Experiences in multiyear combined state-parameter estimation with an ecosystem model of the North Atlantic and Arctic Oceans using the Ensemble Kalman Filter.

• Published in: Journal of Marine Systems
• Main Authors: Simon, Ehouarn, Samuelsen, Annette, Bertino, Laurent, Mouysset, Sandrine
• Other Authors: Algorithmes Parallèles et Optimisation (IRIT-APO), Institut de recherche en informatique de Toulouse (IRIT), Université Toulouse Capitole (UT Capitole), Université de Toulouse (UT)-Université de Toulouse (UT)-Université Toulouse - Jean Jaurès (UT2J), Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université de Toulouse (UT)-Toulouse Mind & Brain Institut (TMBI), Université Toulouse - Jean Jaurès (UT2J), Université de Toulouse (UT)-Université de Toulouse (UT)-Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Université Toulouse Capitole (UT Capitole), Université de Toulouse (UT), Institut National Polytechnique (Toulouse) (Toulouse INP), Nansen Environmental and Remote Sensing Center Bergen (NERSC), Hjort Center for Marine Ecosystem Dynamics
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2015
• Subjects: Data assimilation; Ensemble Kalman filter; Combined state–parameter estimation; Clustering analysis; Ecosystem modeling; [INFO.INFO-DC]Computer Science [cs]/Distributed; Parallel; and Cluster Computing [cs.DC]; [SDE.BE]Environmental Sciences/Biodiversity and Ecology; [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG]; Arctic; Arctic Ocean; Longhurst; North Atlantic; Zooplankton
• Online Access: https://hal.science/hal-02960624; https://hal.science/hal-02960624/document; https://hal.science/hal-02960624/file/simon_15387.pdf; https://doi.org/10.1016/j.jmarsys.2015.07.004

International audience A sequence of one-year combined state–parameter estimation experiments has been conducted in a North Atlantic and Arctic Ocean configuration of the coupled physical–biogeochemical model HYCOM-NORWECOM over the period 2007–2010. The aim is to evaluate the ability of an ensemble-based data assimilation method to calibrate ecosystem model parameters in a pre-operational setting, namely the production of the MyOcean pilot reanalysis of the Arctic biology. For that purpose, four biological parameters (two phyto- and two zooplankton mortality rates) are estimated by assimilating weekly data such as, satellite-derived Sea Surface Temperature, along-track Sea Level Anomalies, ice concentrations and chlorophyll-a concentrations with an Ensemble Kalman Filter. The set of optimized parameters locally exhibits seasonal variations suggesting that time-dependent parameters should be used in ocean ecosystem models. A clustering analysis of the optimized parameters is performed in order to identify consistent ecosystem regions. In the north part of the domain, where the ecosystem model is the most reliable, most of them can be associated with Longhurst provinces and new provinces emerge in the Arctic Ocean. However, the clusters do not coincide anymore with the Longhurst provinces in the Tropics due to large model errors. Regarding the ecosystem state variables, the assimilation of satellite-derived chlorophyll concentration leads to significant reduction of the RMS errors in the observed variables during the first year, i.e. 2008, compared to a free run simulation. However, local filter divergences of the parameter component occur in 2009 and result in an increase in the RMS error at the time of the spring bloom.