Modelagem hidrológica para extremos de inundações e secas para o município de Boa Vista em Roraima.

Abstract

The present research is based on statistical methods applied as an analysis tool in the study of the interaction between ocean and atmosphere in the up and down behavior of the Branco river in Boa Vista. Individual associations for years of floods and droughts conditioned to the oceanic component, evaluated by Sea Surface Temperature (SST) in the monitored areas of Niño 1 + 2, Niño 3.4, Niño 3, Niño 4 and North Atlantic Tropical (ATN); (PD) and Tahiti (PT) regions were evaluated through the acquisition of monthly climatic data from the Climate Prediction Center in 1982-2016. These associations aimed to investigate if these areas present favorable indicatives for extreme years of floods and droughts. Significant correlations were found above 0.5 in most flood and drought events in the following areas: Niño 1 + 2, Tropical Atlantic North, both with a lag time of 4 months, and the Darwin and Tahiti regions, but the effects of these variables to change the fluviometric regime of the White River in Boa Vista is 6 months. This information obtained through the calculation of the correlation coefficient (r) allowed the use of the Least Squares Method to model the prediction of the variability of flood and drought events induced by the seasonality of the Branco river. The long-term trends and numerical oscillations reproduced by the model for both scenarios were compared with the level measurements for the period 2011-2016. The results showed good performance of the model, with a percentage error of 30% for the prediction of drought events and 34% for those of floods, thus indicating that the selected input components exert a great contribution in the predictability of hydrological extremes in Boa Vista. Given this, it is suggested that this study can become operational in the monitoring centers of the state of Roraima, as a tool to support planning actions in the period of floods and droughts.