thumbnail

Surface drift prediction in the Adriatic Sea using hyper-ensemble statistics on atmospheric, ocean and wave models: Uncertainties and probability distribution areas

Journal of Marine Systems

By:
, , and
DOI: 10.1016/j.jmarsys.2007.02.015

Links

Abstract

Despite numerous and regular improvements in underlying models, surface drift prediction in the ocean remains a challenging task because of our yet limited understanding of all processes involved. Hence, deterministic approaches to the problem are often limited by empirical assumptions on underlying physics. Multi-model hyper-ensemble forecasts, which exploit the power of an optimal local combination of available information including ocean, atmospheric and wave models, may show superior forecasting skills when compared to individual models because they allow for local correction and/or bias removal. In this work, we explore in greater detail the potential and limitations of the hyper-ensemble method in the Adriatic Sea, using a comprehensive surface drifter database. The performance of the hyper-ensembles and the individual models are discussed by analyzing associated uncertainties and probability distribution maps. Results suggest that the stochastic method may reduce position errors significantly for 12 to 72??h forecasts and hence compete with pure deterministic approaches. ?? 2007 NATO Undersea Research Centre (NURC).

Additional Publication Details

Publication type:
Article
Publication Subtype:
Journal Article
Title:
Surface drift prediction in the Adriatic Sea using hyper-ensemble statistics on atmospheric, ocean and wave models: Uncertainties and probability distribution areas
Series title:
Journal of Marine Systems
DOI:
10.1016/j.jmarsys.2007.02.015
Volume
69
Issue:
1-2
Year Published:
2008
Language:
English
Larger Work Type:
Article
Larger Work Subtype:
Journal Article
Larger Work Title:
Journal of Marine Systems
First page:
86
Last page:
98