Initialization of Soil Moisture in a Global Climate Model: A North American Case Study

Accurate initialization and forecasting of land surface soil moisture in fully-coupled global climate models is critical for seasonal-to-interannual climatological and hydrological prediction, because of its feedback to precipitation and atmospheric circulations, through its control on partitioning of the land-atmosphere water and energy fluxes. To properly initialize the land surface in the NASA Seasonal-to-Interannual Prediction Project (NSIPP), a one-dimensional Kalman filter assimilation of near-surface soil moisture observations has been added to the Catchment-based Land Surface Model (CLSM) used in the NSIPP coupled prediction model. In this project, the CLSM is run off-line from the atmospheric and ocean simulation models, forced by monthly-mean observation corrected European Centre for Medium-Range Weather Forecasts (ECMWF) re-analysis data. Near-surface soil moisture observations from the Scanning Multi-channel Microwave Radiometer (SMMR) satellite 6.6GHz (C-band) channel, covering the period 1979 to 1987, are assimilated using an extended Kalman filter to correct for soil moisture forecast errors resulting from incorrect initial conditions, inaccurate meteorological forcing data and imperfect model parameterization. The soil moisture estimates from the assimilation are compared with a limited number of ground-based point measurements of soil moisture; 19 stations in Illinois, 6 stations in Iowa and transect of 89 points in New Mexico.