Answers are grounded in this paper and cite their sources.
The numerical ensemble prediction is a well accepted method for improving the performanceof atmospheric models. In the context of ocean wave modeling little has been researched ordocumented about this technique. An essential study of the method of ensemble predictionapplied to deep water waves has been carried out. A framework is defined for obtaining perturbationsof the directional wave spectra and for employing an ensemble of wind fields generatedby an atmospheric model. The third-generation global wave model WAM is used with realatmospheric conditions to investigate the effect on wave predictions of perturbed initial conditionsand atmospheric forcing. Due to spectral shape stabilisation, perturbing wave initialconditions has limited utility in ensemble prediction. However, the members could be used inwave data assimilation schemes in an interactive way. Using ensembles of the atmosphericcondition can generate diverging solutions, justifying the ensemble procedure by itself. In thecases studied, it is observed that the ensemble mean outperformed the other members. Thesolution behaviour suggests using a lower-order approximation of the model to generateensemble members with less computational cost.
No takes yet. Share an insight, caveat, or question.
Leandro Fariña (2002) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: