Predictive Analytics for Identifying Surface Water Sources for Domestic Water Supply in Phuentsholing, Bhutan

Surface water is a primary source of drinking water in Bhutan. To ensure a reliable supply thatmeets both quality and quantity requirements, it is crucial to identify suitable sources. This studypresents an integrated approach using Geographic Information Systems (GIS) and machine learningto identify potential surface water sources. Satellite imagery from Landsat-8 and Sentinel-2 wasutilized to generate geospatial datasets. Five key variables influencing the spatio-temporal presenceof water—rainfall, temperature, soil type, Normalized Difference Vegetation Index (NDVI), andtopography were analyzed within a GIS environment. The Random Forest (RF) algorithm, knownfor its robustness in handling nonlinear and high-dimensional data, was employed to predictpotential water sources. Model outputs were validated through field surveys and spectral analysisusing the Normalized Difference Water Index (NDWI). The study identified 50 viable water sourcelocations situated above 450 meters in elevation. The model achieved an area under the curve (AUC)score of 0.99, indicating a strong correlation between predicted and actual water sources. Theseresults confirm that integrating machine learning with remote sensing and GIS is an effectiveapproach for surface water resource planning in Bhutan's hilly terrain.