Land Cover Classification Using Irs Liss Iii Satellite Image and Digital Elevation Model in Hilly Environment - a Case Study in Nongkhyllem Wildlife Sanctuary, Meghalaya

Abstract

IRS-LISS III satellite imagery corresponding to Nongkhyllem Wildlife Sanctuary area of Ri-Bhoi District, Meghalaya was used for remote sensing analysis of the landuse pattern and vegetation types occurring thereat. Maximum likelihood classifier algorithm of ERDAS Imagine 9.1 version was used to secure supervised classification of pixels into various landuse types and vegetation types among the forest class cover. For the purpose of preparation of training sets thematic maps of the area, and knowledge accruing from extensive personal field visitswere taken aid of. Sample field plots were laid at 30 different locations of the Sanctuary to carry out accuracy assessment. Normalized Difference Vegetative Index value of the LISS III satellite imagery was also computed. Digital Elevation Model of the Sanctuary was erected in the GIS domain. Such GIS database was integrated with remote sensing data in proof of Integrated Geographic Information System capabilities to achieve higher accuracy in classification. There was indeed marked increase in classification accuracy on account of such integration. Bivariate correlation analysis was performed between spectral and DEM variables to cross check the results.