Causes and Consequences of the Late 1960s Great Salinity Anomaly

The second half of the 20th century showed a series of decadal-scale anomalies of salinity, temperature and sea ice cover in the northern North Atlantic. One pronounced event, the 'great salinity anomaly' (GSA) is observed in the late 1060s and the early 1970s (Dickson et al., 1988). This anomaly can be linked to the sea ice volume out of the Arctic through Fram Strait, which represents a major source of freshwater (Aagaard & Carmack, 1989; Schmith & Hansen, 2003). Increased sea ice export from the Arctic Ocean stabilizes the upper water column in the North Atlantic, diminishes the production of intermediate and deep water masses through ocean convection and can influence the large-scale ocean circulation (Häkkinen, 1999). An important part of sea ice export from the Arctic is forced by specific atmospheric structures (Hilmer et al., 1998; Häkkinen and Geiger, 2000; Cavalieri 2002). These can be parts of coupled ocean-atmosphere modes of climate variability with large-scale projections. The large-scale Atlantic meridional overturning circulation (AMOC) is a key component of the climate system. It can be defined as the part of the global ocean circulation, which is driven by density differences. The relatively dense surface waters in the North Atlantic generate downward mixing and southward movement of deep-water masses. This is partially balanced by a transport of saline waters by surface ocean currents from the tropics to midlatitudes. Changes in the AMOC can be very abrupt and have a worldwide climatic impact as demonstrated in theoretical and numerical studies (Stommel, 1961; Rahmstorf, 1994). There is significant evidence that the past abrupt climate changes (Dansgaard et al., 1993; Heinrich, 1988) were linked to AMOC variations (EPICA, 2006). One of the largest climate changes during the last 10.000 years occurred about 8.200 years ago (Alley, 2001) and it had a large impact over the Northern Hemisphere. It was proposed that this event included a freshwater pulse entering the North Atlantic (Barber et al., 1999) and modeling simulations suggest that it was associated with AMOC reduction (Renssen et al., 2002). Given the prominence of the GSA and its potential worldwide impact through the AMOC, it is of significant importance to understand the physical mechanisms associated to this outstanding climatic anomaly. Therefore, the goal of this study is to investigate the causes and the consequences of the GSA. The "climate mode" represents an useful concept to reduce the complexity of climate variations and can be defined as a set of physical processes with coherent spatial variability and quasi-periodic variations in time. Based on North Atlantic sea ice and sea level pressure