Seasonality and forcing factors of the Alaskan Coastal Current in the Bering Strait from July 2011 to July 2012

Abstract

A relatively narrow (~85km) and shallow (~50m) Bering Strait is the only connection between the Pacific and Arctic oceans. Flow through this strait dominates water properties of the Chukchi Sea, impacts Arctic sea ice and stratification, and may influence global climate through freshwater input to the sinking zones of the global thermohaline circulation. A buoyant coastal current, the Alaskan Coastal Current (ACC), typically present in the eastern Bering Strait from approximately late April to late December, contributes significantly to the total heat and freshwater fluxes through the strait, (adding approximately 1020J/yr of heat (Tref=-1.9ºC) and 600km3/yr of freshwater (Sref=34.8psu)), and drives much of the spatial variability in water properties in the eastern Chukchi Sea. However, the seasonal variability of this current has not yet been quantified in any detail. We use temperature, salinity, and velocity data from a 6-mooring array deployed across the eastern channel of the Bering Strait from July 2011 to July 2012 to study the seasonality and driving mechanisms of the ACC. We find the ACC is present (and flowing strongly) in July 2011, but disappears from the strait in November 2011. It starts to reappear in May 2012, and is well established by July 2012. Building on the known high correlation between flow and local wind (r~0.7), we examine a single value decomposition of the meridional velocity flow structure in the strait and see that the dominant mode (59% of the variance) contains a strong surface intensified signal trapped to the eastern coast reminiscent of the ACC. Through the use of a simple linear interpolation “box” method, we improve previous transport estimates during the months of strongest ACC transport by ~34%. Using this analysis, we find the greatest monthly mean of ACC transport is in August 2011 at 0.44 ± 0.06 Sv, equivalent to ~29% of the entire Bering Strait throughflow. At the ACC’s peak in August 2011, mooring data suggests its monthly mean salinity was at least 2.1 psu fresher and 3.9 ºC warmer than the main Bering Strait throughflow, although summer hydrographic sections suggest this is an underestimate, likely because the upper mooring instrumentation is ~18 m below the surface. We compare many parameters from theoretical studies of buoyant coastal currents (with or without wind forcing) with the mooring observations and conclude that our observations poorly constrain the current’s width and many different theories poorly estimate the current’s depth. Theory also suggests that the ACC is mainly driven by buoyancy forcing rather than wind forcing. We present an in-depth analysis on three different ways that wind forcing can impact a buoyant coastal current, viz., separation from the coast, flow reversals, and isopycnal tilting. We find that theory from Csanady (1977) successfully predicts ACC separation from the coast during strong southward wind events, consistent with separation observed in hydrographic data from August 2018. Additionally, we find an approximately 8 m/s southward wind is required to reverse flow of the ACC, and finally that two different theoretical parameters for estimating isopycnal tilting time bound observations, with the parameter 𝑡𝑎 from Moffat and Lentz (2012) being more accurate than 𝑡𝑡𝑖𝑙𝑡 from Whitney and Garvine (2005). Data from the United States Geological Survey shows the Yukon River discharge is most highly correlated (r=0.76) to ACC freshwater transport with a 12-15 day lag, and that the Yukon River discharge, combined with other rivers from the Alaskan peninsula, as well as the Unimak Pass freshwater discharge, are all probable ACC freshwater sources.