Integrated remote sensing, machine learning and geospatial approach for site selection of sewage treatment plants in the metropolitan city

Effective wastewater management is vital for environmental and public health in rapidly growing urban areas. This study presents an integrated framework combining remote sensing (RS), geographic information systems (GIS), machine learning (ML), and multi-criteria decision analysis (MCDA) to identify suitable sites for sewage treatment plants (STPs). Key geospatial factors such as slope, elevation, soil type, land use/land cover, drainage, proximity to urban areas, and hydrology were standardized and weighted. A Random Forest (RF) algorithm was applied to enhance spatial prediction. The study area (363.57 sq.km) was categorized into five suitability classes. Only 1.19 sq.km (0.33%) was found to be very highly suitable, with 8.03 sq.km (2.21%) highly suitable, and the majority (90.71%) falling into the low or very low suitability classes. These results highlight the scarcity of optimal land, emphasizing the need for strategic planning. This approach offers a scalable, data-driven tool for sustainable and resilient urban wastewater infrastructure development.