Scholarlink

In early September 2017, Hurricane Irma made landfall in Saint-Martin and generated major storm surges. The present study reveals the strong sensitivity of such storm surge modeling to the choice of atmospheric forcing. Several simulations are performed with the Coastal and Regional Ocean Community model under various atmospheric forcings: either a typical global reanalysis, or a parametric vortex with two tested resolutions (0.25° and 0.08°). The best results are obtained with the higher resolution parametric vortex forcing, leading to a maximum storm surge of 1.83 m in agreement with observations. By contrast, using a global reanalysis forcing that strongly underestimates the hurricane intensity only captures 16% of the observed surge. Limitations associated to using a parametric forcing are finally unveiled, with an overestimated negative surge as the hurricane is leaving the island, most likely attributed to missing local vortex asymmetries and topographic effects on the wind field. A small time lag between the modeled and observed surges also likely results from uncertainties in the exact position of the cyclone.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Sensitivity of Storm Surge Modeling to the Representation of Atmospheric Forcing: The Case Study of Hurricane Irma

  • 会议论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Aline Zribi,
  • Swen Jullien,
  • Guillaume Dodet,
  • Xavier Bertin,
  • Nils Herry

摘要

In early September 2017, Hurricane Irma made landfall in Saint-Martin and generated major storm surges. The present study reveals the strong sensitivity of such storm surge modeling to the choice of atmospheric forcing. Several simulations are performed with the Coastal and Regional Ocean Community model under various atmospheric forcings: either a typical global reanalysis, or a parametric vortex with two tested resolutions (0.25° and 0.08°). The best results are obtained with the higher resolution parametric vortex forcing, leading to a maximum storm surge of 1.83 m in agreement with observations. By contrast, using a global reanalysis forcing that strongly underestimates the hurricane intensity only captures 16% of the observed surge. Limitations associated to using a parametric forcing are finally unveiled, with an overestimated negative surge as the hurricane is leaving the island, most likely attributed to missing local vortex asymmetries and topographic effects on the wind field. A small time lag between the modeled and observed surges also likely results from uncertainties in the exact position of the cyclone.