Climatological features represented by the CPTEC/COLA Global Circulation Model

Seasonal climate predictions using GCMs have been performed in some research centers around the world and show results comparable to observations in some specific areas and specific seasons. The model result anomalies depend on the model climatology that must represent the main features of the observed climatology. The interpretation of results needs to take into account the response of the model to monthly observed SST during several years. In order to analyse the CPTEC/COLA model climatology, a set of nine integrations using different initial conditions is performed from november 1981 to november 1991. This model is a spectral model developed by the Center for Ocean, Land and Atmosphere Studies (COLA) and modified by the Center for Weather Forecasts and Climate Studies (CPTEC). Observed monthly SST boundary conditions are applied in the model version with T62L28 resolution. The ensemble seasona results are compared to the NCEP/NCAR reanalyses data, Xie and Arkin precipitation data and OLR measured from satellite. The main features of climatological precipitation are well represented by the model, as those associated with the ITCZ, SPCZ, SACZ and in the Indonesia, Asia and India regions. The position of the Atlantic ITCZ is consistent with observations, displaced northwards in the austral winter and spring and southwards in the austral summer and autumn. In autumn the ITCZ is tilted towards Northeast of Brazil showing its influence on the rainy season of this region. The SACZ is well defined in the austral summer. In the austral winter the lack of convection over Central South America is well represented by the model. The seasonal changes in the atmospheric circulation at high levels are well detected such as the formation of the Bolivian High over South America, the Tropical Atlantic Trough in the austral summer and the Tibetan High in the boreal summer. The right positions of the subtropical jet streams are identified in the Northern and Southern Hemisphere winter season. In this study skill scores are also analysed considering the whole globe, Northern and Southern Hemisphere, South America, Brazil and some Regions of Brazil, for each season. The results obtained show that the CPTEC/COLA model is able to represent the global and regional climatological observed features.