An objective and efficient method for estimating probabilistic coastal inundation hazards

Abstract

The joint probability method (JPM) is the traditional way to determine the base flood elevation due to storm surge, and it usually requires simulation of storm surge response from tens of thousands of synthetic storms. The simulated storm surge is combined with probabilistic storm rates to create flood maps with various return periods. However, the map production requires enormous computational cost if state-of-the-art hydrodynamic models with high-resolution numerical grids are used; hence, optimal sampling (JPM-OS) with a small number of (~ 100-200) optimal (representative) storms is preferred. This paper presents a significantly improved JPM-OS, where a small number of optimal storms are objectively selected, and simulated storm surge responses of tens of thousands of storms are accurately interpolated from those for the optimal storms using a highly efficient kriging surrogate model. This study focuses on Southwest Florida and considers ~ 150 optimal storms that are selected based on simulations using either the low fidelity (with low resolution and simple physics) SLOSH model or the high fidelity (with high resolution and comprehensive physics) CH3D model. Surge responses to the optimal storms are simulated using both SLOSH and CH3D, and the flood elevations are calculated using JPM-OS with highly efficient kriging interpolations. For verification, the probabilistic inundation maps are compared to those obtained by the traditional JPM and variations of JPM-OS that employ different interpolation schemes, and computed probabilistic water levels are compared to those calculated by historical storm methods. The inundation maps obtained with the JPM-OS differ less than 10% from those obtained with JPM for 20,625 storms, with only 4% of the computational time.

A sensitivity study of rising compound coastal inundation over large flood plains in a changing climate

“…The coupled surge-wave model CH3D-SWAN, driven by a synthetic parametric wind model 50 , was used with the Joint-Probability-Method with Optimal Sampling (JPM-OS) 21 , 51 – 53 . First, the wind and atmospheric pressure fields were calculated based on the location of the storm, pressure at the center, radius to maximum winds, and translational velocity of the storm.…”

“…First, the wind and atmospheric pressure fields were calculated based on the location of the storm, pressure at the center, radius to maximum winds, and translational velocity of the storm. Next, an ensemble of straight-line storm tracks, with all parameters kept constant until landfall and then allowed to decrease, was generated 53 . Finally, storms were simulated by a 2D vertically averaged version of CH3D-SWAN, which accurately (with 10% of relative Root Mean Square Error) simulated the observed High Water Marks (HWMs) in the region during Hurricane Irma 19 .…”

Risk assessment for typhoon-induced storm surges in Wenchang, Hainan Island of China

“…To reduce the computation burden, the JPM-OS is developed, which selects storms for simulation through optimal parameter selection with associated weighting and interpolation methods (Irish et al 2009;Resio et al 2009). Yang et al (2019) compares the results of JPM and JPM-OS, and finds that results of JPM-OS and JPM are very similar. In addition, they compared the performance of various JPM-OS methods.…”

Storm Surge Risk Assessment for Non-Life Insurance

“…Although it is possible to reduce the calculation load by one digit compared to JPM [17], the problem is that uncertainty arises during interpolation. In order to reduce the uncertainty, research is underway on JPM-OS interpolation methods (eg Yang et al [21]). Yang et al compared the calculation results of the inundation depth for different JPM-OS interpolation method for each return period in Florida, USA.…”