A Hierarchical Bayesian Model based Nonstationary Frequency Analysis of Extreme Sea Level under Climate Change along the Shorelines in South Korea

Abstract

Urban development and population increases have continuously progressed in the coastal areas in Korea; thus, it is expected that vulnerability towards coastal disasters by sea level rise (SLR) would accelerate. This study investigated the trend of the sea level data using Mann-Kendall (MK) test, and the results showed that the increasing trends of annual average sea level at 17 locations were statistically significant. For annual maximum extremes, seven locations exhibited statistically significant trends. In this study, nonstationary frequency analysis for the annual extreme data, together with average sea level data as a covariate, was performed. Nonstationary frequency analysis results showed that sea level at the coastal areas of the Korean Peninsula would be increased from a minimum of 60.33mm to a maximum of 214.90nun. The future mean sea level rise (MSLR) increases in the year 2100 estimated in this study suggested that rises in the sea level at the sea near the Korean Peninsula would be rather less compared with global MSLR by RCP(AR5) scenarios. Nevertheless, the SLR is still very high and we need to evaluate the risk for the marine structures and coastal areas due to the acceleration of the sea level and to prepare adaptation strategies at the national level. Furthermore, it is necessary to carry out a nonstationary frequency analysis considering the representative concentration pathways (RCP) scenario and more reliable information for acceleration characteristics as a future study.