9626380 Horel The National Science Foundation and the National Oceanic and Atmospheric Administration (NOAA) have established a joint program to support research of mutual interest. Generally, research supported under this program involves activities that are interesting research efforts, address a need of NOAA, and ones that can benefit from the resources and computer models of the NOAA National Center for Environmental Predictions. This research will be supported under this joint program and is aimed at evaluating and using the NOAA Eta forecast model and its associated adjoint model that have been developed at the NOAA NCEP. The ongoing development of the mesoscale Eta model makes it possible to evaluate mesoscale weather phenomena that have not been adequately documented before, especially in the western United States, where mesoscale atmospheric motions are tightly coupled to the underlying complex terrain. The Principal Investigator will build and expand on research underway now and continue to evaluate the performance of the mesoscale Eta modal over the Western United States. The adjoint of the Eta model has been developed at NCEP for using an advanced data assimilation method know as 4-dimensional variational (4DVAR) data assimilation. 4DVAR assimilation has been shown in several case studies to lead to improved forecast skill for Eta model forecasts. As this data assimilation procedure begins to be tested on a routine basis at NCEP, additional information will be available to model developers and, eventually, operational personnel regarding the sensitivity of the model forecasts to initial observational error and other factors. The major goal of the proposed work is to participate in the development and application of the Eta adjoint model and associated data assimilation methods. Specifically, the solutions in terms of both the initial conditions and adjoint sensitivities obtained as part of the data assimilation method will be evalua ted as a function of differing synoptic situations. ***
