3M.1
SREPS: Multimodel Ensemble for the Short Range
PAPER WITHDRAWN
Jose A. Garcia-Moya, Instituto Nacional de Meteorologia, Madrid, Spain; and C. Santos, A. Callado, J. Simarro, and D. Santos
Atmospheric predictability is dependent on forecast length and meteorological event. Extreme daily temperatures can be forecast up to several days in advance. Convective precipitation events cannot. Probabilistic forecast may be useful to increase the capability of models to forecast short-range, mesoscale events. The Spanish Weather Service (INM) is building an ensemble prediction system suitable for the short-range (SREPS) in ordre to improve surface parameter forecast (mainly precipitation and wind) and to catch extreme events. Multimodel approach seems to be the best method to get enough spread in the forecast up to 72 hours. As a result of our research, INM SREPS is using four different numerical models (Hirlam, MM5, HRM and UM) with initial and boundaries from four deterministic global models (ECMWF, AVN, GME, UM). Results from a daily pre-operational suite will be show in the presentation. The system will become operational with four runs a day (00, 06, 12 and 18) and a super-ensemble of the last four ensembles will be used as a tool by operational forecasters in Spain.
Session 3M, Mesoscale Processes, Dynamics, and Predictability
Wednesday, 26 October 2005, 10:30 AM-1:00 PM, Alvarado GH
Next paper