Transient halocline and freshwater dynamics of the Arctic's Beaufort Gyre

Georgy E Manucharyan, Woods Hole Oceanographic Institution, Woods Hole, MA, United States and Michael A Spall, WHOI, Woods Hole, MA, United States
Abstract:
Freshwater content (FWC) in the Arctic Ocean is continuously changing with a large contribution associated with changes in halocline properties of the Beaufort Gyre. Transient dynamics of the gyre is, in turn, linked to variations in surface stress forcing arising from anticyclonic atmospheric winds and modified by the presence of sea ice. Here, we present a theoretical study which explores major mechanisms of the halocline adjustment to time-dependent surface stress forcing. Using an idealized eddy resolving numerical model of the Beaufort Gyre we show that for steady anticyclonic forcing the halocline deepening due to Ekman pumping (and hence the accumulation of FWC) is dominantly counteracted by the mesoscale eddy transport that acts to flatten the halocline. Achieving this major balance results in a decadal spin-up time scale of the gyre — a time scale needed for the mesoscale eddies to exchange water masses between the gyre interior and its coastal boundaries. Following the Transformed Eulerian Mean framework and making use of eddy salt flux parameterization we obtain analytical expressions for halocline depth and its adjustment time scale that are in agreement with the eddy resolviong simulations. In case of oscillating surface stress forcing the amplitude and phase of the halocline response is also well explained by assuming that a residual between the eddy and the Ekman transports drives halocline deepening. Thus, e.g. for decadal and longer periods the halocline is in phase with forcing, whereas for a seasonal cycle it lags surface forcing by a quarter of a period. We further discuss implications for the time evolution of the total (area integrated) FWC of the gyre. We demonstrate that only the gyre-scale wind stress perturbations are most efficient at changing its total FWC, with specially localized perturbations being least efficient. Finally, we suggest a potentially useful index for monitoring Beaufort Gyre’s total FWC by relating its time evolution to observations of surface stress and halocline slopes located only along the gyre boundaries.