Previsão sazonal da precipitação no nordeste do Brasil com base na temperatura da superfície do mar.

Abstract

In this study, the correlation of the pluviometric precipitation anomaly of seven homogeneous groups of pluviometric stations in the Northeats of Brazil (NEB) with anomalies of sea surface temperature (SST) in the Pacific and Atlantic oceans was analyzed. The general objective was to develop multiple regression statistical formulas with the purpose for elaborating quantitative forecast of precipitation on the NEB during the rainy season of each group from SST data with lag up to -6 months. The period studied was from 1994 a 2016 and the analysis shows that there is influence of the Pacific and Atlantic oceans on NEB rains, with a higher correlation in the North Pacific and North Atlantic (with a correlation greater than 0.6, and p < 0.05). These correlations were evidenced by Kayano et al. (2018) in which they show the influence of two low frequency dominant modes, the Pacific Decadal Oscillation (PDO) and the Atlantic Multidecadal Oscillation (AMO), in the climate-related teleconnections related to NEB precipitation. In the analysis of the homogeneous groups, seven groups were developed: Group 1, corresponding to the rainy season from October to January; Group 2 has the rainy season from November to February; Group 3 represents the rainy season from Decembre to March; Group 4 has the rainy season from January to April; Group 5 a rainy block is from February to May; Group 6 corresponds to rainy season from March to June; and Group 7 has the rainy season from April to July. The predictive models that were developed for the seven groups had variations in their responses, that is, the correlations and levels of explanation varied according to each group. The best representation was seen in Group 2, with a correlation coefficient of 83.91%, followed by Group 5, with 67.56%. Group 3 presented the worst correlation coefficient with 49.21%, followed by groups 1, 4, 6 and 7, with correlation coefficient of 51.34%, 55.00%, 55.14% and 55.19%, respectively.