Sy, A. et al. (2002): Upper Ocean Climate Ship-of-Opportunity Programme of BSH - A Status Report 
17 
Koltermann et al., 1999; Curry and McCartney, 2001). Since Rogers (1984), the NAO pattem 
has been described by an index of that pressure difference between Portugal/Azores and 
Iceland (Koslowski and Löwe, 1994; Hurrel, 1995; 1996; Löwe and Koslowski, 1998). A high 
NAO index, characterized by an intense Iceland low and a strong Azores high, indicates strong 
westerly winds, while in a low NAO index case the signs of this dipole are reversed. The NAO 
seesaw-like mode of variability is particularly dominant in winter months, and time-series have 
shown that the NAO index has undergone major low-frequency variations during the last 
century (Koslowski and Löwe, 1994; Hurrell, 1995). However, the role of this complex dynamic 
ocean-atmosphere-land system of the North Atlantic and the mechanisms of its interaction are 
not yet fully understood and are, therefore, investigated intensively. (CLIVAR, 1998; DFG, 
2000). A central question is whether the ocean is a passive participant in climate change merely 
responding to atmospheric conditions, or a more active component of the climate (McCartney, 
1997) 
Different explanations have been proposed to explain the low-frequency NAO variability, e.g. as 
a response to changes in the ocean or to changes of external forcings such as solar radiation, 
or internal generation in the atmosphere. However, the lack of long-term observation records 
makes it difficult to confirm one of the different hypothesis. 
On the other hand, low-frequency variability of the ocean is poorly understood and, as is to be 
expected, the strong seesaw-like atmospheric changes have been held responsible for a wide 
range of phenomena observed in all layers of the ocean. Some selected examples are briefly 
described below. 
The largest known dislocation of the freshwater balance in the surface layer of the subpolar 
gyre, the so-called Great Salinity Anomaly which circulated this gyre for a 14-year period from 
1968 to 1982 (Dickson et al., 1988), was explained by the NAO minimum in the 60s due to an 
extreme amplification of the Icelandic low. 
Observations from WOCE reveal surprisingly large and rapid changes in the water mass 
distribution of the intermediate and upper layers of the North Atlantic. In the Subpolar Mode 
Water layer, significant changes in the baroclinic structure along the eastern margin of the sub 
polar gyre were observed in the mid 90s coinciding with the strong decrease of the NAO Index 
(Bersch et al., 1999; Flatau et al., 2002). 
The layer below the main thermocline, where Labrador Sea Water (LSW) dominates the 
intermediate depth level and which had been assumed to have a nearly constant temperature 
and salinity, is undergoing major changes at present. It was the most important oceanic 
occurrence of the 90s in the North Atlantic, associated with the evolution of the NAO from its 
extreme negative state recorded in winter 1960 to its extreme positive state in the early 1990s. 
This event was characterized by marked cooling of intermediate waters which proceeded at 
annual intervals (cascades), and the newly formed so-called “1988 LSW cascade” (Sy et al., 
1997a, b) was fresher, colder, and denser than at any other time in the history of deep 
measurements in that area. Its signal spread from its source area towards the European shelf 
with a mean speed of about 1.5 to 2 cm/s (Sy et al., 1997a). That is three to four times faster 
than previously estimated by Read and Gould (1992). With a delay of 6 years the signal 
appeared in the deep water of the subtropical basins near Bermuda (Curry et al., 1998). The 
Labrador Sea is one of the two convective cells of the North Atlantic Deep Water production. 
There is evidence that the NAO-induced substantial changes in the formation of North Atlantic 
Deep Water affect the Meridional Overturning Circulation (MOC), which is the driving power 
engine at the northern terminus of the THC, and thus the key element of global THC (Dickson 
et al., 2002). Data from the last 40 years indicate that the MOC is subject to strong and natural 
variability on time scales of 10 to 30 years which is correlated with the NAO (Koltermann et al., 
1999; Lorbacher, 2000). However, the question as to cause and effect still remains.