Tendências sazonais da evapotranspiração de referência e do índice de aridez para o Nordeste do Brasil

Abstract

Northeastern Brazil (NEB) is the region most vulnerable to climate change in Brazil, with an emphasis on the irregularity and distribution of water resources in most of the region. With the advent of the global average global temperature increase, NEB is expected to make the air drier and consequently the demand for water in the atmosphere increases, making water for human, animal and agricultural use, more scarce. Therefore, the general objective of this study was to analyze the variability (spatial and temporal) of two important drought indicators: Reference Evapotranspiration (ET0) and Aridity Index (IA) for NEB, checking for signs of changes climate change. In order to do so, we used ET0 monthly meteorological databases and the variables necessary to calculate AI, available from Xavier et al. (2015) in the period from 1980 to 2013 (34 years) and with spacing of 0.25 ° x 0.25 °, making a total of 2042 data points. The IA was calculated according to the criteria recommended by Thornthwaite (1948). The Cluster analysis method was adopted to classify subregions in the NEB with homogeneous characteristics for ET0, then characterized the seasonal variability of ET0 and IA. The respective periodicities of these variables, by sub-region, were verified through the Wavelet analysis. Seasonal analyzes in temporal series were applied using the non-parametric tests of Mann-Kendall (trend), Sen (magnitude) and Pettitt (beginning). Statistical significance was 5% or 1% for all tests. The main results for NEB indicated that there are 5 sub-regions with homogeneous characteristics for ET0, with Sub-region 3 (S3) having the highest average accumulations (2098.0 mm / year) and S5 the lowest (1362, 8 mm / year). For IA, precipitation was the dominant variable and in the less rainy period (Spring) the AI approaches the maximum value (IA = 1.00) in practically the entire NEB. Regarding the temporal analyzes, different trend signals for ET0 were identified within the same season, being predominantly positive with statistical significance at 1% in the southwestern state of Bahia (summer, autumn, winter and spring) and negative at 5 (NE), especially in the eastern part of the state of Rio Grande do Norte (RN), northeast of Bahia (Primavera), eastern Paraíba and RN (summer), and practically along the entire NEB east coast (Winter), except part of the south of Bahia. The respective magnitudes of ET0 positive trends ranged from 1.6 mm / season (Autumn) to 2.4 mm / season (Spring), while negative trends ranged from -1.5 mm / season (Autumn ) to -2.2 mm / season (Spring), obtaining a majoritarian start in the 1990s. For AI, the significantly positive trends were concentrated in southwestern Bahia (summer and winter) and negative ones (predominantly in winter ) occurred in eastern RN east of Pernambuco, state of Alagoas and northeast of Bahia. On the other hand, the negative magnitudes (absent in summer and autumn) ranged from -0.007 (Spring) to -0.016 (winter) ). The 1990s saw a greater frequency of onset of trends for AI in the autumn and winter, while in the spring the respective beginnings of AI trends were concentrated between 2000 and 2004. From the results found, it was concluded that the variations spatial and temporal variations of ET0 and IA in NEB were heterogeneous and showed opposite differences of trend signals, even in the same season of the year, however, it was observed that predominantly to the west of NEB, positive tendencies of ET0 predominate, and to the east the opposite occurs.