Paper Title
Statistical Modeling of Extreme Rainfall Characteristics over Guam

Abstract
The behaviours of extreme variables, in hydrological applications, are often essential for planning, risk assessment and the design of hydraulic structures such as dams, reservoir, and storm sewers. Given a site with sufficient rainfall data, frequency analysis of annual maximum rainfall can be performed. Results from this analysis are often summarized by an “Intensity-Duration-Frequency” relationship for a given site. Several probability models have been developed to describe the distribution of extreme rainfalls at a single site. However, these models are only accurate for the specific time frame associated with the data used. Recently, climate change is considered as having a profound impact on precipitation process. Global Climate Models (GCMs) can represent reasonably well main features of the global distribution of basic climate parameters, but outputs from these models are usually at a resolution that is too coarse for many impact studies. Hence, Yeo et al. (2020, 2021) propose spatial-temporal downscaling approaches to estimate sub-hourly/daily annual extreme rainfall series using historical records from rain gauge network available in South Korea and Canada. Unfortunately, this type of frequency analyses, coupled with climate change impacts, are not introduced to the pacific islands. In light of the above-mentioned issue, the present study proposes a temporal downscaling approach to describe the linkage between observed daily annual maximum precipitations (AMPs) and sub-daily/hourly AMPs using scale invariant properties. The temporal downscaling model’s outputs are numerically and graphically evaluated with observed and estimated quantile methods using L-moment and Non Central Moment. Promising results are observed at 7 rain gage stations over Guam. Further studies are planned to apply the developed temporal downscaling model to climate change scenarios and to stormwater management model (SWMM).