Improvement to Flooding Risk Assessment of Storm Surges by Residual Interpolation in the Coastal Areas of Guangdong Province, China

Xina Wang,Xianwei Wang,Jianqing Zhai,Xiucang Li,Huabing Huang,Chunmei Li,Jing Zheng,Haiyan Sun

DOI: https://doi.org/10.1016/j.quaint.2016.12.025

IF: 2.454

2017-01-01

Quaternary International

Abstract:Estimation of the extreme Total Water Levels (TWLt) is critical for inundation modeling and risk analysis. TWLt consists of the extreme Storm Surge (SSt) and Astronomic Tide (AT(t)), which are strongly affected by local topography and have large spatial heterogeneity. One common method is to compute the TWLt using the long-term TWL recorded at the limited standard tide gauges and then interpolate them into areas without gauges (here called total interpolation), often generating large uncertainties. This study develops a residual interpolation method to estimate the extreme TWLt for those areas without standard tide gauges. The core of this method is first to compute the extreme AT(t) for the 2-year return period using the tidal datum at the dense secondary stations, and then the residuals of AT(t) between return periods of T (T = 10, 20, 50, 100, 200 and 500 years) and 2 years at the standard stations are interpolated into the secondary tide stations; meanwhile, the extreme SSt computed at the standard stations are interpolated into the secondary tide stations as well. The sum of AT(t-2), AT(t) residuals and SSt forms the maximum water level and is converted into the extreme TWLt, which are further interpolated into elevation grids for inundation modeling and risk analysis. Results show that cities in the Pearl River Delta face extremely high storm surge risk, and other regions, such as ChaoZhou and ShanTou in the north and ZhanJiang in the south also have large areas within the extremely high risk zones. The residual interpolations can better capture the spatial variability of the storm tide than the traditionally total interpolation. Accordingly, the flooding risk uncertainty is greatly reduced by the residual interpolation, thus offering better supports for coastal development planning and flooding hazards risk management. (C) 2016 Elsevier Ltd and INQUA. All rights reserved.