AUTHOR=Shampa  , Muktadir Hussain Muhammad , Nejhum Israt Jahan , Hussain Md. Majurul , Bhowmik Anup , Ananto Nazmul Islam , Aungon Fardin Anam , Biswas Showvik , Islam Mohammad Muddassir 

TITLE=A riverbank erosion early warning system (EWS-RE) for braided river: a case of the Brahmaputra–Jamuna

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2025.1570577

DOI=10.3389/feart.2025.1570577

ISSN=2296-6463

ABSTRACT=Braided rivers are distinguished by numerous bars and channels, along with their regular alterations. This unique character of braided rivers induces bank adjustment through erosion or accretion, and when erosion occurs in populated regions, it results in significant disasters. When such erosion affects inhabited areas, it can cause severe damage. To address this calamity, we introduce here an early warning system for riverbank erosion, named EWS-RE. This system issues alerts based on numerical simulations from a two-dimensional hydro-morphological model of the Brahmaputra–Jamuna River. The forecast model incorporates GloFAS seasonal forecasts and over six decades of historical hydraulic data to define boundary conditions. High-resolution bathymetry of the braided river was generated using an enhanced passive bathymetry technique, which combines satellite imagery with limited in-situ cross-sectional measurements. Model outputs and predicted erosion patterns were validated against actual riverbank changes observed in the year 2019. The system achieved a spatial erosion prediction accuracy of 88% and a sensitivity of 88%, both within a 95% confidence interval. Field testing in the year 2023 showed an overall accuracy of 70% in detecting erosion-prone areas along the river. We anticipate that this framework will contribute significantly to reducing the adverse impacts of riverbank erosion through timely and reliable early warnings.