Unit:
Κατεύθυνση Φυσική ΩκεανογραφίαLibrary of the School of Science
Author:
Karagiorgos Ioannis
Supervisors info:
Σοφιανός Σαράντης, Επίκουρος Καθηγητής, Τμήμα Φυσικής, ΕΚΠΑ
Τόμπρου-Τζέλλα Μαρία, Καθηγήτρια, Τμήμα Φυσικής, ΕΚΠΑ
Τριανταφύλλου Γεώργιος, Διευθυντής Ερευνών, Ινστιτούτο Ωκεανογραφίας, ΕΛΚΕΘΕ
Original Title:
Quantification of ocean model uncertainties in an ensemble of high-resolution Bay of Biscay simulations
Translated title:
Quantification of ocean model uncertainties in an ensemble of high-resolution Bay of Biscay simulations
Summary:
Quantifying model uncertainty and error bounds is a key outstanding challenge in ocean state estimation and prediction. In this study, we investigate the impact of different ensemble generation methods in terms of forecast uncertainty representation within the context of a regional/coastal model. As a case study for open-ocean and coastal shelf dynamics, we use a coupled ocean–biogeochemical configuration of the ocean model NEMO (Nucleus for European Modelling of the Ocean) for the Bay of Biscay at 1/36 ◦ resolution. An ocean ensemble forced by the atmospheric model ensemble product of the European Centre for Medium-Range Weather Forecasts (ECMWF-EPS) is compared with an ensemble that generated through applying stochastic perturbations to an unperturbed (control) run. The stochastic perturbations are chosen with the aim to represent potential model forecast uncertainties from assumptions subject to erroneous atmospheric forcing, improper ocean model parameterizations, and ecosystem state uncertainties. The comparison is made through ensemble statistics focusing on the spatio-temporal variability of model spread for the coupled system. In summary, we find that the ECMWF-EPS forcing does not produce a large wind spread compared to the stochastic modelling approach. As a result of this, the stochastic method can be considered as the primary approach to generate upper-ocean model uncertainties. This study is concluded with a qualitative evaluation of the model forecasts with respect to the observations.
Main subject category:
Science
Keywords:
Ocean modelling, Ensemble forecast, Model uncertainties, Stochastic modelling, NEMO, Bay of Biscay
