Geospatial Approach for Evaluating LULC Pattern in the Doddahalla Watershed, Chitradurga District, India

Abstract

The land use and land cover provide essential information for modelling and understanding many Earth phenomena. Regional LULC changes are necessary because it can enhance understanding of the complex interactions between human activities and global climatic change. The research work demonstrates the effectiveness of the normalised difference vegetation index, normalised difference built-up index and normalised difference water index images to map the LULC classes in the Doddahalla watershed. These indices compress the satellite images and are easy to handle the data in an efficient manner. Cloud-free three decadal multispectral images from the Landsat satellite series for the year 1994–1995, 2000–2001 and 2014–2015 for all the major seasons are processed for evaluating the LULC changes in the drought-prone and water-scarce watershed. Survey of India toposheet and high-resolution satellite images from Google Earth are used for the preparation and update the land cover changes. Fourteen LULC classes are mapped up to level III classification system. The classified images are found to be consistent with the original colour composite image. The field verification of classified classes has assessed the accuracy with ground truth points collected across the study area with a representative of the individual land use classes. The classified image has been cross validated with a land use classification image obtained from NRSC (level I classified image from AWiFS acquired on March 8, 2008). The major changes are seen in the cropland, fallow land, forest land, water bodies and built-up land. The fallow lands are converted into built-up land due to the increase in the population density and process of urbanisation.