Use of the Gridpoint Statistical Interpolation for hourly assimilation within the Rapid Refresh system
Dezso Dévényi, NOAA/ESRL/GSD and CIRES-University of Colorado, Boulder, CO; and M. Hu, S. S. Weygandt, S. G. Benjamin, and C. R. Alexander
The Gridpoint Statistical Interpolation (GSI) is an integrated analysis program and part of the Rapid Refresh system, intended to replace the operational Rapid Update Cycle at NCEP in 2010. GSI has been developed by NCEP in cooperation with other institutions. The version of GSI for Rapid Refresh is a 3-dvar analysis with modifications for a 1-hour analysis cycle. The most significant modification to GSI for the Rapid Refresh is the addition of a non-variational cloud analysis scheme developed at the Global Systems Division and the Center for Analysis and Prediction of Storms of the University of Oklahoma (Hu et al., this conference). We are planning to extend this scheme to operate under the GSI variational framework in the future.
Other modifications include preprocessing of surface observations to account for elevation difference between model and actual terrain, scaling observation innovations near shorelines of oceans and lakes to avoid unrealistic innovation values, and use of mixed-layer height from the model 1-h forecast background to propagate surface information aloft in a well mixed boundary layer. These additions to GSI have been tested both in retrospective runs and real time cycles. The presentation will provide detailed description of these modifications and their impact on the analysis will be presented.
Session 12, Assimilation of observations (ocean, atmosphere, and land surface) into models III
Thursday, 21 January 2010, 11:00 AM-12:15 PM, B207
Previous paper Next paper
Browse or search entire meeting
AMS Home Page