Satellite‐Based Assessment of Land Surface Energy Partitioning–Soil Moisture Relationships and Effects of Confounding Variables

Abstract

Abstract Land surface energetic partitioning between latent, sensible, and ground heat fluxes determines climate and influences the terrestrial segment of land‐atmosphere coupling. Soil moisture, among other variables, has a direct influence on this partitioning. Dry surfaces characterize a water‐limited regime where evapotranspiration and soil moisture are coupled. This coupling is subdued for wet surfaces, or an energy‐limited regime. This framework is commonly evaluated using the evaporative fraction–‐soil moisture relationship. However, this relationship is explicitly or implicitly prescribed in land surface models. These impositions, in turn, confound model‐based evaluations of energetic partitioning‐–soil moisture relationships. In this study, we use satellite‐based observations of surface temperature diurnal amplitude (directly related to available energy partitioning) and soil moisture, free of model impositions, to estimate characteristics of surface energetic partitioning–‐soil moisture relationships during 10–‐20‐day surface drying periods across Africa. We specifically estimate the spatial patterns of water‐limited energy flux sensitivity to soil moisture (m) and the soil moisture threshold separating water and energy‐limited regimes (θ*). We also assess how time evolution of other factors (e.g., solar radiation, vapor pressure deficit, surface albedo, and wind speed) can confound the energetic partitioning–‐soil moisture relationship. We find higher m in drier regions and interestingly similar spatial θ* distributions across biomes. Vapor pressure deficit and insolation increases during drying tend to increase m. Only vapor pressure deficit increases in the Sahelian grasslands systematically decrease θ*. Ultimately, soil and atmospheric moisture availability together play the largest role in land surface energy partitioning with minimal consistent influences of time evolution of other forcings.