Article ID Journal Published Year Pages File Type
4450582 Atmospheric Research 2010 13 Pages PDF
Abstract

A short-range ensemble precipitation forecast system has been constructed to be applied to over the Mediterranean area down to a 25-km grid spacing using ten members from the fifth-generation Pennsylvania State University-National Center for Atmospheric Research Model (MM5). The ensemble system consists of ten members, each run with a different combination of two different initial conditions from global models and five different subgrid-scale physics configurations for a 2-weeks period of October 2006. The mesoscale verification is made by using observational precipitation data of Spanish Climatic Network. In terms of the total daily precipitation, spatially averaged over the region, time evolution of bias and root mean square error have shown larger departures in the first and final part of the selected period, maintaining quasi-zero values in the middle of the period. Although the created short-range ensemble shows high-spread-skill correlation values for daily precipitation, the asymmetric shape of the rank histogram shows some underdispersion, suggesting a bias behaviour. Confidence intervals have been built to show observational errors, showing more errors in the bins that suggest bias and underdispersive behaviour of the EPS and less errors in the first bins. Additionally, mean spatial forecast rainfall distributions have shown similar patterns to the observed mean precipitation ones, matching the largest accumulated rainfall values and settling down over the same areas. Relative Operative Characteristics diagrams, showing very outstanding areas, and the attribute diagram are indicative of usefulness of the forecasting system, depicting in general a good agreement between forecast probability and the mean observed frequency. Epsgrams of the spatially averaged total daily precipitation for each ensemble member and for the ensemble system illustrate, in general, good agreement among the different members, showing the spread of the ensemble system with departures between them through the time period.

Related Topics
Physical Sciences and Engineering Earth and Planetary Sciences Atmospheric Science
Authors
, , , ,