Trend analysis of droughts and their geospatial and temporal variability assists decision-making about water resources management around the world and decreases the harmful effects of drought that affect the entire population. This work aimed to analyze short-, medium-and long-term droughts and their trends in the Brazilian state of Paraíba from 1998 to 2015 using Tropical Rainfall Measuring Mission (TRMM) data and applying the Mann-Kendall test and Sen's slope estimator method, based on the standardized precipitation index (SPI). TRMM data were validated by comparison with data from 267 rain gauges in the region, which showed the consistency of the satellite data. Therefore, 187 monthly TRMM rainfall time series were used, each with 216 months. The series were equally distributed over the entire study area. At the significance level of 0.01, a new geospatial classification of drought severity is proposed, through which it is possible to determine exactly which types of drought events affected or did not affect a given region based on the SPI and the trend of the analyzed SPI time series, which shows the situation of drought risk analysis. The results of the comparison between long-and short-term droughts indicate that the wettest regions of the state of Paraíba are strongly affected by extreme drought events and show trends with increasingly negative slopes. In this way, the proposed geospatial classification is proved to be a useful tool because it provides information about the current drought situation of a given region, simultaneously showing the trend slope with respect to short-, medium-and long-term droughts.

The present study investigates the influence of different atmospheric teleconnections on the annual precipitation variability in Northeast Brazil (NEB) based on the annual precipitation data from the Global Precipitation Climatology Center (GPCC) from 1901 to 2013. The objective of this study is to analyze the influence of different atmospheric teleconnections on the total annual precipitation of NEB for the 1901–2013 period, considering the physical characteristics of four subregions, i.e., Mid-north, Backwoods, Agreste, and Forest zone. To analyze the influence of different atmospheric teleconnections, GPCC data were used, and the behavior of the teleconnections was assessed using Pearson correlation coefficient, Rainfall Anomaly Index (RAI), and cross-wavelet analysis. The Pearson correlation was used to analyze the influence on the annual precipitation for the studied region. RAI was used to calculate the frequency of atmospheric patterns and drought episodes. The cross-wavelet analysis was applied to identify similarity signals between precipitation series and atmospheric teleconnections. The results of the Pearson correlation assessed according to Student's t test and cross-wavelet analysis showed that the Atlantic Multidecadal Oscillation (AMO) exerts a more significant influence on the Backwoods region at an interannual scale. In contrast, the Pacific Decadal Oscillation (PDO) exerts greater control over the modulation of the climatic patterns in NEB. The results of the study are insightful and reveal the differential impacts of teleconnections such as the AMO, PDO, MEI, and NAO on precipitation in the four sub-regions of NEB. The Atlantic circulation patterns strongly influence the interannual and interdecadal precipitation in the Agreste, Backwoods, and Mid-north regions, possibly associated with the Intertropical Convergence Zone (ITCZ) position. Finally, this study contributes to understanding internal climatic variability in NEB and planning of water resources and agricultural activities in such a region. Graphic abstract

Universidade Federal do CearáRESUMO: O objetivo deste trabalho foi estudar a dinâmica da variabilidade climática espacial e temporal da pluviosidade na Amazônia Ocidental, utilizando as técnicas multivariadas de Análise de Componentes Principais (ACP) e Análise de Agrupamento (AA). Foram utilizadas médias anuais da precipitação pluvial de quatro índices climáticos pluviais obtidas a partir de dados diários de 36 pontos de grade, extraídos da reanálise ERA-40 do ECMWF (European Centre for Medium-Range WeatherForecasts), para o período de 1970 a 2001. Na análise de componentes principais aplicada para os índices de precipitação pluvial PRCPTOT e SDII observaram-se que as cinco primeiras componentes apresentaram autovalores superiores ou igual a um e explicaram mais de 83% da variância total, para os demais índices, DCM foram retidas as nove primeiras componentes que explicaram 81,69% e para o DCS foram necessárias as quatro primeiras que explicaram 87,98%. Com relação à determinação de áreas homogêneas verificou-se cinco regiões para PRCPTOT e SDII, quatro para DCS e seis para DCM. Conforme as análises da regionalização homogêneas usando CPs, conclui-se que o grande número de regiões obtidas reflete os sistemas de tempo produtores de precipitação na Amazônia Ocidental e a interação entre eles, bem como a interação deles com os efeitos locais. Palavras-chave: Regiões homogêneas, Índices climáticos, Precipitação. MULTIVARIATE ANALYSIS OF CLIMATIC INDICES IN THE WESTERN AMAZONABSTRACT: The aim of this work was to study the dynamics of spatial and temporal climate variability of rainfall in western Amazonia, using multivariate techniques of Principal Component Analysis (PCA) and Cluster Analysis (AA). Average annual rainfall of four stormwater climate indices derived from daily data from 36 grid points , extracted from the ERA-40 reanalysis of the ECMWF (European Centre for Medium -Range Weather Forecasts) for the period 1970-2001 were used. In principal component analysis applied to the indices of rainfall and SDII PRCPTOT observed that the first five components had eigenvalues greater than or equal to one and explained over 83 % of the total variance for the other indices, the nine DCM were retained first parts that explained 81.69 % and DCS were necessary for the first four which explained 87.98 % . With respect to the determination of homogeneous areas there was five regions for PRCPTOT and SDII, four for DCS and six for DCM. As the analysis of homogeneous regionalization using CPs, it is concluded that the large number of regions obtained reflects the systems producing time of precipitation in western Amazonia and the interaction between them as well as their interaction with local effects.

No abstract

São aplicadas técnicas multivariadas às séries de precipitação pluvial para a Amazônia Ocidental. Utilizaram-se séries de dados médios mensais de precipitação oriundas do Instituto Nacional de Meteorologia (INMET), para o período de 1970-2001. Usaram-se os métodos de Componentes Principais e Análise de Agrupamentos para determinar os padrões espacial e temporal e regiões homogêneas com relação à precipitação pluvial. Os resultados obtidos mostraram que três regiões climaticamente homogêneas podem caracterizar a distribuição da precipitação na Amazônia Ocidental, quais sejam: a identificada como região 1 que representa a parte noroeste com altos valores de precipitação associados à condensação do ar úmido trazido pelos ventos de leste da ZCIT, que são elevados quando o escoamento sobe os Andes provocando chuvas orográficas.; a identificada como região 2, observada no extremo sul da Amazônia Ocidenttal sofre influência da ZCAS e ZCIT; a região 3, que se encontra no estado de Roraima, os totais pluviométricos mensais são influenciados pela topografia e tem como principal sistema indutor e distribuição das chuvas a ZCIT. Palavras-chave: Precipitação, Regiões Homogêneas, Agrupamento. A B S T R A C T Multivariate techniques are applied to the series of rainfall for the Amazônia Ocidental. We used data series average monthly rainfall derived from the Instituto Nacional de Meteorologia (INMET) for the period 1970-2001. They used the methods of Principal Component and Cluster Analysis to determine the spatial and temporal patterns and homogeneous regions with respect to rainfall. The results showed that three climatically homogeneous regions can characterize the distribution of precipitation in the western Amazon, which are: a region identified as one that represents the northwestern part with high amounts of precipitation associated with the condensation of humid air brought by easterly winds ZCIT, which are high when the flow rises causing the Andes orographic rainfall.; identified as the region 2, observed in the extreme south of the state of Amazônia Ocidental influenced the ZCAS and ZCIT, region 3, which in the state of Roraima, monthly rainfall totals are influenced by topography and its main inducer system and the ZCIT rainfall distribution. Keywords: Precipitation, Regions Homogeneous Grouping.