Investigação de tendência de índices de precipitação e temperatura do ar no Nordeste do Brasil.

Abstract

This study aimed to determine possible trends of climate extreme indices, dependent on precipitation and air temperature in range of monthly, seasonal and annual time, in northeastern Brazil, using the Mann-Kendall test. Ten indices of climate extremes, two daily precipitation and eight daily maximum and minimum temperatures out of twenty-eight grid cells covering the entire Northeast were used in the period from January 1, 1971 to December 31, 2001. In calculating the trends, we used the technique of linear least squares regression applied directly in the series of indices, called direct regression, and monthly, seasonal and annual values obtained in the Mann-Kendall test, called the Mann-Kendall. It was observed that the regression coefficients using the two paths were very close, except for the months of August and September, possibly because they were months of dry season with little rain in most part of the Northeast. It was found a 20% decrease, approximately, with the trends of the indices the maximum rainfall a day and five days in a row area. Positive trends were also observed, i.e., the increase in maximum and minimum temperatures and the number of hot days and nights. Another important observation was the decrease in the number of cold days and nights. We conclude that the reduction of maximum rains of one day and five consecutive days, the increase in maximum and minimum temperatures, the decrease in the number of cold days and nights and an increase in the number of days and warm nights were due to natural and anthropogenic local and global factors.