Ensemble simulation of interannual climate variability using the CPTEC/COLA Global Climate Model for the period 1982-1991

The ability of an atmospheric model to simulate the observed climate and its variability varies with scale and variable, with the radiative effects of clouds and the land-surface and sea-air interactions remaining a complex area of study. Given the correct sea surface temperature (SST), most atmospheric climate models (GCMs) (a) simulate the observed large-scale climate showing better skill for some areas as

compared to others, and (b) give a useful indication of some of the observed regional and global interannual climate variations and trends. Even though the ability of the model, submitted to the observed SST, in reproducing the observed mean interannual variability of climate is an important aspects of its performance, part of this variability may be produced by the internal variability of the atmosphere and climate system. The evaluations implemented in this paper include the assessment and validation of selected aspects of rainfall, radiation, circulation, temperature and their interannual variability. The model results are obtained with the atmospheric GCM of the Center for Weather Forecasts and Climate Studies (CPTEC) in Brazil, which is a modified version of the COLA GCM. In order to diagnose the potential predictability of seasonal total precipitation, convection and circulation, an ensemble of 9 members in a 10 years run (1982-91) was implemented with prescribed observed monthly SST.

This study is mainly exploratory as we try to assess the ensemble characteristics of the interannual climate variability simulation, to point out the skill of the model and the predictability at regional scale. The performance of the CPTEC/COLA GCM is analyzed to identify the deficiencies and uncertainties with a view to further model improvement. Skill values of precipitation, convection and circulation anomalies are calculated in equatorial, tropical and extratropical regions. Preliminary results are promising, depicting quite well the circulation and rainfall anomalies during the El Nino’s 1982-83 and 1986-87, and the La Nina’s 1985-86 and 1988-89, especially in regions such as Northeast Brazil, Northern Amazonia, the Indian monsoon, Southern South America and the Pacific sector.