Extreme Rainfall Events over the Amazon Basin Produce Significant Quantities of Rain Relative to the Rainfall Climatology

Abstract

Although much effort has been made to characterize and understand extreme rainfall's causes and effects, little is known about their frequency and intensity. Moreover, knowledge about their contribution to the total rainfall climatology is still minimal, especially over the Amazon where rainfall data are very scarce. In this paper we propose to classify extreme rainfall events by type and analyze their frequency and intensity over South America with a focus on the Amazon basin. Gridded daily data from the MERGE/CPTEC product over a period of 15 years (1998-2013) was used. An adaptation of Rx5d climate index was applied to select different kinds of extreme rainfall for the purpose of quantifying their frequency and intensity as well as their contribution to the accumulated rainfall climatology. According to the results, all kinds of extreme rainfall events can be observed over the studied area. However, the quantity of rainfall produced by each type is different, and consequently their percent contributions to the total accumulated rainfall climatology also differ. For example: in the Amazon region EET-I is responsible for 15% -40% of the total accumulated rainfall. Moreover, in the Brazilian northeast there are regions where EET-I exceeds 40% of the total rainfall. In northeast Brazil EET-II is responsible up to 30% of the total accumulated rainfall. EET-III is responsible for 5% -15% in the Amazon basin, 25% -45% in northeast Brazil and 10% -45% over Roraima State. Area-mean time variation shows that the quantity of rainfall extremes over the

An Analysis of Rainfall Extremes in the Northern South America and Their Behaviors for Future Climate Based on A1b Scenario

“…Pará, Maranhão and Acre State present approximately half of their extensions with similar values. These results partially agree with the increased tendency obtained from Brito et al (2014), which observed the highest values for 1998-2013 in northern Pará and Southern Amazonas State, indicating that the observed patterns should continue in the future. Negative biases of less than 1mm/day were found for NEB in 2071-2100 time-slice (MARENGO AND BERNASCONI, 2015) using CMIP3 HadCM3 data, downscaled by ETA model .…”

“…The Roraima State, one of the areas highlighted by Brito et al (2014), presents a decrease in the accumulated rainfall for heavy and very heavy events, agreeing with the behavior evaluated by the authors. This pattern was _________________Revista Brasileira de Climatologia_________________ ISSN: 1980-055x (Impressa) Another parameter calculated in this work was the percentage contribution of each type of extreme event to the total amount of rainfall precipitated over the continent.…”

“…The Roraima State, one of the areas highlighted by Brito et al (2014), presents a decrease in the accumulated rainfall for heavy and very heavy events, agreeing with the behavior evaluated by the authors. This pattern was also observed for coastal areas (east to west, north of the continent).…”

“…discussing the CMIP5 projections for NEB, downscaled by ETA model simulations, for the 4 RCP (Representative Concentration Pathways, 2.6,4.5,6.0,8.5) found a broader area with positive bias, the higher the RCP, the broader the area in the north (Maranhão, north of Piauí, Ceará and Rio Grande do Norte) of the northeast region with biases ≥ 0.5mm/day. As can be observed, depending on the different projections used (CMIP3 and CMIP5) you may have different results for NEB, and the most recent ones points out to a not so dry future for the previous months of the peak of the rainy season, depending on the RCP considered.In Colombia, Venezuela and Peru a similar spatial pattern is noticed when the bias obtained for heavy, very heavy and rare events show changes from negative to positive, indicating a higher precipitation quantity associated with this type of event also for these countries.The Roraima State, one of the areas highlighted by Brito et al (2014), presents a decrease in the accumulated rainfall for heavy and very heavy events, agreeing with the behavior evaluated by the authors. This pattern was _________________Revista Brasileira de Climatologia_________________ ISSN: 1980-055x (Impressa) Another parameter calculated in this work was the percentage contribution of each type of extreme event to the total amount of rainfall precipitated over the continent.…”

“…However, to quantify and select extreme rainfall events by intensity, the adaptation proposed by Brito et al (2014) was applied. Considering our data set, composed of 2 different decades, 1980-1990 and 2089-2099, the December-January-February trimester of each year (10 trimesters per decade) was analyzed.…”