As part of the effort by the Environmental Modeling Center (EMC) of the National Centers for Environmental Prediction (NCEP) to upgrade and evolve the land-surface and hydrology package of the NCEP mesoscale Eta Model, the problem of initializing soil moisture content is being addressed by the testing of a data assimilation method.
The proposed data assimilation scheme consists of variational methods wich apply the adjoint of the land-surface and hydrology model and observational products matching any or some of the model's output or state variables. One such state variable, the skin temperature, is available from GOES-derived surface products.
The tests are performed with the land package running off-line and validation is obtained through the surface fluxes observed at specific sites from the NOAA/ARL Integrated Surface Irradiance Study (ISIS).