The oceanography of winter leads

Abstract

Leads in pack ice have long been considered important to the thermodynamics of the polar regions. A winter lead affects the ocean around it because it is a density source. As the surface freezes, salt is rejected and forms more dense water which sinks under the lead. This sets up a circulation with freshwater flowing in from the sides near the surface and dense water flowing away from the lead at the base of the mixed layer. If the mixed layer is fully turbulent, this pattern may not occur; rather, the salt rejected at the surface may simply mix into the surface boundary layer. In either event the instability produced at the surface of leads is the primary source of unstable buoyancy flux and, as such, exerts a strong influence on the mixed layer. Here as many as possible of the disparate and almost anecdotal observations of lead oceanography are assembled and combined with theoretical arguments to predict the form and scale of oceanographic disturbances caused by winter leads. The experimental data suggest the velocity disturbances associated with lead convection are about 1–5 cm s−1. These appear as jets near the surface and the base of the mixed layer when ice velocities across the lead are less than about 5 cm s−1. The salinity disturbances are about 0.01 to 0.05 psu. Scaling arguments suggest that the geostrophic currents set up by the lead density disturbances are also of the order of 1–5 cm s−1. The disturbances are most obvious when freezing is rapid and ice velocity is low because the salinity and velocity disturbances in the upper ocean are not smeared out by turbulence. In this vein, lead convection may be characterized at one extreme as free convection in which the density disturbance forces the circulation. At the other extreme, lead convection may be characterized as forced convection in which the density disturbance is mixed rapidly by boundary layer turbulence. The lead number Lo, which is the ratio of the pressure term to the turbulence term in the momentum equation, and the turbulent lead number Lot, which is the ratio of buoyant production to shear production in the turbulent kinetic energy equation, define the boundary between the free and forced regimes. For Lo and Lot less than one, both the large‐scale circulation and the turbulence are forced by surface stress. For Lo and Lot greater than one, both the large‐scale circulation and the turbulence are forced by the buoyancy flux. The magnitudes of velocity and salinity disturbances from a model developed elsewhere, suitable to free convection, agree with what few observations we have. The results of a forced convection model, developed here, suggest salinity disturbances of the order of 0.01–0.02 practical salinity units, with the maximum occurring at the surface of the lead and decreasing substantially below 5–10 m. This unstable gradient is a unique characteristic of lead convection. Though the salinity disturbances may be small when ice velocities are large, the buoyancy flux in leads has a major effect on the boundary layer turbulence.

Insight into chemical, biological, and physical processes in coastal waters from dissolved oxygen and inert gas tracers

“…However, other processes may enhance near-surface turbulence in the presence of sea ice including convection associated with heat loss and brine rejection (Morison et al, 1992;Smith and Morison, 1993), boundary layer shear between ice and water (McPhee, 1992;Saucier et al, 2004), and wave interactions with drifting ice (Kohout and Meylan, 2008).…”

Currents and convection cause enhanced gas exchange in the ice–water boundary layer

“…Without replenishment, the gas partial pressure differential would become diminished, leading to a decreased gas flux and corresponding decrease in the estimate of k. Under some circumstances, the same phenomenon might occur in the real iceÁocean boundary layer; however, Ekman turning is a persistent feature of the IOBL, leading to continual relative motion between the ice and the water beneath it (McPhee, 1992). Additionally, lead openings produce their own local circulation (Morison et al, 1992), which would also lead to surface gas renewal. The low surface renewal might explain the low values of k during experiments 16 and 18; however, the current speed was nearly as low during experiment 15 when the tracers predicted a much greater value of k. Aside from the uncertainty in the tracer mass balance when the open water area is low and fluxes are small, we have no adequate explanation for these apparently contradictory results.…”

Sensitivity of simulated CFC‐11 distributions in a global ocean model to the treatment of salt rejected during sea‐ice formation

“…The motivation for the Test simulation is that brine rejected during sea-ice formation in the real ocean typically sinks through the mixed-layer to the pycnocline that lies beneath the mixed layer [Morison et al, 1992]. The depth of 160 m to which we sink this salt in our Test simulation was chosen to produce This parameterization is intended to demonstrate the potential importance of processes associated with sea-ice formation that help maintain stratification.…”