ABSTRACT Deepor Beel, a Ramsar‐designated wetland in Assam, India, is experiencing severe ecological degradation from urban expansion, industrial effluents, and leachate from the nearby Boragaon solid waste site, threatening its biodiversity and ecosystem services. This study demonstrates the application of Geographic Information Systems (GIS) integrated with remote sensing for continuous and cost‐effective water quality monitoring. Multi‐sensor satellite data (Landsat 8, Sentinel‐2) combined with in‐situ measurements of turbidity, TDS, and Biochemical Oxygen Demand (BOD) were processed within a GIS framework to develop spatiotemporal pollution models. Machine learning algorithms, including Multiple Regression, Artificial Neural Networks, Random Forest, and Gradient Boosting, were embedded in the GIS environment, with Gradient Boosting delivering the highest predictive accuracy for BOD ( R 2 = 0.9778, RMSE = 0.1611). GIS‐enabled mapping identified persistent pollution hotspots near the Boragaon dump and revealed seasonal patterns, with BOD peaking during monsoons (4.74 mg/L in May 2019). A critical outcome was the GIS‐based demonstration that hydrological expansion of the wetland provides negligible dilution, proving that localized anthropogenic loads dominate pollution dynamics. The key contribution of this work lies in developing a GIS‐integrated and empirically validated Machine Learning (ML) framework for high‐resolution wetland water quality monitoring, contextualized for the Deepor Beel ecosystem. Beyond the Deepor Beel case study, the proposed GIS‐integrated remote sensing and machine learning framework is transferable to other data‐scarce wetlands and provides actionable decision support for targeted pollution hotspot management and seasonal water quality risk early‐warning systems.
Talukdar et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: