Sea‐ice interaction with the thermohaline circulation

Abstract

Linkages have been suggested between observed interdecadal variability of sea‐ice and salinity in the North Atlantic. A plausible mechanism for generating such variability through the interaction of sea‐ice and the thermohaline circulation (THC) is examined in a zonally‐averaged THC model coupled to a thermodynamic ice model. A self‐sustaining interdecadal oscillation arises through the feedbacks between salinity anomalies induced by the sea‐ice melting‐freezing process and anomalous meridional heat transport associated with the THC. The oscillation time scale is not associated with any oceanic time scales but fundamentally depends on ice‐THC coupling. The period is quite robust to stochastic forcing, although the regularity is strongly affected.

Low-frequency climate variability of an aquaplanet

“…The periodic appearance and disappearance of a stable polar halocline are also an important aspect of our low-frequency oscillation. However, the results of Yang and Neelin (1993) differ from ours not only in the time scale (interdecadal instead of centennial). Since we analyze an aquaplanet, the concept of the thermohaline circulation with a sinking region in the North Atlantic cannot be accurately adapted.…”

“…The meridional overturning circulation (MOC) of the ocean plays a central role in the analyses of these long-term climate variations (e.g., Rahmstorf 2001, andMikolajewicz andMaier Reimer 1990), and variations in the meridional heat transport of the ocean have often been interpreted as a direct consequence of the varying strength of the MOC, for example Weaver et al (1991). Broecker et al (1990) and Yang and Neelin (1993), as well as other studies, find that a strong overturning circulation causes higher oceanic heat transport and, thus, higher temperatures, while a weaker MOC phase leads to colder climate phases. However, here, we present results which suggest that warm climates are not necessarily connected to a stronger overturning circulation.…”

The anomalous sea‐ice extent in Hudson bay, Baffin bay and the Labrador sea during three simultaneous NAO and ENSO episodes

“…Alternatively, it may result from natural decadal oscillations in the thermohaline circulation (Weaver and Sarachik, 1991). We propose that it may be either a natural oscillation in the coupled ice-ocean system of the type described by Yang and Neelin (1993) or Zhang et al (1995), or a decadal-scale regional climate cycle involving feedbacks between the atmosphere, ocean, sea-ice cover and runoff of a form similar to that proposed in Mysak et al (1990) or Mysak and Power (1992). It is well known that runoff stiongly affects sea-ice extent in marginal seas (Prinsenberg, 1988;Manak and Mysak, 1989;Mysak and Power, 1991); thus runoff variations may also be important for Labrador Sea ice-cover variations (Myers et al, 1990).…”

“…The simplest approach is to use Boolean delay equations (e.g., Darby and Mysak, 1993), which in conjunction with the concept of a feedback loop (Mysak et al, 1990;Mysak and Power, 1992), allows one to experiment with a variety of lagged climate processes in an efficient manner. One could also use box (dynamical system) models to crudely represent different regions of the coupled atmosphere-ocean-ice system (e.g., Birchfield, 1989;Robitaille et al, 1995), and either two-dimensional ice-ocean models (Yang and Neelin, 1993) or threedimensional ice-ocean models (Weisse et al, 1994;Zhang et al, 1995). It is significant to note that Weisse et al (1994) detected a 10-to 40-year broad band signal in the output of their ice-ocean model of the North Atlantic which involved the generation of quasi-periodic salinity anomalies in the Labrador Sea, which were subsequently advected into the North Atlantic.…”

The interaction between sea ice and salinity-dominated ocean circulation: implications for halocline stability and rapid changes of sea ice cover

“…Using a Stommel-type box model coupled to an energy-balance representation of the atmosphere, Jayne and Marotzke (1999) found that sea ice related feedbacks destabilized the thermally-dominated mode of the thermohaline circulation. In their model, feedbacks between sea ice, meridional temperature gradients and atmospheric moisture transport were of key importance, whereas brine rejection associated with sea ice formation played a more central role in the box model of Yang and Neelin (1993). However, less attention has been paid to how sea ice feedbacks affect a salinity-dominated thermohaline circulation.…”