Validação da precipitação do modelo BAM em eventos de ZCAS

Abstract

One of the main meteorological variables for defining the climate of each region is precipitation. In Brazil, each region has a characteristic rainfall regime, according to the atmospheric systems that operate. Regions such as southeastern Brazil, for example, where there is a high demographic, industrial and commercial concentration, the lack of rain causes chaos in the society and economy of the country. On the other hand, when excessive precipitation occurs, society suffers from the disruption of cities in the face of high rainfall levels. Facts like these are very common when systems such as the South Atlantic Convergence Zone (SACZ) occur, a deep convection zone that acts on the southeast and midwest of Brazil. The SACZ occurs during the extended summer, from October to March and is essential for the maintenance of the pluviometric regime in the region. However, the representation of precipitation by numerical models requires calibrations for Brazilian conditions. In view of this, this study aimed to analyze the rainfall simulations of the Brazilian Global Atmospheric Model (BAM), during the period from 1990 to 2013 and compare with observed data, in order to evaluate the model’s performance in the representation of rain in Brazil . The data used in this study were, daily data of precipitation, wind at 200hPa and 850hPa of the BAM model (1°x1°), observed daily data of precipitation from Brazil, interpolated in high spatial resolution (0.25°x0.25°) and daily wind data from the ERA-Interim reanalysis (0.75°x0.75°). The analyzes were divided into two chapters, first the climatology of precipitation in Brazil was addressed, during the year, during the rainy season and during 149 ZCAS events reported in the analyzed period, then the simulations of the BAM model in the representation of rain and atmospheric circulation during two selected ZCAS events. In the analyzes of the first chapter, the BAM simulations consistently represented the spatial distribution of the rainfall, but there was an underestimation of the amount of accumulated rain. The results obtained in chapter 2, the BAM simulations represented the precipitation and atmospheric circulation during the ZCAS events, but with peculiarities of the model, such as the delay in the configuration of the atmospheric circulation patterns expected during ZCAS events, compared to the reanalysis data.