Significant Influences of Global Mean Temperature and ENSO on Extreme Rainfall in Southeast Asia

Abstract

Abstract This study investigates the changes in annual and seasonal maximum daily rainfall (RX1day) in Southeast Asia, obtained from gauge-based gridded precipitation data, to address the increasing concerns about climate change in the region. First, the nonparametric Mann–Kendall test was employed to detect significant trends in RX1day. Then, maximum likelihood modeling, which allows the incorporation of covariates in the location parameter of the generalized extreme value (GEV) distribution, was conducted to determine whether the rising global mean temperature, as well as El Niño–Southern Oscillation (ENSO), is influencing extreme rainfall over the region. The findings revealed that annual and seasonal RX1day is significantly increasing in Indochina and east-central Philippines while decreasing in most parts of the Maritime Continent during the past 57 yr (1951–2007). The trends in RX1day were further linked to the rising global mean temperature. It was shown that the location parameter of the GEV—and hence the RX1day on average—has significantly covaried with the annually averaged near-surface global mean temperature anomaly. Such covariation is pronouncedly observed over the regions where significant trends in RX1day were detected. Furthermore, the results demonstrated that, as ENSO develops in July–September, negative covariations between the location parameter of the GEV and the ENSO index, implying a higher (lower) likelihood of extreme rainfall during La Niña (El Niño), were observed over the Maritime Continent. Such conditions progress northward to the regions of Indochina and the Philippines as ENSO approaches its maturity in October–December and then retreat southward as the ENSO weakens in the ensuing seasons.