Sensitivities of Modelling Storm Surge to Bottom Friction, Wind Drag Coefficient, and Meteorological Product in the East China Sea

Dongdong Chu,Jicai Zhang,Yongsheng Wu,Xiaohui Jiao,Suhui Qian
DOI: https://doi.org/10.1016/j.ecss.2019.106460
IF: 3.229
2019-01-01
Estuarine Coastal and Shelf Science
Abstract:In this study, effects of meteorological product, wind drag coefficient, and the bottom drag coefficient on the modelling storm surge in the East China Sea were investigated by using a high-resolution model based on FVCOM (Finite Volume Community Ocean Model). The model was first evaluated against the observational storm surge caused by Typhoon Winnie; the sensitivities of modelling surge variations to different factors were then examined, including four different meteorological products (ERA-Interim, ERA5, CCMP, NCEP-CFSR), seven formulae of wind drag coefficient (Peng & Li, Large & Pond, Garratt, Wu, Large & Yeager, Edson, and Zijlema), and six cases of bottom drag coefficient. The results indicated that all the experiments could capture temporal variations of the surge elevations. However, NCEP-CFSR wind field performs the best among the four wind field products. The wind drag coefficient formulae of Large & Yeager produce better results than the other formulae. The formulae of Edson, Wu, and Garratt produce higher surge elevations than those of the Large & Pond and Zijlema at the time of peak surge. Decreasing the bottom friction has a greater impact on surge elevations and current velocities than increasing the bottom friction. The non-linear interaction between tides and surge was studied as well, and the results showed that the non-linear effect contributed by 37% to the peak surge. The best combination of wind field and parameters derived from the sensitivity studies was used for the other three different storms (Chan-Hom, Herb and Mireille), and the simulations indicated that the best combination of forcing and drag coefficient obtained in this study in general can improve the performance of storm surge models.
What problem does this paper attempt to address?