Drought severity based on the SPI index and its relation to the ENSO and PDO climatic variability modes in the regions North and Northwest of the State of Rio de Janeiro - Brazil

Abstract

Drought severity in regions North and Northwest of Rio de Janeiro state (SRJ) was evaluated using the Standardized Precipitation Index (SPI) for the monthly (SPI-1) and annual (SPI-12) scales, as well as its relationship with El Niño - Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO). Data of 14 rainfall stations from a time series considering the period of 1967–2013 was used. The SPI calculation was based on the SCI package available at R software version 3.4.2 library. SPI categories used are Very Dry, Moderately Dry and Extremely Dry associated with the hot and cold phases of the PDO, followed by ENSO events according to the Oceanic Niño index (ONI) in El Niño region 3.4 using an interaction matrix. The matrix was evaluated by the coefficients of Kendall (τ), Sperman (ρ), Pearson (r) and R2 determination, followed by Willmott's concordance index (d), Standard Error of Estimation (SEE) and Student's t-test. The intense activity of La Niña in the cold 1st phase of the PDO increased rainfall with greater intensity in the Northwest region compared to the North. In the 2nd cold phase of the PDO, the events of El Niño, La Niña and Neutral years of the ENSO mutually contribute to the changes in rainfall regime of both regions. The most frequent SPI categories are Moderately Dry and Extremely Dry. SPI-1 was efficient in perceiving extreme droughts, while SPI-12 results were not significant in the revealing of extreme droughts. In addition, there was no relational pattern through statistical indexes between droughts detected by SPI, ENSO and PDO. Nevertheless, when ENSO cycles were persistent within both phases of the PDO, droughts occurred, mainly on the monthly scale.