Land Surface Temperature Coupled Assimilation Scheme Based On BCC_CSM Climate System Model

At present,most of the land surface data assimilation systems in the world are offline assimilation systems.A coordinated and stable land surface data assimilation scheme,has been a key issue for the current international numerical forecasting centers to promote the development of their coupled climate models.In order to obtain land surface information that incorporates observational information and coordinates with the climate system,this paper uses an ensemble Kalman filter(En KF)algorithm to develop a land surface temperature assimilation scheme suitable for the Beijing Climate Center Climate System Model(BCC?CSM)coupled climate model.The effects and causes of land surface temperature assimilation on other variables in the climate system were analyzed.First,an offline En KF land surface temperature assimilation scheme was developed in the land surface model BCC?AVIM of the BCC?CSM model.The Moderate Resolution Imaging Spectroradiometer(MODIS)land surface temperature observation data was assimilated into the scheme.The assimilation result proved that the scheme can be stably operated in BCC?AVIM,and that the assimilation scheme can effectively reduce the simulated deviation of the model on the land surface temperature.The assimilation scheme proved to be suitable for BCC?AVIM.Second,because polar-orbiting satellite observations have global coverage observation and inconsistent observation times at different locations,in order to analyze the influence of the daily variation information of land surface temperature observation on the assimilation effect,and to select the appropriate observation data time interval for the land surface temperature assimilation,assimilation experiments were performed using observation data at different time intervals.Overall,a relatively good assimilation effect can be obtained when assimilated by observation data at 6-hour intervals in the land surface model.Third,an En KF land surface temperature assimilation scheme was established in the coupled climate model BCC?CSM,and MODIS land surface temperature observation data was used to test the scheme.The results show that the scheme can be stably operated in the coupled model,and the observation information entered the model successfully after assimilation.The deviation and root mean square error are slightly reduced,but there are many areas where the deviation and root mean square error increase after assimilation.This is not as good as the assimilation result of the land surface model,which may be due to the effect of atmospheric variables in the coupled model.Finally,the effects of the coupled model land surface temperature assimilation on other land surface variables and atmospheric variables were analyzed.In the coupled climate model,there is a high-frequency real-time interaction between the land surface and the atmosphere.The assimilation of the land surface temperature not only affects the land surface variables,but also has a close relationship with the atmospheric model variables.The results show that the assimilation of land surface temperature causes the change of atmospheric temperature in each layer from bottom to top,as sensible heat flux and of net radiation.Experiments on representative areas show that,to some extent,the assimilation of land surface temperature can correct the deviation of the precipitation position and precipitation range that occurs in the precipitation simulation.Where there are many missing values in land surface temperature observations,the simulation results may be deviated,which may cause deviation of precipitation results,which in turn will affect the simulation of land surface temperature.This paper developed a land surface temperature assimilation scheme suitable for China's coupled model.The scheme is an improvement on the coupled model simulation results as a whole,as it reduces the simulated deviation of the land surface temperature on a global scale,and also offers some corrections for atmospheric variables that frequently interact with the land surface.