Improving Soil Water Content and Surface Flux Estimation Based on Data Assimilation Technique

Abstract

Land surface model is a powerful tool for estimating continuous soil water content (SWC) and surface fluxes. However, simulation error tends to accumulate in the process of model simulation due to the inevitable uncertainties of forcing data and the intrinsic model errors. Data assimilation techniques consider the uncertainty of the model, update model states during the simulation period, and therefore improve the accuracy of SWC and surface fluxes estimation. In this study, an Ensemble Kalman Filter (EnKF) technique was coupled to a Hydrologically Enhanced Land Process (HELP) model to update model states, including SWC and surface temperature (Ts). The remotely sensed latent heat flux (LE) estimated by Surface Energy Balance System (SEBS) was used as the observation value in the data assimilation system to update the model states such as SWC and Ts, etc. The model was validated by the observation data in 2006 at the Weishan flux station, where the open-loop estimation without state updating was treated as the benchmark run. Results showed that the root mean square error (RMSE) of SWC was reduced by 30%~50% compared to the benchmark run. Meanwhile, the surface fluxes also had significant improvement to different extents, among which the RMSE of LE estimation from the wheat season and maize season reduced by 33% and 44%, respectively. The application of the data assimilation technique can substantially improve the estimation of surface fluxes and SWC states. It is suggested that the data assimilation system has great potential to be used in the application of land surface models in agriculture and water management.