Modeling Extreme Precipitation in Thailand

Abstract

Attempts to use the Generalize Extreme Value distribution and Generalize Pareto distribution with the maximum likelihood estimates on the extreme rainfall data at 99 weather stations over Thailand. The rainfall data gathered from January 1984 to December 2014. The research area occupied on 5 regions as the North, Northeast, Central, Eastern and South and estimate return level in the period of 5, 10, 25, 50 and 100 years. The discussion indicates the return level by pluvial maps. The result display the modeling of GEV, Gumbel distribution was a fitting propose for annual extreme rainfall either 5 regions in Thailand. However, some of station in the North and Central region founded Fréchet Distribution and Weibull Distribution. Morover, in the Northest region lay down on the Weibull Distribution while in the South Fréchet Distribution was approximately model. In addition, the modeling of GPD with daily rainfall data in 5 regions had fitting on Exponential distribution. However, some stations in area study had suitable with Pareto Distribution and Gamma Distribution. Certainly, the return level indicated Trat weather station in the South had return level highest than other station.