Riverbank Change Detection in the Jamuna River Basin, Bangladesh: A Remote Sensing and Machine Learning Approach

The Jamuna River Basin in Bangladesh faces severe riverbank erosion, threatening agricultural lands, infrastructure, and livelihoods due to seasonal floods and climate change. The absence of early-warning systems exacerbates these challenges. This study integrates Sentinel-2 imagery, spectral indices, and machine learning within Google Earth Engine to monitor and predict riverbank changes from 2015 to 2023. Advanced cloud-masking, clustering, and predictive models identify erosion hotspots with high accuracy (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> scores of 0.9996–1.0). The methodology enhances shoreline detection and supports proactive mitigation, fostering sustainable land use and community resilience in the Jamuna River Basin.