THE THERMOHALINE OCEAN CIRCULATION: A SYSTEM WITH DANGEROUS THRESHOLDS? An Editorial Comment Threatening scenarios of a breakdown of the Atlantic thermohaline circulation (Figure 1), a collapse of northern European agriculture and fisheries, and of glaciers advancing on Scandinavia and Scotland

Threatening scenarios of a breakdown of the Atlantic thermohaline circulation (Figure 1), a collapse of northern European agriculture and fisheries, and of glaciers advancing on Scandinavia and Scotland have captured the popular imagination in recent years, with a number of newspaper reports, magazine articles and television documentaries covering this topic with a widely varying degree of accuracy. The risk of critical thresholds in the climate system being crossed where some irreversible qualitative change sets in (such as a major ocean circulation change) is taken increasingly seriously in the discussion on anthropogenic climate change. While the 1995 IPCC report (Houghton et al., 1995) only mentioned it in passing, the upcoming third IPCC assessment will devote substantial space to this issue. So where do we stand scientifically? Looking at the growing number of publications on thermohaline circulation stability (reviewed e.g. by Weaver and Hughes, 1992; Rahmstorf et al., 1996) one would be forgiven to be confused, finding the model results contradictory and the data inconclusive. Nevertheless I believe that a fairly consistent picture has emerged during the past years. The first robust conclusion taken for granted now but unknown only two-anda-half decades ago (Oort and Stommel, 1976) is that the thermohaline circulation makes a major contribution to the heat budget of the North Atlantic region (Roemmich and Wunsch, 1985), warming annual-mean surface temperatures locally by up to 10 C (Manabe and Stouffer, 1988). It is still a matter of debate how far the warm anomaly created by the oceanic heat transport extends into the European continent or affects the North American seaboard, but it is likely that at least the northwestern European countries (Great Britain and Ireland, Iceland, Scandinavia, Holland, Belgium) are warmed by several degrees, with the largest effect in winter. The second robust conclusion is that the system is sensitive to the amount of freshwater entering the North Atlantic. This freshwater sensitivity has been the subject of numerous modeling studies and can best be summarised in a simple schematic stability diagram (Figure 2). This shows how under present-day climatic conditions the equilibrium rate of Atlantic overturning changes if the freshwater budget (precipitation and runoff minus evaporation) of the northern North Atlantic is altered.