An assessment of the added value from data assimilation on modelled Nordic Seas hydrography and ocean transports. (March 2016)

• Record Type: Journal Article
• Authors: Lien, Vidar S.; Hjøllo, Solfrid S.; Skogen, Morten D.; Svendsen, Einar; Wehde, Henning; Bertino, Laurent; Counillon, Francois; Chevallier, Matthieu; Garric, Gilles
• Is Part Of: Ocean modelling. Volume 99(2016:Mar.)
• Page Start: 43
• Page End: 59
• Publication Date: 2016-03
• Subjects: Ocean modelling -- Ocean reanalysis -- Atlantic Water -- Model comparison -- Volume transport -- Norwegian Sea; Oceanography -- Periodicals; Océanographie -- Périodiques; Oceanography; Periodicals; 551.46
• DOI: 10.1016/j.ocemod.2015.12.010
• Languages: English
• ISSNs: 1463-5003

Abstract

Highlights: A comprehensive assessment of the effect of including data assimilation. DA improves the model performance as estimated from independent measurements. A free high-resolution regional model performs similarly to coarser models with DA. Abstract: The Nordic Seas is a hotspot both in terms of climate related processes, such as Atlantic–Arctic heat exchange, and natural marine resources. A sustainable management of the marine resources within the Nordic Seas, including the co-existence between fisheries and petroleum industries, requires detailed information on the state of the ocean within an operational framework and beyond what is obtainable from observations only. Numerical ocean models applying data assimilation techniques are utilized to address this need. Subsequently, comprehensive comparisons between model results and observations are required in order to assess the model performance. Here, we apply a set of objective statistics to quantitatively assess the added value of data assimilation in numerical ocean models that are currently used operationally. The results indicate that the inclusion of data assimilation improves the model performance both in terms of hydrographic properties and volume and heat transports. Furthermore, we find that increasing the resolution towards eddy resolving resolution performs similarly to coarser resolution models applying data assimilation in shelf areas.