Description: Hydrographic, nutrient and halocarbon tracer data collected in July–August 1994 in the Norwegian Sea, the Faroe Bank Channel (FBC), the Iceland and Irminger Basins and the Iceland Sea are presented. Special attention was given to the overflow waters over the Iceland–Scotland Ridge (ISOW). The Iceland–Scottland overflow water (ISOW) was identified along its pathway in the Iceland Basin, and entrainment of overlying water masses was quantified by multivariate analysis (MVA) using principal component analysis (PCA) and Partial Least Square (PLS) calibration. It was concluded that the deeper portion of the ISOW in the FBC was a mixture of about equal parts of Norwegian Sea Deep Water (NSDW) and Norwegian Sea Arctic Intermediate Water (NSAIW). The mixing development of ISOW during its descent in the Iceland Basin was analysed in three sections across the plume. In the southern section at 61°N, where the ISOW core was observed at Full-size image (〈1 K) depth, the fraction of waters originating north of the ridge was assessed to be 54%. MVA assessed the fractional composition of the ISOW to be 21% NSDW, 22% NSAIW, 18% Northeast Atlantic Water (NEAW), 11% Modified East Icelandic Water, 25% Labrador Sea Water (LSW) and 3% North East Atlantic Deep Water. It may be noted that the fraction of NEAW is of the same volume as the NSDW. On its further path around the Reykjanes Ridge, the ISOW mixed mainly with LSW, and at 63°N in the Irminger Basin, it was warmer and fresher (θ=2.8°C and S=34.92) than at 61°N east of the ridge Full-size image (〈1 K). The most intensive mixing occurred immediately west of the FBC, probably due to high velocity of the overflow plume through the channel, where annual velocity means exceeded Full-size image (〈1 K). This resulted in shear instabilities towards the overlying Atlantic waters and cross-stream velocities exceeding Full-size image (〈1 K) in the bottom boundary layer. The role of NSAIW as a component of ISOW is increasing. Being largely a product of winter convection in the Greenland Sea when no Greenland Sea Deep Water (GSDW) is formed, it spreads above the older and denser deep water in the Nordic Seas. Little or no GSDW, which earlier was considered to be the principal overflow water, has been formed since 1970. This shows that the Iceland–Scotland overflow may also be maintained with intermediate waters as the principal overflowing component. Decadal variability in ISOW properties has not been insignificant, as since the early 1960s there has been a decrease in salinity and temperature, by 0.06 and up to 0.5°C, respectively. Such a trend applies also to the LSW, particularly in the Irminger Basin, where it was warmer, saltier and less dense in the late 1950s and early 1960s Full-size image (〈1 K) than in 1994 Full-size image (〈1 K). CFC tracers were used to assign apparent ages of water masses, showing that the NSDW had an apparent age of about 30 years and that the age of Iceland Sea Deep Water exceeded 25 years. NSAIW observed in the southern Norwegian Sea was estimated to be 6–16 years old. An upper age limit of LSW in the Iceland Basin was found to be 18–19 years. It was further concluded that the products of the onset of intense wintertime convection in the Labrador Sea in the late 1980s were not yet observed in the northern central part of the Iceland Basin. The LSW in the Irminger Basin was found to be significantly younger. Two layers were found there. A shallower layer at a depth of 1000–Full-size image (〈1 K) depth was older than the layer beneath by about 4 years, while the deeper layer at 1500–Full-size image (〈1 K) depth was assessed at an apparent age ranging between less than 1 (formed during the previous winter) and 4 years.

Description: In summer 1996, a tracer release experiment using sulphur hexafluoride (SF6) was launched in the intermediate-depth waters of the central Greenland Sea (GS), to study the mixing and ventilation processes in the region and its role in the northern limb of the Atlantic overturning circulation. Here we describe the hydrographic context of the experiment, the methods adopted and the results from the monitoring of the horizontal tracer spread for the 1996–2002 period documented by ∼10 shipboard surveys. The tracer marked “Greenland Sea Arctic Intermediate Water” (GSAIW). This was redistributed in the gyre by variable winter convection penetrating only to mid-depths, reaching at most 1800 m depth during the strongest event observed in 2002. For the first 18 months, the tracer remained mainly in the Greenland Sea. Vigorous horizontal mixing within the Greenland Sea gyre and a tight circulation of the gyre interacting slowly with the other basins under strong topographic influences were identified. We use the tracer distributions to derive the horizontal shear at the scale of the Greenland Sea gyre, and rates of horizontal mixing at ∼10 and ∼300 km scales. Mixing rates at small scale are high, several times those observed at comparable depths at lower latitudes. Horizontal stirring at the sub-gyre scale is mediated by numerous and vigorous eddies. Evidence obtained during the tracer release suggests that these play an important role in mixing water masses to form the intermediate waters of the central Greenland Sea. By year two, the tracer had entered the surrounding current systems at intermediate depths and small concentrations were in proximity to the overflows into the North Atlantic. After 3 years, the tracer had spread over the Nordic Seas basins. Finally by year six, an intensive large survey provided an overall synoptic documentation of the spreading of the tagged GSAIW in the Nordic Seas. A circulation scheme of the tagged water originating from the centre of the GS is deduced from the horizontal spread of the tracer. We present this circulation and evaluate the transport budgets of the tracer between the GS and the surroundings basins. The overall residence time for the tagged GSAIW in the Greenland Sea was about 2.5 years. We infer an export of intermediate water of GSAIW from the GS of 1 to 1.85 Sv (1 Sv = 106 m3 s−1) for the period from September 1998 to June 2002 based on the evolution of the amount of tracer leaving the GS gyre. There is strong exchange between the Greenland Sea and Arctic Ocean via Fram Strait, but the contribution of the Greenland Sea to the Denmark Strait and Iceland Scotland overflows is modest, probably not exceeding 6% during the period under study.