A Dynamic Approach to Addressing Observation-Minus-Forecast Bias in a Land Surface Skin Temperature Data Assimilation System

Abstract

Abstract In land data assimilation, bias in the observation-minus-forecast (O − F) residuals is typically removed from the observations prior to assimilation by rescaling the observations to have the same long-term mean (and higher-order moments) as the corresponding model forecasts. Such observation rescaling approaches require a long record of observed and forecast estimates and an assumption that the O − F residuals are stationary. A two-stage observation bias and state estimation filter is presented here, as an alternative to observation rescaling that does not require a long data record or assume stationary O − F residuals. The two-stage filter removes dynamic (nonstationary) estimates of the seasonal-scale mean O − F difference from the assimilated observations, allowing the assimilation to correct the model for subseasonal-scale errors without adverse effects from observation biases. The two-stage filter is demonstrated by assimilating geostationary skin temperature Tskin observations into the Catchment land surface model. Global maps of the estimated O − F biases are presented, and the two-stage filter is evaluated for one year over the Americas. The two-stage filter effectively removed the Tskin O − F mean differences, for example, the Geostationary Operational Environmental Satellite (GOES)-West O − F mean difference at 2100 UTC was reduced from 5.1 K for a bias-blind assimilation to 0.3 K. Compared to independent in situ and remotely sensed Tskin observations, the two-stage assimilation reduced the unbiased root-mean-square difference (ubRMSD) of the modeled Tskin by 10% of the open-loop values.