Dynamically vs. empirically downscaled medium-range precipitation forecasts

  • For three small, mountainous catchments in Germany two medium-range forecast systems are compared that predict precipitation for up to 5 days in advance. One system is composed of the global German weather service (DWD) model, GME, which is dynamically downscaled using the COSMO-EU regional model. The other system is an empirical (expanded) downscaling of the ECMWF model IFS. Forecasts are verified against multi-year daily observations, by applying standard skill scores to events of specified intensity. All event classes are skillfully predicted by the empirical system for up to five days lead time. For the available prediction range of one to two days it is superior to the dynamical system.

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar Statistics
Author:Gerd Bürger
Series (Serial Number):Postprints der Universität Potsdam : Mathematisch-Naturwissenschaftliche Reihe (paper 126)
Document Type:Postprint
Year of Completion:2009
Publishing Institution:Universität Potsdam
Release Date:2010/07/16
Tag:Models; Scenarios; Verification; Weather-service
Source:Hydrology and earth system sciences 13 (2009), 7, S. 1649 - 1658
Organizational units:Mathematisch-Naturwissenschaftliche Fakultät / Institut für Geoökologie
Dewey Decimal Classification:5 Naturwissenschaften und Mathematik / 55 Geowissenschaften, Geologie / 550 Geowissenschaften
Licence (English):License LogoCreative Commons - Attribution 3.0 unported
Notes extern:
The article was originally published by Copernicus Publications:
Hydrology and Earth System Sciences. - 13 (2009), 7, S. 1649-1658
ISSN 1027-5606