Ensemble mean wave height or wave height of ensemble mean? (Abstract/Poster in convegno)

  • Ensemble mean wave height or wave height of ensemble mean? (Abstract/Poster in convegno) (literal)
  • 2014-01-01T00:00:00+01:00 (literal)
Alternative label
  • Pezzutto, Paolo; Cavaleri, Luigi (2014)
    Ensemble mean wave height or wave height of ensemble mean?
    in Waves in Shallow Water Environment (WISE), ECMWF, Reading (UK), 8 to 12 June 2014
  • Pezzutto, Paolo; Cavaleri, Luigi (literal)
  • Poster (literal)
  • CNR - ISMAR - Istituto di scienze marine (literal)
  • Ensemble mean wave height or wave height of ensemble mean? (literal)
  • Operational wave ensemble prediction systems are based on a number of realizations of sea states, i.e. a number of parallel third-generation wave model runs. It is common practice to build forecasts (and thus verify the system skills) by evaluating ensemble mean and estimating ensemble spread of relevant physical quantities. Typically significant wave height (HS ), mean period (Tm02 ) and mean direction (?m). These are member by member integrated quantities that are defined on geographic space and time. Integration of the spectrum in frequency and direction necessarily causes loss of knowledge on energy distributions. In principle, one would pursue the best results with the least loss of informations. For example, for each point in geographic space and time it could be more sound to evaluate the average sea state (mean energy per frequency and direction) and then integrate it to derive the expected main spectral moments. As an example with significant wave height, it's easy to show that from an ensemble mean sea state one would retrieve an expected HS that is strictly greater than the average of all single EPS wave heights. If the proposed procedure turns out to be effective, it could be worth to dress a wave EPS system with a post processor. The latter should receive, time-step by time-step, all EPS members spectra and derive the integrated quantities and their uncertainties as expected values and spread of the ensemble of sea states. Within the MyWave project, the limitation of data storage resources made difficult to keep detailed memory of the five dimension grid data. Note that spectra are necessary for the evaluation of energy ensemble spread. However, accounting for ensemble mean only, the estimate of difference with respect to common practice is easily retrievable case by case. We are therefore able to present differences between ensemble mean HS and Tm02 and corresponding quantities as integrals of ensemble mean spectra, for the NETTUNO-WAM EPS system. The dataset covers the whole Mediterranean sea for six the month testing period (July-December 2013). (literal)
Prodotto di
Autore CNR

Incoming links:

Autore CNR di