International Journal of Innovative Research and Review ISSN: 2347 – 4424 (Online) An Online International Journal Available at http://www.cibtech.org/jirr.htm 2016 Vol. 4 (1) January-March, pp.26-37/Das and Pal Research Article © Copyright 2014 | Centre for Info Bio Technology (CIBTech) 26 IDENTIFICATION OF WATER BODIES FROM MULTISPECTRAL LANDSAT IMAGERIES OF BARIND TRACT OF WEST BENGAL *Rajib Tarani Das and Swades Pal University Gour Banga, Malda, West Bengal, India *Author for Correspondence ABSTRACT Present work concentrates on extraction of water bodies of the Barind tract of west Bengal using multi- spectral Landsat imageries. This work also aims to investigate the suitability of indices used for extracting water bodies from Landsat imageries. Barind tract is drained a good number of rivers with their complex evolution and shifting of courses. As a result of that 1% of total area is covered by water bodies of different kinds like oxbow lakes, left channel, water logging palaeo-channels, seasonally inundated wetlands etc. over fast changing land use, land cover scenario, the wet lands are also exposed to vulnerability. Climate change issue, changing land surface temperature etc., are also some vital challenges which wetlands are experiencing that are found to be difficult for combating. So, scientific assessment of wetland area change is very vital for monitoring and providing decision support for wetland management. In present work six indices are used for extracting water bodies and each index are compared with higher resolution google earth image based product. Ultimately, it is forwarded that which index is least suited for accurate water body extraction. NDVI, NDWI, MNDWI, NDMI, WRI and AWEI are applied for extracting water bodies and NDWI shows more accurate estimate of water body. If such accurate measurement is possible, certainly it will provide a good data support and based on which water body conservation planning strategy will be devised. Keywords: Water Body Extraction, Indices, Barind Tract, Multispectral Images and Threshold Limit Identification INTRODUCTION Wetlands are complex and rich diverse ecosystem showing land transitional between terrestrial and aquatic system (Cowardin et al., 1979). In a broader sense it is defined as shallow water area or low land or depressions that are inundated or saturated by surface or ground water temporarily or permanently with a prevalence of vegetation, animal and microorganism. Wetlands may be fall into different categories, on the basis of hydrology (shallow marsh, deep marsh, shallow open water), (Sather, 1976) geomorphology (lacutrine, riverine, shoreland wetland and flood plain wetland etc), (Cowardin et al., 1979) abiotic and biotic components etc. (Mitra et al., 2005). Identification of accurate category is essential for assessment of wetlands. At present wetlands cover only 6 % of total land area of the earth, but it contributes significant ecological value as well as economic assistance (Curie et al., 2007). From ecological point of view wetlands carried out essential services like water quality improvement, flooding control, sediment traps, ground water replenishment, retention of phosphorous and nitrogen, and recycling of others element etc (Curie et al., 2007). But its economic benefits are not directly valued and appreciated by community, except some common economic value such as fish production, (Rai, 2008; Liu et al., 2010) water supply to agricultural field, timber production and recreation sites etc (Brander et al., 2006). Growing awareness about the importance of wetlands has drowned attention of many scientist and environmentalist. For last 50 years wetlands remained a vital topic on which numbers of article and research paper have been written. Considering its importance numbers of protocol and policies have been taken at national and international level to conserve wetlands. Yet there is a gap between policy making and implementation of these policies. This is because of unlike developed countries, in developing countries like India; there is lack of