Dependence properties of spatial rainfall extremes and areal reduction factors: An application for rain stations in the North of Vietnam

Abstract

    Dpendence properties of spatial rainfall extremes provides information regarding the simultaneous or non-simultaneous occurrence of extreme rainfall at different locations in a catchment. The lack of this information will lead to inaccuracies in flood studies, such as the estimation of the area reduction factors (ARFs) for a given catchment. Currently, there is also an disagreement about whether extreme rainfall has the properties of asymptotic dependence or asymptotic independence. This paper has shown that the extreme rainfall in the northern region of Vietnam has the properties of asymptotic independence, so the asymptotic independence model has been selected to conduct extreme rainfall
simulation for the study area. The results of calculating the area reduction factors (ARFs) from simulated and observed rainfall data are very suitable for long return periods (20, 50, and 100 years), and ARFs calculated from simulated rainfall are also useful for extrapolating to rarer frequencies. The paper also recommends that future studies should use asymptotic independence models to simulate extreme rainfall for the study area.