From Sources to Sinks: Advancing Surface Water Management Through Satellite Remote Sensing

Abstract

Surface water is essential for sustaining life, supporting economic activity, and ensuring food security. Historically, since the dawn of remote sensing from space, we have been able to determine where water is located and its spatial extent (Area - L2). However, for deeper process-based insights and to inform decision-making, knowledge of spatial extent is not sufficient. We require knowledge of volume (L3) and fluxes (L3/T) to understand the fate and transport of surface water. This challenge motivates this study, which seeks to answer the overarching question: How can satellite remote sensing be utilized to enhance surface water management?To address this question, the research develops a set of integrated frameworks based on a Source-to-Sink (S2S) perspective, monitoring both sources, where surface water is stored (e.g., lakes and wetlands), and sinks, where it is consumed (e.g., irrigated agriculture). The first component integrates satellite remote sensing and citizen science to develop an innovative, cost-effective methodology for monitoring and managing the volume and flux of surface water. A time-averaged uncertainty metric revealed higher errors in mountainous regions due to shadowing and pixel misclassification. This approach establishes a pre-SWOT (Surface Water and Ocean Topography satellite mission) benchmark for satellite-based monitoring and demonstrates how quantified uncertainty can inform water management decisions, such as designing urban water supply systems for minimum lake volumes and flood infrastructure for maximum levels. Building on this work, the monitoring strategy for surface water is advanced by developing a gauge network optimization framework that integrates hydro-physical factors such as temperature, precipitation, and terrain roughness, along with citizen science elevation data. The framework introduces two novel methods to model lake area-volume relationships, providing benchmark volume estimates for regional and global assessments. A probabilistic mapping approach identifies optimal gauge placements to maximize representativeness and return on investment. Furthermore, the framework supports the integration of satellite-derived virtual stations and facilitates future expansion of gauge networks in coordination with missions such as SWOT. The analysis also showed that strategically deploying around 30 gauges across one-fourth of Bangladesh (Northeastern Bangladesh, characterized by flat topography) could significantly enhance surface water monitoring through more frequent and representative observations in the region. Extending the S2S framework to the domain of water consumption, the next component of this research focuses on irrigation, which is the largest consumer of global water resources. A decision-support framework for gravity-fed canal systems was developed to assist water managers (water providers) in allocating water based on crop-specific demand. By integrating Earth observation data with global numerical weather models, the framework quantifies the net water requirements for each canal within the network. The results demonstrate that satellite-informed allocation can significantly enhance irrigation efficiency, reducing water supply requirements by approximately 26% of the supply. Finally, an open-source Python package that unites both water providers and consumers within a single digital umbrella. Building on earlier efforts that focused primarily on providers, this package integrates satellite, in-situ, and local weather station data thereby extending beyond reliance on global models. Its key contribution is its design philosophy that emphasizes transparency, scalability, and accessibility through the use of open-source tools, publicly available datasets, detailed documentation, and flexible user configurations, ultimately lowering barriers to adoption across diverse agricultural and technological contexts. Collectively, this research highlights the potential of coupling satellite remote sensing with citizen science and cloud-based platforms for quantifying surface water storage (L³) and enhancing surface water management. Developed in consultation with stakeholders, the frameworks operationalize the S2S approach, bridging scientific innovation and real-world implementation. The results offer a scalable pathway toward more equitable, efficient, and sustainable surface water management in a changing world.