Efeitos do aumento de gases do efeito estufa na frequência e intensidade dos eventos extremos de precipitação na região norte da América do Sul: análise de dados modelados

Abstract

The present study analyses the changes in the frequency and intensity of extreme precipitation events in northern South America for two future climate simulations: 2040-2050 and 2090-2100. Both numeric experiments were performed with the regional ETA model, using data from the CCSM3 global model as initial and boundary conditions for the high emissions scenario SRES A2. The future climate was compared to the actual one, represented here by the decade 1980-1990 for the three months period JFM. To identify the extreme precipitation events we used a methodology known by R5d, indicated by the World Meteorological Organization to study extreme climate events. R5d selects the maximum rainfall value for a period of 5 days. We calculated the mean value from this subset and analyzed again the original series, considering values higher or equal to this mean as extremes. Once identified, the extreme events were classified. Extreme Event Type (EET) – I consist of the events that are equal or higher than the extreme average (cmax) and smaller than the extreme average (cmax) plus once standard deviation (dpmax). EET-II are the extreme events equal or higher than cmax plus once dpmax and smaller than cmax plus twice dpmax. The last type, EET-III, are all events higher than cmax plus twice dpmax. In summary, the spatial analysis shows that the Central Amazon Region (RCA) presents intense positive anomalies for the average of EET-I, EET-II and EET-III, while the Northeastern Region (RNE) and the Northern Region (RNO) present negative anomalies for most of their areas for the three types of extremes. Frequency anomalies do not present clear spatial patterns. Quantitative analysis confirms what was observed spatially for the RCA; Positive anomalies for the mean values of the three types of EET and for the frequency in EET-I and EET-II for both time period and a decrease in frequency to EET-III for both time periods. Likewise, RNE and RNO present negative anomalies for the mean values and in frequency of the EET-I and EET-II for both decades. For EET-III in RNE there were positive anomalies in frequency and in the events average for 2040-2050 decade and negative anomalies in both variables for the 2090-2100 decade. In RNO the average anomalies were positive for both analyzed periods and the frequency had an increase in the half of the XXI century and reduction for the final of the XXI century. The intensification tendencies for extreme precipitation events in RCA and their reduction in RNE and RNO show the same results as previous studies using different models and emission scenarios.