9712925 Sanders This project represents a collaborative effort between Dr. Sanders and Dr. Mullen of the University of Arizona (ATM-9714397). It involves an investigation of the impact of initial data uncertainty and model error on, and the usefulness of ensemble forecasting techniques for, medium range (3-5 day) prediction of explosive cyclogenesis. The principals investigators are employing fields from a long-term climate simulation with NCAR's Community Climate Model (CCM2) as well as analyses and operational forecasts of a classic case of explosive cyclogenesis which was observed during the Experiment on Rapidly Intensifying Cyclones over the Atlantic (ERICA) project. Critical issues to be explored include the design of initial perturbations, the minimum number of members for the ensemble and the minimum resolution required for ensemble forecast purposes. The results of this research could be of enormous practical benefit to operational weather forecasting.