Temperature-Salinity Criterion for Inhibition of Deep Convection

Dan E. Kelley Oceanography Department, Dalhousie University, Halifax, Nova Scotia, Canada

Search for other papers by Dan E. Kelley in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Malmberg's salinity criterion for the inhibition of oceanic deep convection is extended here to account for increases in salinity caused by evaporation brine rejection, and mixed-layer deepening. Roughly speaking, accounting for evaporation permits deep convection for waters up to ∼0.2 psu fresher than Malmberg's critical salinity of 34.7 psu. An additional 0.2 psu of freshness is permitted in regions of ice formation at rates such as that in the Greenland Sea. Typically a further 0.1 psu is permitted because of the salinizing effect of mixed-layer deepening. On a global ocean scale, the difference between Malmberg's criterion and the present criterion is relatively minor: both criteria suggest that the North Atlantic is salty enough to feed the global thermohaline overturning cell but that the North Pacific is too fresh. On a regional scale, the difference between the criteria is more significant. This is illustrated with surface salinity maps for the Greenland Sea, a region known to produce bottom water for the overturning cell. Each criterion predicts that the relatively fresh inshore waters are not capable of deep convection but that the saltier offshore waters are. However, the new criterion places the onshore-offshore dividing line 100–200 km closer to Greenland that does Malmberg's criterion. A large geographical area lies between the two dividing lines and would thus be misjudged by Malmberg's criterion. Furthermore, the new stability boundary is shifted inshore by a distance equal to the offshore decay scale of the buoyancy flux associated with cold air outbreaks from the continent. In effect, then, the buoyancy fluxes driving the ocean convection are much larger than otherwise would have been predicted.

Abstract

Malmberg's salinity criterion for the inhibition of oceanic deep convection is extended here to account for increases in salinity caused by evaporation brine rejection, and mixed-layer deepening. Roughly speaking, accounting for evaporation permits deep convection for waters up to ∼0.2 psu fresher than Malmberg's critical salinity of 34.7 psu. An additional 0.2 psu of freshness is permitted in regions of ice formation at rates such as that in the Greenland Sea. Typically a further 0.1 psu is permitted because of the salinizing effect of mixed-layer deepening. On a global ocean scale, the difference between Malmberg's criterion and the present criterion is relatively minor: both criteria suggest that the North Atlantic is salty enough to feed the global thermohaline overturning cell but that the North Pacific is too fresh. On a regional scale, the difference between the criteria is more significant. This is illustrated with surface salinity maps for the Greenland Sea, a region known to produce bottom water for the overturning cell. Each criterion predicts that the relatively fresh inshore waters are not capable of deep convection but that the saltier offshore waters are. However, the new criterion places the onshore-offshore dividing line 100–200 km closer to Greenland that does Malmberg's criterion. A large geographical area lies between the two dividing lines and would thus be misjudged by Malmberg's criterion. Furthermore, the new stability boundary is shifted inshore by a distance equal to the offshore decay scale of the buoyancy flux associated with cold air outbreaks from the continent. In effect, then, the buoyancy fluxes driving the ocean convection are much larger than otherwise would have been predicted.

Save