Use of NDVI and PCA To Analyze Land Use and Land Cover Change in Upper Cauvery River Basin (UCRB), India

Using aerial imagery to classify land use and land cover (LULC) is a crucial method for tracking changes to the earth's surface. Studies show that using conventional ground survey methods to track variation in land use land cover (LULC) requires significant time and labor. The most recent technologies for assessing changes in land cover are geographic data systems and satellite remote sensing data. Due to their low cost, quick turnaround, and reliable results, scientists have used remote sensing and Geographic Information Systems (GIS) techniques extensively. This study utilized images from Google Earth of the Upper Cauvery River Basin (UCRB) in India, along with the normalized difference vegetation index (NDVI) classification and Principal Component Analysis (PCA). The Landsat 7-ETM+ images for the years 2012 and Landsat 9-OLI/TIRS images for the year 2022 were selected. Six different categories—agricultural land, built-up area, shrub land, wasteland, forest, and water body have been used to classify the basin LULC. The outcomes from both classifications showed consistent assessment accuracy and a reasonable degree of agreement. The water, forest, and built-up area are improved by 0.78%, 1.2%, and 1.82%, respectively, and overall accuracy (OA) and kappa coefficient are 91.50 and 0.90, respectively, according to the results. Wasteland, agricultural land, and shrubland are reduced by 0.01%, 2.45%, and 1.33%, respectively. The study's conclusions could be applied to decision-making and the preparation of eco-friendly, evidence-based policies for an urbanizing watershed and other environments with a similar setting, ultimately enhancing the quality of the environment.