water quality parameters and the type of Remote Sensing technique useful for water quality monitoring is given in Table 1.
REMOTE SENSING IN WATER RESOURCES MANAGEMENT
Remote Sensing techniques have been used extensively in the past for various applications in the management of global water resource. A summary of such applications in various scientific investigations is given below:
Runoff and Hydrological Modelling
Although it is not possible to directly measure surface runoff by Remote Sensing techniques, they can be used in research areas like (a) determining watershed geometry, drainage network and other map-type information for distributed hydrologic models and for determining empirical flood peak, annual runoff or low flow equations and (b) providing input data like soil moisture or delineated land use classes, which are used for determining runoff co-efficient.
Remotely sensed data, particularly Landsat, Thematic Mapper (TM), Système Pour l'Observation de la Terre (SPOT), MLA data and Indian Remote Sensing Satellite (IRS) data has been used for calculation of drainage basin area and stream network density. Quantitative geomorphic information has been extracted from analysis of Landsat imagery (Haralick et al., 1985).
Remote Sensing can provide attribute data of suitable spatial resolution, which is extremely valuable as model inputs. Panchromatic stereo imagery from IRS-1C and IRS-1D can be used to develop a Digital Elevation Model (DEM) with horizontal resolution of 5.8 m whereas Cartosat-1 panchromatic data can be used with horizontal resolution of 2.5 m. A new technology, known as Interferometric Synthetic Aperture Radar (InSAR) has beenused to demonstrate similar horizontal resolution and about 2 m of vertical resolution (Zebker et al., 1992).
Distributed hydrologic models require information on spatial land use. Most of the research on hydrological modeling using Remote Sensing has involved the Soil Conservation Service (SCS) Runoff Curve Number Model (USDA, 1972). Hydrologic Engineering Center (HEC-1) model was used to demonstrate the efficacy of using Remote Sensing for re-computation of hydrologic variables. HEC-1 model was used in integrating detailed land use data from Landsat TM, which resulted in 90% cost cuts in upgrading dams and spillways constructed on Sable River, Australia (Mettel et al., 1994). Distributed Hydrologic Modeling was done in the USA where a Group Response Unit based on interpretation of satellite data was evoked (Kouwen et al., 1993). Hydrological modeling and GIS has been used in similar studies in small watersheds in India (Hari Prasad et al., 1997).
GIS techniques can be used to integrate spatial data forms like topography and soil maps with hydrologic variables like rainfall distribution or soil moisture. The impact of land use change on Mosel River Basin, USA was quantified by using hydrologically similar units by DEM data, soil maps and satellite based land use data. The satellite data was also used to determine Normalized Difference Vegetation Index (NDVI) and Leaf Water Content Index (LWCI), which were combined to delineate areas where a subsurface supply of water is available to vegetation (Ott et al., 1991, Schultz, 1993, Schultz et al., 2000).
India is one of the most disaster prone geographical zones in the world. It has been estimated that a loss worth more than $300 million (about Rupees 1500 Crore) was incurred annually as a result of damage caused by flood and cyclone in India. In the list of worst flood affected countries, India stands second after Bangladesh (Agarwal and Narain, 1991) and accounts for one fifth of global death count due to floods. It has been estimated that about 40 million hectare (Mha) or nearly one eighth of India’s geographical area is flood prone.
In every aspect of flood management like preparedness, prevention and relief, space technology has made substantial contribution. Information acquired through Remote Sensing covers wide area, periodicity and spectral characteristics and is especially useful in comparing data before and after the flood. The utility of satellite Remote Sensing has been demonstrated operationally for mapping flood inundated areas. Major floods and cyclones that occurred in India were mapped in near real time and information was provided toconcerned agencies. Partially cloud free data was acquired, analysed and interpreted in near real time by IRS series satellites. IRS-1C, IRS-1D, IRS-P6, Cartosat-1, Cartosat-2, Radarsat and Earth Resource Satellite (ERS) were used and are being used for flood inundation mapping, estimation of flood damage and infrastructure loss.
Watershed management is an integral part of any water resources project. The prioritization of a watershed is based on calculation of sediment yield potential so that treatment of the watershed would result in minimizing sediment load into the river or a reservoir. Remote Sensing techniques have been used in a distributed parameter watershed model (Leavesley et al., 1990). Satellite data has been used extensively in many watersheds in India like Jurala (Krishna River Basin), Asan (Yamuna River Basin), Ukai (Tapi River Basin) etc. forderiving the parameters of the Sediment Yield Index (SYI) Model to provide quantitative silt load estimates in watersheds developed by the All India Soil and Land Use Survey, Ministry of Agriculture, Government of India.
