Study Of Assimilating FY3C/MWRI Soil Moisture Based On CLDAS

Soil moisture is an important parameter on the land surface,and plays an important role in climate prediction and weather forecast.Both land surface hydrological process model and satellite remote sensing can help to obtain soil moisture in a large area.Data assimilation can combine the advantages of model simulation and remote sensing to improve the simulation of soil moisture.However,the instantaneity of remote sensing in bias correction was ignored in previous studies.And the improvement on skill metrics is overemphasized,while ignoring the true information captured by remote sensing.Focus on these two points,the FY3C/MWRI soil moisture product is assimilated into the CMA Land Data Assimilation System using three time strategies to apply bias correction.Assimilating results are used for irrigation detection ignored in land surface model.The main conclusions are:1)Calculate the CDF of Noah-MP based on three matching strategies to match the soil moisture of model simulation and remote sensing observation.With the improvement of matching degree between Noah-MP and MWRI,CDF curves changed in the areas of western and northern China.M_cdf of model CDF calculated from precise matching strategy is less than that of all-time strategy about 0.024 overall in China.2)Assimilation of MWRI/VSM leads number reducing of stations with high accuracy,and number raising of stations with moderate accuracy.It's because of the relative low accuracy of MWRI/VSM.3)When only stations with significantly positive correlation between MWRI/VSM and ground observation soil moisture are considered,the precise matching strategy with highest correlation coefficient of surface soil moisture in assimilation experiments is better than other two strategies.Assimilation can increase the correlation coefficient of soil moisture at the site by about 0.13 when Noah-MP model skill is 0.2?0.3.But when Noah-MP model skill is greater than 0.3,assimilation reduces the soil moisture correlation coefficient at the site by about 0.03.However,there are too much stations where Noah-MP model skill if greater than 0.03,causing accuracy reducing of soil moisture in overall stations.4)Assimilation satellite observation of soil moisture can be used to detect irrigation events.Based on new wet-dry events,we found that irrigation occurred in Hubei.According to investigations,a drought occurred in Hubei during the period of study,and artificial irrigation of paddy fields is expected to be carried out in summer,all evidence proves that irrigation process detected in Hubei has certain credibility.