Technical paper

A novel approach to generating a hurricane database for the Gulf of Mexico based on numerical weather prediction models

Published

05 Jul 2019

Authors

Liria, J., Coelho, H., Sproson, D., Martinho, P., Webb, C., Oropeza, F., Bradon, J., Smith, R., Peng, Z

DOI

Accurate, reliable and cost effective hindcast databases have become essential tools for the design of offshore and coastal structures. Within the Gulf of Mexico (GoM), many databases exist, which cover operational wave climates, but very few of these fully resolve the complexities of Tropical Revolving Storms (TRS) in terms of both the spatial and temporal resolution.

Accurate, reliable and cost effective hindcast databases have become essential tools for the design of offshore and coastal structures. Within the Gulf of Mexico (GoM), many databases exist, which cover operational wave climates, but very few of these fully resolve the complexities of Tropical Revolving Storms (TRS) in terms of both the spatial and temporal resolution. Here, we present the results of our hindcast analysis, covering all TRS affecting the GoM from 1979 to the present. This database represents all the physical and dynamic aspects of the TRS, reproducing not only the cyclonic winds, but also how the hurricane interacts with the atmospheric mean flow in the area surrounding the eyewall. The Weather Research and Forecasting model (WRF-ARW), with two nested fixed domains (18 km and 6 km), and uncoupled Hurricane WRF (HWRF) have been used to model the atmospheric conditions. Waves have been modeled using Wave Watch III (WW3), a third-generation wave model, driven by the 18 km WRF modeled winds. Boundary conditions from our inhouse global wave database have been used, forced by CFSR winds. Most of the existing databases produce a set of TRS based on the records of wind, track, pressure and radius of maximum speed, producing a uniform field of wind and pressure. We have developed a state of the art wind and wave database of TRS based on fully convective, non-hydrostatic numerical models. Model results have been validated against observational data from buoys, meteorological stations, airborne measurements, satellite data and against tracks from the International Best Track Archive for Climate Stewardship (IBTrACS) database. Results show that the modeled TRS are quite sensitive to the initial state, obtaining the best results when initializing the model with the vortex in a mature state. The hurricane intensification and wind distribution are well reproduced by the HWRF model, being in good agreement with both the track and measurements. This approach provides a realistic, cost effective and up to date hindcast of the wind and wave conditions during hurricane events throughout the GoM. Using cutting edge models, we have been able to reproduce hurricane conditions over the whole GoM, obtaining results that are physically representative.

Proceedings of the Annual Offshore Technology Conference, 2, pp. 769-793.

Download the video here:
https://www.onepetro.org/presentation/OTC-28909-PT

Read the full paper here:
https://www.onepetro.org/conference-paper/OTC-28909-MS