Typhoon wind hazard estimation by full-track simulation with various wind intensity models

Mingfeng Huang, Qing Wang, Qiang Li, Renzhi Jing, Ning Lin, Lizhong Wang

Research output: Contribution to journalArticlepeer-review

24 Scopus citations

Abstract

A general framework of typhoon full-track simulation for the Western North Pacific (WNP) basin is presented in this paper. Specifically, an empirical TC track model integrated with two intensity models, i.e., a classical Vickery intensity regression model (VIRM) and a novel Markov Environment-Dependent Hurricane Intensity Model (MeHiM) are developed. The applicability of MeHiM for typhoon simulation in WNP basin is validated by comparing several metrics (i.e., intensity spatial distribution, probabilistic distributions) with those from observation data. The full-track model parameters are calibrated and modified for the local area of the WNP basin, and storm tracks from the genesis to lysis are generated. The proposed full-track typhoon simulation framework together with a widely-used wind field model is employed to estimate design wind speed with a given return period for ten major cities of Southeast China. The results show that the estimated 100-year return period design wind speeds in simulated tropical cyclone climates are in good agreement with that prescribed in the current design code for some cities. For other cities, the comparison is complicated.

Original languageEnglish (US)
Article number104792
JournalJournal of Wind Engineering and Industrial Aerodynamics
Volume218
DOIs
StatePublished - Nov 2021

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Renewable Energy, Sustainability and the Environment
  • Mechanical Engineering

Keywords

  • Extreme wind speed
  • Full-track simulation
  • Monte Carlo simulation
  • Typhoon
  • Wind intensity model

Fingerprint

Dive into the research topics of 'Typhoon wind hazard estimation by full-track simulation with various wind intensity models'. Together they form a unique fingerprint.

Cite this