Experiments of soil moisture data assimilation system based on ensemble Kalman filter

摘要

Ensemble Kalman filter is a new sequential data assimilation algorithm which was originally developed for atmospheric data assimilation. It calculates background error covariance matrix using Monte-Carlo method and is able to resolve the nonlinearity and discontinuity existed within model operator and observation operator. When observation data are assimilated at each time step, background error statistics were estimated from the phase-space distribution of an ensemble of model states were used to calculate the Kalman gain matrix and the analysis increments. In this work, we develop a one-dimensional soil moisture data assimilation scheme based on ensemble Kalman filter and simple biosphere model (SiB2) to assimilate soil moisture observation. We also do some assimilation experiments using GAME-Tibct observation data from July 6 to August 9,1998, at the MS3608 site on the Tibetan plateau. Once every 12 hours, in situ observations of soil moisture at the depth of 4, 20, 100 cm are assimilated into land surface model (SIB2) and the best estimations of soil moisture at the surface layer, the root zone and the deep layer are calculated. The results indicate that data assimilation can significantly improve the soil moisture estimation in the surface layer, the root zone and the deep layer. And we think that the Ensemble Kalman filter is both practical and effective for assimilating in situ observation into land surface models.