A dual-phase satellite data simulation system: Framework and preliminary evaluation over China

It is very crucial for developing satellite land data assimilation system by directly assimilating the gridded satellite brightness temperature (TB) data to simulate gridded satellite observation data. A dual-phase satellite data simulation system framework is developed, which consists of the Community Land Model (CLM), microwave Land Emissivity Model (LandEM), Shuffled Complex Evolution algorithm (SCE-UA) and the gridded Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E) TB data, and it is implemented in two phases: the parameter optimization and calibration phase and the satellite data simulation phase. The SCE-UA algorithm is used to optimize the LandEM parameters and calibrate microwave wetland surface emissivity by minimizing the difference between the simulated and observed BT. In this paper, the monthly mean microwave wetland surface emissivity calibrated at HeFei in 2003 are transferred to East Asia region, the dual-phase satellite data simulation system is mainly evaluated over China region. Experimental results indicated that the vertically polarized TBs (6.925 GHz and 10.65 GHz) simulated by the dual-phase satellite data simulation system are basically matched with those observed by AMSR-E sensor and the differences between the simulated and observed TBs are less than 15 K, which indicates that the calibrated microwave wetland surface emissivity possesses excellent transportability and the dual-phase satellite data simulation system is feasible and practical over China. This study provides reference for developing China satellite land data assimilation system by directly assimilating the gridded AMSR-E TB data (low-frequency and vertical polarization) for model grids contained various land cover types, especially for those model grids including wetland cover type, which will greatly improve land data assimilation study.