Assimilation Techniques to Improve the use of Satellite Water Vapor Radiances

The effective use of satellite water vapor sensitive radiance measurements is essential in producing realistic humidity fields. New assimilation techniques are currently being reviewed with the National Centers for Environmental Prediction's (NCEP) global version of their Gridpoint Statistical Interpolation (GSI) to further enhance the methodology to exploit the infrared and microwave water vapor radiances. Without the proper 3D-VAR constraints to the background error matrix (both static and ensemble-based), observations, and the minimization, negative q and supersaturation solutions can become excessive. We will outline the technique we have developed to improve the radiance assimilation and reduce these negative q and supersaturation solutions in the GSI. Changes to various fields (humidity, temperature, clouds, etc.) in the NCEP Global Data Assimilation System / Global Forecast System analyses and forecasts from using these new techniques will be presented.