LAUSR

Abstract The main factors responsible for the greatest natural disasters are related to extreme weather events for which no country is really well-prepared, particularly in Latin America, and specifically in Brazil. This research was developed to determine the spatiotemporal climatic similarities of municipalities along a longitudinal gradient from the east to north-east of Pernambuco State, in northeastern Brazil, from 1993 to 2018, based on applied statistics. The daily meteorological database (rainfall, air temperature, humidity, pressure, wind and evaporation) registered at eight manual meteorological stations of the National Institute of Meteorology, with a historical period of 26 years. The meteorological data were subjected to descriptive statistics, the cluster analysis (CA), and the principal components analysis (PCA), to identify which components are most affected in the climatic patterns. This study also carried out investigations into the influence of the possible dynamic causes of the El Nino-Southern Oscillation (ENSO) phases on the rainfall patterns of the municipalities. The eigenvalues of the three PCAs >1. PC 1 and PC 2 showed a total variance of 59.50 %. The variables rainfall and relative humidity presented a higher correlation with the municipalities of Recife and Triunfo (group G1). The municipalities of Petrolândia, Cabrobo and Ouricuri were grouped according to the variables evaporation, air temperature and atmospheric pressure (group G2). The CA was able to indicate the cluster of municipalities with similar climatic characteristics. The ENSO phases directly influence the variability of rainfall in both homogeneous groups identified in the study. Based on multivariate statistical analysis, it is possible to determine the meteorological seasonality between the different regions of the State of Pernambuco - northeastern Brazil. Such an analysis is important for the population in the eight municipalities, since it makes it possible to improve the productivity of economic activities, especially crop cultivation and livestock rearing.