Ensemble simulations of the magnetic field induced by global ocean circulation: Estimating the uncertainty

• Published in: Journal of Geophysical Research: Oceans
• Main Authors: Irrgang, C., Saynisch, J., Thomas, M.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: 2016
• Subjects: Antarctic; The Antarctic; Antarc*
• Online Access: https://gfzpublic.gfz-potsdam.de/pubman/item/item_1478172; https://gfzpublic.gfz-potsdam.de/pubman/item/item_1478172_4/component/file_1529111/1478172.pdf

The modelling of the ocean global circulation induced magnetic field is affected by various uncertainties that originate from errors in the input data and from the model itself. The amount of aggregated uncertainties and their effect on the modelling of electromagnetic induction in the ocean is unknown. For many applications, however, the knowledge of uncertainties in the modelling is essential. To investigate the uncertainty in the modelling of motional induction at the sea surface, simulation experiments are performed on the basis of different error scenarios and error covariance matrices. For these error scenarios, ensembles of an ocean general circulation model and an electromagnetic induction model are generated. This ensemble-based approach allows to estimate both the spatial distribution and temporal variation of the uncertainty in the ocean induced magnetic field. The largest uncertainty in the ocean induced magnetic field occurs in the area of the Antarctic Circumpolar Current. Local maxima reach values of up to 0.7 nano Tesla. The estimated global annual mean uncertainty in the ocean induced magnetic field ranges from 0.1 to 0.4 nT. The relative amount of uncertainty reaches up to 30 % of the signal strength with largest values in regions in the northern hemisphere. The major source of uncertainty is found to be introduced by wind stress from the atmospheric forcing of the ocean model. In addition, the temporal evolution of the uncertainty in the induced magnetic field shows distinct seasonal variations. Specific regions are identified which are robust with respect to the introduced uncertainties. This article is protected by copyright. All rights reserved.