Clima presente e cenário futuro dos eventos com potencial para causar alagamentos nas cidades de Porto Alegre, Pelotas, Caxias do Sul e Santa Maria

Abstract

This work has as general purpose the analysis of present climate and future behavior of short duration extreme local rainfall events with potential for causing flooding in major cities in Rio Grande do Sul State covering different regions: Porto Alegre, Caxias do Sul, Santa Maria and Pelotas. Based on history of floods of each city and on the rainfall data obtained from meteorological stations of INMET (National Institute of Meteorology), the thresholds that have identified which precipitation rate is an event with the potential to cause flooding in the city were obtained, being such events divided in case of attention (25th percentile) or case of alert (75th percentile) in 48 hours before the occurrence of flooding. The results showed a subtle increase of cases, both as alert and attention in the last 53 years. The city of Porto Alegre was presented the least thresholds for flooding condition, which was expected due to its increased urbanization. In assessing the seasonal behavior of the cases it was found that in this climate, the July-August-September was the time of year most facilitates cases of flooding. Later, cases were correlated with three climate indices that measure the SST (Sea Surface Temperature) anomalies in order to assess whether their occurrences are associate to large-scale climatic phenomena. It was found weak to moderate correlations, but with a high level of statistical confidence for the EMI (ENSO Multivariate Index). The use of regional climate model (ETA-HadCM3) denotes that in a future scenario (A1B following International Panel on Climate Change) for the period 2021-2070 will be in general an increase number of cases of attention for all cities, when compared to those observed in the present climate (1961-2013). In Pelotas there will be a climatological change due to a displacement of the season with more cases: from JAS quarter (present) to OND quarter (future scenario).