Space borne multi spectral data has been used to generate baseline information on various natural resources like soil, forest cover, surface water, groundwater and land use/land cover. Subsequent integration of such information with slope and socio-economic data in a GIS has resulted in generation of location specific management plan for sustainable development of land and water resources within a watershed. A national level project entitled Integrated Mission for Sustainable Development (IMSD) was undertaken by the Department of Space, Government of India, which has covered an area of about 84 Mha spread over 175districts in India. Implementation of rain water harvesting in selected watersheds under the project has demonstrated the benefits by way of increased recharge to groundwater and agricultural development of once barren areas.
The socio-economic life of millions of people is being affected every year by drought, which is one of the worst natural disasters. Due to the highly non uniform distribution of monsoon showers in both space and time, drought has become a menace over many parts of India. It has been estimated that out of the net sown area of 140 Mha, 68% is vulnerable to drought conditions and about 50% of the drought prone area is classified as “Severe” where frequency of drought is bi-annual or less (Rao, 1999).
Timely and reliable information about the onset of drought, its extent, intensity, duration and impact can limit drought related loss of life, minimize human suffering and reduce damage to the economy and environment. Remote Sensing data from geostationary and polar orbiting weather satellites like Indian National Satellite (INSAT), National Oceanic and Atmospheric Administration (NOAA) and other global data are used as major inputs in rainfall predictions ranging from long-term seasonal predictions through medium range predictions to short-term predictions. Vegetation Index derived from satellite imageries are now being used continuously to monitor drought conditions on near real time basis.
National Remote Sensing Centre, Department of Space has introduced a National Agricultural Drought Assessment and Monitoring System (NADAMS), through which early warning is provided on expected yield from major crops at district level by the end of August. The information is updated by the end of September and October with improved accuracy based on the relationship between total crop growth profile and crop yield. The NADAMS uses daily data from NOAA Advanced Very High Resolution Radiometer having 1.1 kmhorizontal resolution and IRS Wide Field Scanner having 188 m horizontal resolution. The high resolution data of bi-weekly or monthly Vegetation Index is used by NADAMS, which provides periodic bulletins and detailed reports to the user agencies. Crop Acreage and Production Estimate (CAPE) programme is operational at district level for providing production forecast before harvest from major crops in winter season. IRS-1D datahaving high spatial resolution and ground based meteorological data are used for forecasting of winter crop production under the CAPE programme.
Irrigation Command Area Management
A major factor in increasing agricultural production is development of irrigation practices, which in turn is essential for economic development of the country. In most of the Irrigation Command Areas (ICA) in India, the present scenario demands a more efficient water management programme. The ICAs in India suffer from problems of inadequate and unreliable water supply, wide gap between created and utilized irrigation potential, temporal imbalances of water demand and supply, excessive seepage loss and rise of groundwatertable leading to water logging and salinity problems.
Remote Sensing techniques can be immensely helpful in inventory of irrigated land, identification of crop types, crop extent, crop condition and estimation of crop yield, as demonstrated in various investigations in India and in other countries. Periodic satellite monitoring of Irrigation Command Areas has helped in evaluating increase in irrigation utilization and improvement in agricultural productivity over a period of time. Remote Sensing methods have been successfully applied in delineating saline and alkaline soils and detecting areas having ineffective water management practices leading to decrease in crop yield. Remote Sensing techniques are now increasingly applied in land use planning and in identifying areas suitable for sustained irrigated cropping with the help of “irrigability maps” prepared from satellite data. Vegetation Indices and demand-supply analysis is used in many Irrigation Command Areas in India to evaluate irrigation potential.
Remote Sensing techniques have been successfully applied for performance evaluation of Mahi Right Bank Canal command with an area of 212,000 ha in Gujarat State (Ray et al., 2002). During this study, IRS-1C Linear Imaging and Self Scanning-III (LISS-III) and Wide Field Sensor (WiFS) data were used for calculation of Adequacy Index, Equity Index and Water Use Efficiency (WUE) for characterization of the Irrigation Command Area. Themulti-temporal Remote Sensing data was finally used for crop inventory, generation of vegetation spectral index profiles and estimation of crop evapotranspiration.
