Improved Estimation of Evapotranspiration and Recharge from a Dynamic SVAT Model through Assimilation of Microwave Brightness

University Of Florida, Gainesville FL

Investigators

Abstract

0337277 Judge Accurate knowledge of energy and moisture fluxes at the land-atmosphere and at the vadose-saturated zone interfaces is essential for realistic predictions of atmospheric, hydrologic, and crop growth processes. These coupled processes are typically predicted using Soil-Vegetation-Atmosphere-Transfer (SVAT) models. SVAT models have become increasingly sophisticated over the past decade, with biophysically based parameterizations of the land-surface processes. However, even if the biophysics of the system is well-captured by these models, their flux estimates will diverge from reality over time due to lack of proper initial conditions and accumulation of parameterization and computational errors. Because moisture in the root-zone is arguably the most important factor governing the fluxes, a promising way to improve model performance is to assimilate remotely sensed observations that are sensitive to soil moisture. Microwave brightness observations at less than 10 GHz are sensitive to soil moisture in the upper 5 cm for vegetated surfaces with significant biomass. This surface soil moisture is linked to the deeper vadose zone moisture through the SVAT model. The PIs propose to improve evapotranspiration (ET) and recharge estimates by assimilating microwave observations into a dynamic SVAT model. Toward this, a coupled SVAT-Microwave Brightness (MB) model will be developed for the full growing season of cotton grown in north-central Florida. This dynamic SVAT-MB model will be calibrated/validated with observations from a field experiment for one growing season. Microwave observations from a second growing season will be assimilated in the model to improve ET and recharge estimates.