An evaluation of a general circulation model (GCM) and the NCEP–NCAR reanalysis data for winter precipitation in Greece

• Published in: International journal of climatology 2006-06, Vol.26 (7), p.935-955
• Main Authors: Tolika, K., Maheras, P., Flocas, H. A., Arseni‐Papadimitriou, A.
• Format: Article
• Language: English
• Subjects: canonical correlation analysis; Earth, ocean, space; Exact sciences and technology; External geophysics; global circulation model; Greece; Meteorology; NCEP reanalysis data; principal components analysis; variability; Water in the atmosphere (humidity, clouds, evaporation, precipitation); winter precipitation
• ISSN: 0899-8418; 1097-0088
• DOI: 10.1002/joc.1290

In this study, an attempt is made to evaluate winter precipitation data derived from the general circulation model (GCM)—HadAM3P (atmospherical circulation model—Hadley Center) and from the National Center for Environmental Prediction–National Center of Atmospheric Research (NCEP–NCAR) reanalysis project for the period 1960–1990. These data sets are compared on a seasonal basis (winter) with observational data derived from stations over Greece. More specifically, the suitability of their average distributions and their spatial and temporal variability is examined. The methods used are principal component analysis (PCA) and canonical correlation analysis (CCA). The connection between the precipitation data sets and five cyclonic circulation types, from a daily catalogue, is also studied. With respect to the spatial distribution of the mean precipitation amounts, NCEP data show better agreement with the observational data compared to the modeled data. However, both NCEP and HadAM3P values are generally underestimated. Moreover, it seems that the HadAM3P is able to capture and reproduce quite satisfactorily the relationship between the precipitation data and the circulation types, but inadequately simulates the impact of orography and the Aegean Sea. Finally, the links between the precipitation variability in Greece and the European 500 hPa variability given by CCA seem to be fairly well reproduced by the model. Copyright © 2006 Royal Meteorological Society