Mixing processes in the changing Arctic Ocean

Abstract

The Arctic has undergone tremendous changes the last decades, including a strong decline in sea ice extent and thickness. The rapid pace of Arctic changes relative to the global changes are known as Arctic amplification, and has been referred to as the ‘canary in the coalmine’ of the present climate changes. Factors contributing to the accelerated changes are the ice-albedo effect, and the vast heat reservoir of Atlantic water flowing in the ocean below. This study has aimed to describe and quantify the influence of oceanic heat on the heat budget at the ocean’s upper boundary. There is a delicate heat balance at the interface between the atmosphere, the sea ice and the ocean. A small change in heat flux can have large effect on the ice cover. While the Arctic Ocean is generally not a very energetic one, the recent changes has raised concern about whether internal wave energy and the importance of vertical mixing processes are increasing. Reductions in sea ice extent may allow for more momentum transfer from the atmosphere to the ocean, either mixing the surface layer directly, or initiating inertial oscillations in the boundary layer. Near-inertial internal waves may propagate into the interior and cause mixing away from the surface boundary layer. An increase in vertical mixing in the Arctic Ocean may bring up more heat from the underlying warm Atlantic Water, posing a further threat to the diminishing Arctic sea ice. The study is based on observations from two different campaigns, both located in the region north of Svalbard. First, under-ice boundary layer and upper ocean measurements made during the winter-to-spring drift campaign N-ICE2015. Second, a yearlong deployment of three moorings on the slope of the Yermak Plateau is used to study the near-inertial wave field by the plateau. From an under-ice turbulence mast, a unique data set of winter-time measurements over the deep basin is obtained. Direct measurements of heat fluxes are weakly positive, even in winter, which are roughly doubled during storm events. Individual events can cause an order of magnitude increase in fluxes. A one-dimensional vertical diffusion model based on the observations from the drift satisfactorily reproduced observed changes in upper ocean winter hydrography. The model further suggests that observed salinity increase in the mixed layer was dominated by entrainment of saline water from below, rather than brine rejection from ice growth. In spring, coincident with drift over the shallower topography, where the warm Atlantic Water is found at shallow depths, heat fluxes below the sea ice are much higher. Varying by one to two orders of magnitude, heat fluxes are highly dependent on the depth of the warm water layers, the wind forcing and its effect on the ice cover. Highest heat fluxes exceeded 100Wm−2 over several hours, during a wind event in the marginal ice zone. From a subset of the under-ice turbulence measurements, during sea ice melt in June, heat and salt fluxes are found to be inversely correlated. This is contrary to expectations of positive heat- and salt fluxes during sea ice melt. This is hypothesized to origin in salt released from the melting sea ice. Objective criteria are used to identify 131 salty plumes descending past the measurement volume, accounting for 9% of the salt fluxes in only 0.5% of the time. The accumulated salt flux indicate a near full desalination of the sea ice. The reduction in bulk salinity of two nearby ice cores, taken three days apart, agree with accumulated salt flux within a factor of two. Plumes have previously only been observed from land-fast ice in a Svalbard fjord. The study confirms its existence on drifting Arctic sea ice, with implications for the understanding of salt and freshwater distribution in the under-ice boundary layer, and brine drainage in sea ice. From the southwestern Yermak Plateau, the near-inertial field was analyzed in yearlong records from three moorings. The near-inertial signal is clockwise dominant, indicative of downward energy propagation. The clockwise polarization is stronger closer to the surface, further suggesting surface generation by winds. Examples of wind-generated near-inertial wave propagation are presented, and wave group properties are calculated. At mid-depth and in the deep, episodic events of elevated near-inertial horizontal kinetic energy can be caused by surface generation at a remote location, or by tidal currents interacting with the rough topography. Theoretical characteristic beam paths initiated at the shelf break are consistent with the mid-depth elevation in near-inertial horizontal kinetic energy. The sum of these observations further highlight the importance and complexity of ocean mixing processes, both at the ice-ocean interface and at depth. The Yermak Plateau is a region of significant internal wave generation and energetic turbulence, and will be an important and interesting region for further studies. The diapycnal mixing taking place here is key in determining the vertical exchange of heat between inflowing Atlantic water and the surface, and the fate of this heat in the Arctic basins.