Hyper-resolution Land Surface Modeling and Earth Observation: Bridging Data Gaps for Improved Plant-Soil-Water Representation in Earth System Models

Abstract

Due to soil moisture and vegetation's critical role in controlling land-atmosphere interactions, detailed and accurate hydrological and ecological information is essential to understand, monitor, and predict hydroclimate extremes (e.g., droughts and floods), natural hazards (e.g., wildfires and landslides), irrigation demands, and weather and climate dynamics. While in-situ soil moisture and vegetation biomass measurements can provide detailed information, their representativeness is limited, and networks of sensors and field data are not widely available. Satellite observations offer global coverage, but retrievals can be infrequent or too coarse to capture the local extremes. This observation data gap often limits the use of remote sensing information for locally relevant processes understanding and policy implementation. To bridge this gap, recent advances in hyper-resolution land surface models (LSMs) – operating at the ~100m to 1km resolution – along with the ever-increasing availability of satellite and big environmental datasets, machine learning, and high-performance computing, provide a pathway forward to bridge data scales. This study highlights recent advancements in plant-soil-water dynamics in LSMs through improved realism in hyper-resolution land surface models and satellite land data assimilation. Previous research enabled soil moisture information at unprecedented scales across continental domains, exemplified by SMAP-HydroBlocks – the first 30m resolution satellite-based surface soil moisture dataset in the United States. To refine the representation of plant-soil-water dynamics at hyper-resolution scales, we introduce the assimilation of monthly 250m resolution vegetation carbon biomass estimated from satellite observations, such as canopy height (GEDI) and leaf area index (MODIS), into the vegetation dynamics of the NOAA GFDL Earth System Model. We assess how vegetation biomass assimilation impacts root zone soil moisture, runoff, vegetation biomass, surface temperature, and evapotranspiration. By improving the spatiotemporal accuracy of plant-soil-water process estimates at the local scales, we can quantify and map their spatiotemporal variability and scaling behavior, which is critical to identifying freshwater-dependent ecological hotspots. Such advancements contribute to improved capabilities to detect droughts, their impacts on crop yields and ecosystems, and understand their interactions with climate at spatial scales required for many water and food security applications (~100s meters). In a rapidly evolving climate, bridging such data gaps and harnessing the ever-increasing potentials of satellite observation and Earth system models will continue to play a critical role. Integrating these technologies and methodologies offers a robust framework for responding to contemporary challenges in plant-soil-water processes and vegetation and surface water monitoring.