A review of AI-driven Google Earth Engine applications in surface water monitoring, assessment, and management
Jahangeer Jahangeer, Pranjay Joshi, Aditya Kapoor, Zhenghong Tang
Abstract
Geospatial technologies now allow routine observation of lakes and wetlands across large areas, but turning those observations into timely, actionable insight still requires scalable computing. Google Earth Engine (GEE) provides a web-based platform that brings multi-sensor Remote Sensing (RS) archives and parallel processing together in one environment. This review synthesizes how artificial intelligence (AI), machine learning (ML) and deep learning (DL) have been paired with GEE to map and monitor surface water quantity and quality. We summarize recent methods, compare model families commonly used on GEE, and discuss frequent processing pitfalls. To ground the review, we include a case study of three Nebraska lakes (2022-2023) that demonstrates month-to-month tracking of water extent and indicators of water quality. The results demonstrated the effectiveness of GEE in providing timely and accurate insights for surface water monitoring and assessment while also revealing current limitations and opportunities for improvement. Overall, we find that coupling AI methods with GEE can strengthen operational surface water assessment and inform decision-making under increasing environmental pressures. Supplementary Information: The online version contains supplementary material available at 10.1007/s44288-025-00255-x.