Combined Data Assimilation of Satellite-Based Total Water Storage and Soil Moisture Data to Improve Global Evaporation Estimation

Evaporation is a fundamental process in the global water cycle, playing a critical role in regulating climate, sustaining ecosystems, and managing water resources. Despite its importance, accurately estimating evaporation on a global scale remains a significant challenge due to its spatial and temporal variability and the scarcity of direct ground-based observations, especially in water-limited regions. Satellite observations of key land surface processes offer a potential solution to these challenges, providing consistent and high-resolution observations that can enhance model-based evaporation estimates. In this study, we assimilate observations from the Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-On (GRACE-FO) into the Global Land Evaporation Amsterdam Model (GLEAM). GRACE provides measurements of terrestrial water storage changes by detecting variations in Earth's gravity field, offering critical insights into large-scale hydrological processes that are otherwise difficult to observe. GLEAM, a widely used model for land evaporation, integrates meteorological data, vegetation dynamics, and satellite-based soil moisture to provide comprehensive estimates of evaporation through the computation of its main components (interception loss, bare soil evaporation, and transpiration). GLEAM4 is able to represent groundwater-sourced transpiration, making it suitable for improvements via GRACE data assimilation. More specifically, we investigate the impact of assimilating GRACE data into GLEAM4 and compare its performance across three data assimilation scenarios: (1) when only GRACE data are assimilated, (2) when only ESA-CCI soil moisture data are assimilated, and (3) when both GRACE and ESA-CCI soil moisture data are assimilated. This comparative analysis evaluates the ability of GLEAM to incorporate complementary remote sensing products to better capture evaporation-related processes, thus reducing uncertainties and improving accuracy in global evaporation estimates. Our findings reveal that the integration of both GRACE and soil moisture data can marginally but consistently improve the model’s ability to represent the spatial and temporal variability of evaporation, particularly in water-limited regions, where accurate evaporation estimates are the most needed. This study highlights the potential of combining satellite-based datasets synergistically to address challenges in global evaporation estimation. By advancing the understanding of evaporation dynamics, these results contribute to improved hydrological and climatic assessments and water resource management in the context of climate change.