Performance comparison of meso-scale ensemble wave forecasting systems for Mediterranean sea states. (August 2016)

• Record Type: Journal Article
• Authors: Pezzutto, Paolo; Saulter, Andrew; Cavaleri, Luigi; Bunney, Christopher; Marcucci, Francesca; Torrisi, Lucio; Sebastianelli, Stefano
• Is Part Of: Ocean modelling. Volume 104(2016:Aug.)
• Page Start: 171
• Page End: 186
• Publication Date: 2016-08
• Subjects: Ensemble forecast -- Mediterranean sea -- Limited area model -- Short-range -- Ocean waves forecast; Oceanography -- Periodicals; Océanographie -- Périodiques; Oceanography; Periodicals; 551.46
• DOI: 10.1016/j.ocemod.2016.06.002
• Languages: English
• ISSNs: 1463-5003

Abstract

Highlights: We estimate the skill of two new short-range wave EPS for the Mediterranean Sea. We compare two wave forecast systems driven by different atmospheric scale models. The coarse wind EPS achieves similar or better results than the high resolution control member. The right result is achieved, ensemble probabilistic additional value is verified. Abstract: This paper compares the performance of two wind and wave short range ensemble forecast systems for the Mediterranean Sea. In particular, it describes a six month verification experiment carried out by the U.K. Met Office and Italian Air Force Meteorological Service, based on their respective systems: the Met Office Global-Regional Ensemble Prediction System and the Nettuno Ensemble Prediction System. The latter is the ensemble version of the operational Nettuno forecast system. Attention is focused on the differences between the two implementations (e.g. grid resolution and initial ensemble members sampling) and their effects on the prediction skill. The cross-verification of the two ensemble systems shows that from a macroscopic point of view the differences cancel out, suggesting similar skill. More in-depth analysis indicates that the Nettuno wave forecast is better resolved but, on average, slightly less reliable than the Met Office product. Assessment of the added value of the ensemble techniques at short range in comparison with the deterministic forecast from Nettuno, reveals that adopting the ensemble approach has small, but substantive, advantages.