As part of the global ocean circulation the buoyancy driven thermohaline circulation (THC) transports heat and salt from the tropics to the northern North Atlantic, where North Atlantic deep-water (NADW) is formed by cooling and sinking. The system of redistribution of global water masses is commonly known by the “conveyor belt” metaphor [Broecker, 1991] (Figure 1), representing an idealised sketch of the global ocean circulation.

The THC and ultimately deep-water formation in the North Atlantic seems to be a sensitive part of the global ocean circulation. During the last decade a possible reduction of the THC has been demonstrated in several atmosphere-ocean model simulations [Manabe and Stouffer, 1993; Stocker and Schmittner, 1997; IPCC, 2001], caused by warming and freshening of high latitude surface water, associated with global warming. Both effects the high latitude warming and enhanced poleward transport of moisture in the atmosphere cause a reduction of the density in the formation regions of NADW [IPCC, 2001]. A less likely, but not impossible, scenario is a complete shut down of the THC after passing a critical threshold, which would have a dramatic effect on the climate of areas surrounding the North Atlantic [IPCC, 2001]. A shut down of the THC might be an irreversible process because of its multiple equilibria, which have been reported by ocean and climate models of different complexity [Stommel, 1961; Bryan, 1986; Marotzke and Willebrand, 1991; Stocker and Wright, 1991].

It is difficult to assess the likelihood of future changes in the THC, due to uncertainties in the response of the climate system to greenhouse warming. In this context the analyses of abrupt changes in the past offers the possibility to form quantitative hypotheses, about the causes, mechanisms and feedbacks of climate changes.