| description abstract | This study focuses on testing two different soil moisture analysis systems based on screen-level parameters (2-m temperature T2m, 2-m relative humidity RH2m) and 1.4-GHz passive microwave brightness temperatures TB. First, a simplified extended Kalman filter (EKF) system is compared with an optimal interpolation (OI) method assimilating screen-level parameters in a single-column version of the European Centre for Medium-Range Weather Forecasts (ECMWF) numerical weather prediction model. In the second part of this study, the EKF is applied to investigate whether the synergy of T2m, RH2m, and additionally TB in an assimilation framework improves the simulated soil moisture and atmospheric parameters. For a summer period (130 days) during the First International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment (FIFE) 1987 it is shown that the OI and EKF analysis systems give similar results. Both systems distinguish consistently between periods of atmospheric and surface-controlled fluxes. Though the overall soil water is adjusted by the same amount, the EKF system simulates increments increasing from the first to the third layer, whereas in the OI method they are equally distributed. The EKF system is applied for the Southern Great Plains Field Experiment 1997 (SGP97) testing the assimilation of a synergy of T2m, RH2m, and TB. The observed root zone soil moisture is best simulated by the control run and when TB is assimilated. The assimilation of T2m and RH2m worsens the simulated root zone soil moisture compared to observations, because during a 10-day period modeled T2m and RH2m considerably diverge from observations and soil moisture is tuned to compensate for deficiencies in the model. But in comparison to observed net radiation, heat fluxes, and near-surface soil moisture, it is shown that the assimilation of the synergy of observation types (T2m, RH2m, and TB) gives more consistent results than when they are assimilated separately. | |
