The Impact of Soil Moisture Initialization on the Direct Assimilation of Satellite Radiance Data

This study conducted direct assimilation experiments of microwave remote sensing data AMSU-A with the mesoscale numerical weather prediction model WRFV3.3.1 and its 3DVAR system. Our numerical simulation and assimilation experiment research focused on a heavy rainfall event that occurred at Jiangxi Province on June 19, 2010. We changed the soil moisture value in the initial field of the model and analyzed the impact of improved soil moisture accuracy on the model simulation and the directly assimilated emissivity data. Moreover, we adjusted the deviations in the observational data and the background conditions under different soil moisture conditions. The results show that:The output of the China Land Soil Moisture Data Assimilation System (CLSMDAS), after adjusting the soil moisture initial condition, simulates brightness temperature values that are much closer to actual observations. Much more observational data can be entered into the assimilation system after quality control and bias correction, so the improved soil moisture initial condition can be positively adjusted, especially in window channels such as band 3, with a frequency of 50.3 GHz. Soil moisture from the output of the CLSMDAS can better represent the trend of the rain belt, the drop zone of heavy rain, and the rain center and intensity. All these show us that more accurate initial soil moisture values can improve the results of satellite data assimilation, and thus increase the numerical model forecasting capability.