Compound drought and hot events assessment in Australia using copula functions

• Published in: Environmental Research Communications Vol. 6; no. 3; pp. 31002 - 31013
• Main Authors: Páscoa, Patrícia, Gouveia, Célia M, Ribeiro, Andreia F S, Russo, Ana
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.03.2024
• Subjects: Conditional probability; Drought; Evapotranspiration; heatwaves; NHD; NHN; number of hot days; number of hot nights; SPEI; standardized precipitation evapotranspiration index; Summer; Wildfires; Australia
• ISSN: 2515-7620; 2515-7620
• DOI: 10.1088/2515-7620/ad2bb8

Abstract The occurrence of compound drought and hot events has been shown to cause stronger socio-economic, environmental and health impacts than the isolated events. Moreover, the frequency of these compound events has increased unevenly throughout the world and is expected to keep increasing in several regions. In this work, an assessment of compound drought and hot events in the summer months in Australia was made, using copula functions. Drought and hot conditions were identified by the Standardized Precipitation Evapotranspiration Index (SPEI) and the indices Number of Hot Days (NHD) and Number of Hot Nights (NHN) for the summer months, respectively. We analysed drought conditions in the current and the previous 1 to 3 months and the periods 1950–1978 and 1979–2020. The results show that the conditional probability of the occurrence of hot events given drought conditions is very high for the concurrent month in most of the study area, reaching 0.9 in some cases. Considering previous drought conditions, the higher probabilities are obtained in the southeastern region in December and in the north in February but, in most of the study area, these values are higher than for the case of non-drought conditions, pointing to an effect of previous drought conditions on hot events of up to 3 months. Moreover, an increased frequency of compound drought and hot events from the first to the second period was identified in more than half of the study area for lags of 1 and 2 months. We show that, although the conditional probabilities are mostly higher when computed with NHD, NHN is also affected by drought conditions, and should also be considered in this analysis, since nights can have a relieving contribution when impacts in health and wildfires are being analysed.