Indonesia Conference Directory

A multivariate analysis on rainfall variables is needed to understand relationships among characteristics of rainfall variables. The monthly rainfall amount records from three stations during the period of 1985-2014 in Makassar are used. The Standardized Precipitation Index is employed to classify data into wet category. The aim of this study is to investigate the most appropriate joint probability distribution function of wet duration and wet severity based on bivariate Archimedean copulas. The copulas are Gumbel-Hourgaard, Frank, Joe, and Clayton. Parameter of copulas are estimated by Kendall-s tau correlation coefficient. The study result show that the Frank copula was identified as the best copula in joint modelling between wet duration and wet severity at MMP and BRP stations, meanwhile the Gumbel-Hougaard copula at BBMKG station. This results could be useful information for determining return periods of wet condition characteristics.