Multifractal analysis of standardized precipitation index in Northeast Brazil

Abstract

Northeast Brazil (NEB) is one of the world's most densely populous dryland regions most vulnerable to the impacts of climate change. It is expected that the risk of natural disasters related to excess or lack of water (flash floods and severe droughts) will continue increasing in the region until the end of the century. In this work, we study multifractal properties of the Standardized precipitation index (SPI) which is commonly used to evaluate the degree of severity of drought and humid conditions at different time scales and can be compared among regions with distinct climatic conditions. The method Multifractal detrended fluctuation analysis (MFDFA) is used to analyze SPI for 1, 3, 6, and 12-month accumulation periods at 133 locations across the state of Pernambuco, with climate varying from tropical humid in the coastal area to semiarid in the deep inland region. The results show that all SPI series (independently of time scale and climatic conditions) display multifractal dynamics, which can be described by properties of the multifractal spectrum. It was found that SPI series exhibit long-term persistence and multifractality that becomes stronger as accumulation time increases. For all temporal scales, the values of Hurst exponent increase in the west-east direction indicating stronger persistence of dry/wet conditions in the coastal region Zona da Mata, with a tropical humid climate, than in the dry semiarid inland Sertão region. For SPI-1, SPI-3, and SPI-6 which describe short and medium conditions the width of the multifractal spectrum (degree of multifractality) doesn't vary across the studied area, while for SPI-12 which describes long-term conditions the width of the multifractal spectrum decreases in a west-east direction, indicating stronger multifractality and higher complexity in long duration precipitation anomalies in the semiarid Sertão region.