The Dynamics of Equatorially Asymmetric Thermohaline Circulations

Abstract

The three-dimensional dynamics of equatorially asymmetric thermohaline flow are investigated using an ocean general circulation model in a highly idealized configuration with no wind forcing and nearly fixed surface density. Small asymmetries in surface density lead to strongly asymmetric meridional overturning patterns, with deep water formed in the denser (northern) hemisphere filling the abyss. The poleward deep transport in the lighter hemisphere implies that the deep zonal-mean zonal pressure gradient reverses across the equator. Density along the eastern boundary and the zonally averaged density are nearly symmetric about the equator except at very high latitudes; the Southern Hemisphere western boundary thermocline, in contrast, is balanced by weaker upwelling and is hence broader than its northern counterpart. This pattern is explained through the spinup of the asymmetric circulation from a symmetric one, the timescale of which is set through advection by the mean deep western boundary current.For the strength of the interhemispheric transport, a lower bound of one-half the one-hemisphere overturning strength is derived theoretically for small finite forcing asymmetries, implying that the symmetric circulation is unlikely to be realized. Under asymmetric surface forcing, enhanced mixing in the denser hemisphere suppresses interhemispheric transport. Conversely, very strong cross-equatorial transport is caused by enhanced mixing in the lighter hemisphere. These results indicate that, once the surface densities determine that North Atlantic Deep Water is the dominant ventilating source, its export rate from the North Atlantic is controlled by mixing and upwelling in the rest of the World Ocean.

The role of Atlantic Ocean-atmosphere coupling in affecting North Atlantic oscillation variability

“…They involve tropical extra-tropical oceanic teleconnections set rapidly (months) by boundary Kelvin waves [e.g., Kawase, 1987;Johnson and Marshall, 2002] as a response to changes in the rate of deep-water formation at high latitudes, and the subsequent generation of long planetary Rossby waves along the eastern boundary of the North and South Atlantic [Wajowicz, 1986]. Further delay between changes in buoyancy forcing at high latitudes and oceanic changes elsewhere might also be introduced by oceanic advection and storage [e.g., Marotzke and Klinger, 2000]. The possibility that changes in meridional oceanic heat transport might be actively involved in (not simply responding to) the observed interdecadal NAO variability is examined below.…”

Response of the meridional overturning circulation to variable buoyancy forcing in a double hemisphere basin

“…This happens with the same periodicity as the surface forcing strength of the overturning in each hemisphere depends partly on the interhemispheric density gradient as this will, in effect, set the surface to bottom temperature difference. This surface to bottom temperature difference determines the strength of the overturning, which means that a fairly weak temperature difference between the high northern and high southern latitudes leads to significant differences in terms of the strength of the overturning in each hemisphere, as noted by Marotzke and Klinger (2000).…”

“…Here, we extend these results by including a second hemisphere. Indeed, interhemispheric dynamics have been shown to be crucial for the dynamics and stability of the MOC Marotzke 1999, 2000;Marotzke and Klinger 2000) Hence to get a proper hold on the circulation in the North Atlantic, considering the other hemisphere is essential. A double-hemisphere study has the important added complexity of theoretically allowing the existence of two high latitude deep water production sites, a situation much closer to what is observed in today's ocean with deep water production in the North Atlantic and to a lesser extent in the Weddell Sea (Pickard and Emery 1990).…”

Can limited ocean mixing buffer rapid climate change?

“…Various studies suggest as little as several months within a hemisphere (e.g. Kawase, 1987;Johnson and Marshall, 2002) or as much as several decades (McDermott, 1996;Marotzke and Klinger, 2000). However, the relationship between meridional gradients and overturning transport appears to remain valid in the transient state in at least one GCM (Thorpe et al, 2001).…”

“…There is strong evidence that circulation driven by interhemispheric gradients is important (e.g. Rooth, 1982;Rahmstorf, 1996;Marotzke and Klinger, 2000), and that symmetric cells in each hemisphere are unstable to asymmetric perturbations (Bryan, 1986;Vellinga, 1996;Weijer and Dijkstra, 2001;Nilsson et al, 2004). Conceptual models that have helped to explain this exhibit dynamics that are explicitly excluded when considering one hemisphere in isolation.…”