Predictability of Coastal Extreme Wave Heights Based on a Nonstationary Hierarchical Bayesian Model: The Role of the Sea Surface Temperature

Abstract

Kim, Y.-T.; Uranchimeg, S.; Park, M.H., and Kwon, H.-H., 2021. Predictability of coastal extreme wave heights based on a nonstationary hierarchical Bayesian model: The role of the sea surface temperature. In: Lee, J.L.; Suh, K.-S.; Lee, B.; Shin, S., and Lee, J. (eds.), Crisis and Integrated Management for Coastal and Marine Safety. Journal of Coastal Research, Special Issue No. 114, pp. 191–195. Coconut Creek (Florida), ISSN 0749-0208. The increasing frequency of extreme wave heights has been observed in many parts of the world due to recent global warming. Especially, the variability of extreme wave heights in the summer season has increased during the last three decades over the coast of South Korea that is partially attributable to the increase of the climate variability, leading to the changes in frequency and intensity of the extreme wave heights. Further, coastal areas are particularly vulnerable to the increased climate variability, and the associated risk can be exacerbated by the rapid coastal development over South Korea. Most of the extreme wave heights in South Korea are largely caused by a typhoon during the summer season. Recent studies revealed that the typhoon frequency and intensity in the Asian monsoon region are modulated by Sea Surface Temperature (SST) during the summer season from May to November. Traditional frequency analysis methods do not consider the year to year shifts in wave heights risk distributions that are attributed to changes in climate variability that affect the causal structure of coastal inundation risk. In this perspective, a climate-informed nonstationary frequency analysis model is proposed to predict the extreme wave heights and explore the role of the SST in a Bayesian regression framework. The parameters of the nonstationary frequency model are obtained using a Markov Chain Monte Carlo algorithm. The proposed model shows strong potential for predicting the extreme wave heights by linking large-scale climate patterns, providing motivation for developing dynamic coastal flood risk management strategies.