Affordable Access

Probabilistic Forecast of Coastal Waves for Flood Warning Applications at Reunion Island (Indian Ocean)

Authors
  • Lecacheux, Sophie
  • Bonnardot, François
  • Paris, F.
  • Rousseau, Marie
  • Pedreros, Rodrigo
  • Lerma, Alexandre
  • Quetelard, Hubert
  • Barbary, David
Publication Date
May 13, 2018
Source
HAL-UPMC
Keywords
Language
English
License
Unknown
External links

Abstract

La Reunion being a small isolated island, the coastal impacts of tropical cyclones transiting nearby are tightly related to the relative position of the track, wind intensity and size. Thus, the forecast of cyclone-induced waves should be conducted in a probabilistic manner. Using ensemble track and intensity forecasts of Bonnardot and Quetelard (2016), the approach presented here relies on two steps: (1) Considering that the use of parametric models leads to an underestimation of wave heights distant from the cyclone eye, the forecasted vortex is bogused into the large-scale wind field provided by ECMWF's IFS model for each track and time step. A short simulation is conducted with Meso-NH model at 8 km resolution so that it can create realistic wind and pressure fields well balanced with the large-scale conditions (2) A combination of a two-way nested Wavewatch 3 modelling framework enables to compute the waves associated to each track from 10 km to 300 m resolution at the coast. Results extracted at 50 m depth are presented as exceedance values and probabilities of exceedance of wave heights for different coastal segments and thresholds. The comparison with measurements and simulations performed with Best-Track data show that probabilistic wave forecasts performed with the ensemble sets of Bonnardot and Quetelard (2016) upgrade the precision and the level of information provided compared to the deterministic method. Nevertheless, the study also points out that the size and shape parameters of the cyclones should be considered in the ensemble generation method to produce more accurate wave forecast.

Report this publication

Statistics

Seen <100 times