Water quality monitoring and forecasting through integration of geospatial and remotely sensed data

• How can geospatial data from diverse sources, such as satellites, Uncrewed Aerial Vehicles (UAVs), and Uncrewed Surface Vehicles (USVs), be combined to effectively map, monitor, and predict pollution in freshwater and coastal ecosystems?
• How can advanced geo-AI and machine learning techniques be utilized to forecast near-real-time trends in water quality?

Freshwater and coastal ecosystems have experienced significant pollution over the past decade, particularly due to the presence of harmful pathogens such as E. coli and other microbial contaminants. These pollutants often result from intensified industrial and agricultural activities, urban runoff, wastewater discharge, and climate extremes. Pathogens like E. coli pose serious threats to human and animal health, degrade water quality, and disrupt ecosystem biodiversity and aquatic health, making effective monitoring and mitigation critical.

Satellite remote sensing has been extensively used to monitor optically active water quality indicators, such as turbidity and chlorophyll-a, which can serve as proxies for pollution. Advances in geo-AI and cloud platforms have made it easier to conduct time-series monitoring of surface water quality. However, detecting pathogens such as E. coli and other non-optical water quality constituents requires direct in situ sampling, as satellite remote sensing alone cannot estimate these parameters even when enhanced by AI. Furthermore, satellite data must be corrected for atmospheric effects and water column scattering to ensure accurate reflectance, which also depends on in situ measurements. Manual sampling remains costly and hazardous, particularly in contaminated or remote areas. Environmental factors, such as water flow dynamics, wind, and cloud cover, can further limit observation capabilities and introducing uncertainties.

To address these challenges, this research will integrate multimodal, multiscale data from freely available and commercial satellite platforms, Uncrewed Aerial Vehicle imagery, as well as Uncrewed Surface Vehicle. The project will develop a comprehensive framework combining geospatial data and environmental analysis to monitor and map the spatio-temporal distribution of pathogens like E. coli and predict near-real-time pollution trends in both freshwater and coastal ecosystems. This framework will provide actionable insights for proactive management and improved water quality standards.

Prerequisites if applicable

• Knowledge of Remote Sensing;
• Handling geospatial data;
• Coding skills (python);
• Understanding of water quality issues;
• Environmental lab analysis.