An Observing System Simulation Experiment for HYDROS Soil Moisture Retrievals

The Hydrosphere State (HYDROS) mission is designed to provide global estimates of surface soil moisture and freeze/thaw state. In this paper we present results of a simulation study of soil moisture products to be derived from the HYDROS mission. The study illustrates different approaches being developed for HYDROS soil moisture retrieval using L-band radar and radiometer observations, and indicates how retrieval accuracy is affected by landcover heterogeneity and vegetation water content (VWC) at different spatial resolutions. The study is based on a modeled geophysical domain in the south-central United States, with simulated landcover, soil moisture, and soil temperature characteristics over a one-month period. Simulated microwave backscatter and emission measurements were computed from the geophysical fields and averaged to the HYDROS radar and radiometer spatial resolutions of 3-km and 40-km, respectively. Representative instrument sampling and noise characteristics were included. The prototype HYDROS retrieval algorithms were applied to the simulated observations, and accuracies of soil moisture products at 3, 40, and 10 km resolutions were estimated. The simulations afford a capability to examine trade-offs of resolution and accuracy, as well as sensitivity to instrument noise and model error. The simulations indicate the degree to which increasing VWC and degree of heterogeneity degrade the retrieval accuracy. The results confirm that the HYDROS 10-km product acuracy goal of 4% volumetric soil moisture is feasible for footprint-averaged vegetation water content up to approximately 5 kg m-2.