AI-Driven Monitoring of the Terrestrial Water Cycle: Evapotranspiration, Soil Moisture, and Drought Under Climate Change

The terrestrial water cycle is a fundamental component of Earth's climate system, governing the exchange of water between land surfaces and the atmosphere. Evapotranspiration, soil moisture dynamics, and drought represent interconnected processes that regulate ecosystem productivity, agricultural yields, and freshwater availability. Climate change is intensifying these cycles, increasing the frequency and severity of drought events worldwide. Traditional monitoring approaches relying on in-situ measurements and physics-based models face critical limitations in spatial coverage, temporal resolution, and computational cost. The rapid advancement of satellite remote sensing — particularly geostationary platforms such as GOES-R providing near real-time observations — combined with machine learning and explainable artificial intelligence (XAI) techniques, now offers transformative opportunities to monitor, model, and predict terrestrial water fluxes at unprecedented scales.This Research Topic aims to advance the integration of artificial intelligence and satellite remote sensing for monitoring key components of the terrestrial water cycle under climate change. A central challenge lies in estimating ET and its sub-components (transpiration, soil evaporation, canopy interception), characterizing soil moisture variability, and detecting drought conditions across diverse ecosystems with high temporal and spatial resolution. Existing products are often limited by coarse resolution, poor temporal frequency, or lack of interpretability. We aim to consolidate advances in physics-informed and knowledge-guided machine learning, near real-time geostationary satellite retrievals, and explainable AI methods (e.g., SHAP) applied to water-cycle variables. By uniting these approaches, the topic seeks to foster methodological innovations that improve drought early warning, agricultural water management, and hydrological forecasting — particularly in climate-vulnerable and data-scarce regions.We welcome original research articles, reviews, and methods papers addressing: - ML and deep learning models for ET, transpiration, and soil moisture estimation using remote sensing data- Near real-time water flux retrieval from geostationary satellites (e.g., GOES-R, Meteosat)- Physics-informed and knowledge-guided ML approaches for hydrological applications- Explainable AI (XAI) methods for drought index development and water stress monitoring- Climate-change impacts on water cycle components and extreme drought events- ET partitioning methods integrating eddy covariance and satellite observations- Regional case studies in semi-arid and Mediterranean climates. Studies combining multiple data sources, multi-scale analysis, and novel uncertainty quantification approaches are particularly encouraged. Topic editors: Eyyup Ensar Basakin, Paul Christopher Stoy, Sadegh Ranjbar Source: https://www.frontiersin.org/research-topics/82481/ai-driven-monitoring-of-the-terrestrial-water-cycle-evapotranspiration-soil-moisture-and-drought-under-climate-change