Performance of the CPTEC/COLA global model during SACZ episodes using kuo and relaxed arakawa-schubert deep convection schemes / Desempenho do modelo global CPTEC/COLA durante episódios de ZCAS, utilizando os esquemas de convecção profunda tipo kuo e arakawa-schubert relaxada

AUTOR(ES)
DATA DE PUBLICAÇÃO

1999

RESUMO

One of the most important meteorological phenomena during summer in South America is the South Atlantic Convergence Zone (SACZ). Several studies have been accomplished in order to understand the atmospheric dynamics involved in its formation. In this work, the performance of the Center for Weather Forecast and Climate Studies (CPTEC) global model is evaluated during three episodes of SACZ. The evaluation of the model is accomplished based on statistical indexes, such as mean errors (BIAS), root mean square errors and anomaly correlation of geopotential height, virtual temperature, moisture convergence, specific humidity, zonal and meridional winds and precipitable water. The evaluation of the precipitation forecast is achieved by using threat score for rain-no rain, weak rain, moderate rain and heavy rain categories. Two versions of the global model are used, one with the Kuo parameterization for the deep convection scheme, and the other with relaxed Arakawa-Schubert (RAS) scheme. The performance of the two versions is compared for the selected SACZ episodes over the regions: South America (AS), Tropics (TR), Extratropics (EX), Continent (CON) and Oceans (OCE). It is observed that the formation, the intensification and the dissipation of SACZ affect the forecasts quality, determining in some cases, the end of the useful forecast. The version of the model with Arakawa-Schubert parameterization shows better performance in forecasting the geopotential height (500 hPa), moisture convergence (850 hPa), zonal and meridional winds (850 hPa) in all the regions; at 250 hPa, the zonal wind on the regions AS, TR and CON, and the meridional wind on the region TR; the virtual temperature (1000 hPa) on the regions AS, TR, EX and CON, and the precipitable water on the regions AS, TR and CON. The version with Kuo parameterization shows better performance for the specific humidity in all the regions, for the zonal wind (250 hPa) and precipitable water on the regions EX and OCE, and meridional wind on the regions AS, EX, CON and OCE. The analysis of the threat score index indicates that both versions shows good performance in forecasting occurrence/not ocurrence of precipitation and deficiency in forecasting heavy rain. In general, forecasts based on the Arakawa-Schubert scheme produces better results.

ASSUNTO(S)

atmospheric general circulation models modelos de circulação geral da atmosfera convection clouds previsão numérica de tempo south parameterization statistical analysis parametrização nuvens convectivas análise estatística. key words:numerical weather forecasting

ACESSO AO ARTIGO

Documentos Relacionados