Acknowledgement: The authors are thankful to Sh. A. K. Bhatia, the then Regional Director and Dr. R. P. Singh, Scientist ‘D’, Central Ground Water Board, Uttaranchal Region, Dehradun for constant encouragement and suggestion during preparation of this review paper.
- Agarwal, A. and Narain, S, 1991. Book on Floods, Flood Plains and Environmental Myths, Centre for Science and Environment, New Delhi
- Balakrishnan, P., 1986. A technical report on issues in Water Resources Development and Management and the role of Remote Sensing, ISRO-NNRMS-TR67-86
- Engman, E. T. and Gurney, R. J., 1991. “Water Quality”, Remote Sensing in Hydrology, Great Britain, Cambridge University Press, pp. 175-192
- Haralick, R. M., Wang, S., Shapiro L. G. and Campbell, J. B., 1985. Extraction of drainage networks by using a consistent labeling technique, Remote Sensing of Environment, vol. 18, pp. 163-175
- Hari Prasad, V. 1997. Surface soil assessment using Landsat TM middle infrared band data, Asia-Pacific Remote Sensing and GIS Journal, vol. 9(2), pp. 63-68
- Hari Prasad, V and Chakraborti, A. K., 1997. Hydrological modeling of small watershed using hydrological modeling and GIS, Geographical Infromation System and Remote Sensing Applications, Muralikrishna, I. V. (ed.), Allied Publisher, Hyderabad
- Kouwen, N., Soulis, E. D., Pietroniro, A., Donald, J., and Harrington, R. A., 1993. Grouped Response Units for Distributed Hydrologic Modeling, Journal of Water Resources Planning and Management, 119(3), pp. 285-305
- Leavesley, G. H. and Stannard, L. G., 1990. Application of remote sensed data in a distributed parameter watershed model, Proc. Workshop on Applications of Remote Sensing in Hydrology, Kite, G. W. and Wankiewicz, A. (eds), National Hydrology Research Institute, Saskatoon, Saskatchewan, pp. 47-68
- Lillesand, T. R. and Kiefer, R. W., 4th Ed., 2000. Remote Sensing and Image Interpretation, John Wiley and Sons Inc, New York
- Mettel, C., McGraw, D. and Strater, S., 1994. Money Saving Model, Civil Engineering, 64(1), pp. 54-56
- Moore, G. K., 1977. Satellite Surveillance of Physical Water Quality Characteristics, Proc. 12th International Symposium on Remote Sensing of Environment, University of Michigan, Ann Arbor, pp. 44-61
- Ott, M., Su, Z., Schumann, A. H. and Schultz, G. A., 1991. Development of a distributed hydrological model for flood forecasting and impact assessment of land use change in the international Model River Basin, Hydrology for the Water Management of Large River Basins, Proc. Vienna Symposium, IAHS Publ. No. 201
- Rao, D. P., 1999. Space and Drought Management in Asia-Pacific Region, Space Forum, vol. 4, pp. 223-247
- Rao, D. P., 1999. Water Resources Management, UN ESCAP/ISRO Science Symposium on Space Technology for Improving Quality of Life in Developing Countries: A Perspective for the Next Millennium, New Delhi
- Ray, S. S., Dadhwal, V. K. and Navalgund, R. R., 2002. Performance Evaluation of an Irrigation Command Area using Remote Sensing: A Case Study of Mahi Command, Gujarat, India, Agricultural Water Management, vol. 56, 2, pp. 81-91
- Schultz, G. A., 1993. Hydrological modeling based on remote sensing information, Advance Space Research, 13(5), pp. 149-166
- Schultz, G. A. and Engman, E. T., 2000. Remote Sensing in Hydrology and Water Management, Springer-Verlag, Berlin, Germany
- Smith, F. G. F. and Jensen, J. R., 1998. The Multispectral Mapping of Seagrass: Applications of Band Transformations for Minimization of Water Attenuation Using Landsat TM, Technical Papers, ASPRS-RTI 1998 Annual Conference, American Society for Photogrammetry and Remote Sensing, Bethesda, MD, 1998, pp. 592-603
- Zebker, H. A., Madsen, S. N., Martin, J., Wheeler, K. B., Miller, T., Lou, Y., Alberti, G., Vetrella, S. and Cucci, A., 1992. The TOPSAR Interferometric Radar Topographic Mapping Instrument: IEEE Trans. on Geoscience and Remote Sensing, vol. 30, pp. 933-940
D. Bagchi & R. Bussa
Central Ground Water Board, Dehradun