Structure and forcing of the overflow at the Storfjorden sill and its connection to the Arctic coastal polynya in Storfjorden

• Published in: Ocean Science
• Main Authors: Geyer, F., Fer, I., Smedsrud, L. H.
• Format: Text
• Language: English
• Published: 2018
• Subjects: Arctic; Svalbard; Sea ice; Storfjorden
• Online Access: https://doi.org/10.5194/os-6-401-2010; https://os.copernicus.org/articles/6/401/2010/

Storfjorden (Svalbard) is a sill-fjord with an active polynya and exemplifies the dense water formation process over the Arctic shelves. Here we report on our simulations of Storfjorden covering the freezing season of 1999–2000 using an eddy-permitting 3-D ocean circulation model with a fully coupled dynamic and thermodynamic sea-ice model. The model results in the polynya region and of the dense water plume flowing over the sill crest are compared to observations. The connections of the overflow at the sill to the dense water production at the polynya and to the local wind forcing are investigated. Both the overflow and the polynya dynamics are found to be sensitive to wind forcing. In response to freezing and brine rejection over the polynya, the buoyancy forcing initiates an abrupt positive density anomaly. While the ocean integrates the buoyancy forcing over several polynya events (about 25 days), the wind forcing dominates the overflow response at the sill at weather scale. In the model, the density excess is diluted in the basin and leads to a gradual build-up of dense water behind the sill. The overflow transport is typically inferred from observations using a single current profiler at the sill crest. Despite the significant variability of the plume width, we show that a constant overflow width of 15 km produces realistic estimates of the overflow volume transport. Another difficulty in monitoring the overflow is measuring the plume thickness in the absence of hydrographic profiles. Volume flux estimates assuming a constant plume width and the thickness inferred from velocity profiles explain 58% of the modelled overflow volume flux variance and agrees to within 10% when averaged over the overflow season.