ABSTRACT
The automatic extraction of surface water bodies from remotely sensed data has been the most important consideration for the investigation of water bodies, natural resource management, and planning with fast and accurate effectiveness. The present study focused on the spatial variation of surface water bodies in 2019 using multi-spectral Landsat-8 OLI satellite data. Satellite-derived indices including Normalized Difference Water Index (NDWI), Modified NDWI (MNDWI), Water Ratio Index (WRI), and Automated Water Extraction Index (AWEI) are considered for the extraction of surface water. The spectral and spatial performance of each classifier is compared using Pearson’s r. Overall, the NDWI is found to be superior to other indexes and may be used to model for the extraction of surface water bodies. The results also indicate a progressive decreasing trend of water surface area in the study site. Subsequently, a significant variation of seasonal coverage (e.g., pre-monsoon and post-monsoon season) is also observed.
