Remote Sensing and Machine Learning Approaches for Assessing Environmental Dynamics in the Southeastern Watersheds of Madre de Dios, Peru

Abstract

This study evaluates environmental changes in the southeastern basins of Madre de Dios, Peru, using multispectral remote sensing and machine learning techniques. Landsat and Sentinel imagery were processed to compute the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI). Principal Component Analysis was employed for dimensionality reduction, while unsupervised K-means clustering was used to classify land cover. Results reveal a marked reduction in dense vegetation alongside growing extents of bare soil and water bodies, underscoring the effectiveness of our approach for comprehensive environment monitoring and change detection.