Downscalling of the global climate prediction in SE Brazil using RAMS

ain range. Downscalling of the global climate prediction in SE Brazil using RAMS

Pedro L. Silva Dias (1) Ricardo Hallak (2)

(1) Department of Atmospheric Sciences Institute of Astronomy and Geophysics University of São Paulo São Paulo SP Brazil

The Regional Atmospheric Modeling System has been regularly used to generate a downscalling of the climate predictions produced by the Center for Weather Forecasting and Climate Prediction - CPTEC in Brazil. Several methodologies were experimented and best results were obtained with nesting of 2 grids (a coarse grid with 160 km resolution and a fine grid of 40 km). Weak nudging at the interior points and stronger nudging towards the CPTEC ensemble members produced stable results. Specific cases will be discussed, showing the impact of the complex local topography in the SE region of Brazil as well as the role of the SST anomalies along the coast. Higher order statistics of the results are discussed, indicating some skill in predicting local anomalies such as the general characteristics of the prevailing systems, such as frequency of sea breeze penetration and cloudiness along the coastal moun