Summary of changes and trends

• Atlantic waters and adjacent shelf areas had low winter and summer sea surface salinity (SSS) in the mid-late 1970s (associated with the passage of the Great Salinity Anomaly (GSA)), followed by three decades of large inter-annual variability.
• Salinity records from the Faroe Shetland Channel and the Ellett line indicate a recent trend to high salinity.
• SSS averaged over the northern North Sea from 1950 to 2002 shows decreasing salinity since the 1970s and is reflected by observations at fixed locations in the Fair Isle Current and the North Sea fishing grounds.
• There is no discernible trend in mean SSS in the English Channel from 1900 to the early 1980s.
• SSS averaged over the Irish Sea from 1950 to 2002 shows a decrease in both winter and summer.

1. Introduction

The ocean circulation is determined primarily by the forcing due to momentum, heat and water fluxes to and from the atmosphere, and by the distributions of salinity (and temperature) in the ocean that determine its density structure. Changes in salinity cause changes in density and hence in density currents, thus affecting large scale thermo-haline circulation and local dynamics in coastal and estuarine locations. In particular, the density structure affects the ‘meso-scale’ dynamics of fronts and eddies, which are the most energetic motions in the ocean.

Salinity plays an important role in marine ecosystems. For example, nitrogen is a key element limiting primary production and its release from estuarine sediments alters in response to changes in overlying water salinity. An increase in salinity causes a decrease in the rate of denitrification and hence a nitrogen loss, thus determining the year-to-year variation in primary productivity, particularly in beginning and maintaining summer phytoplankton blooms. Also, many organisms are adapted to certain salinity conditions, especially in saline pool, lagoon and estuarine locations. The egg and larval stages of some fish species also depend on salinity tolerance.

Descriptions of the monitoring networks that regularly measure salinity are given in Chapter 1, including details of how to access near real-time data.

Click here for a list of links to monitoring networks and data sets.

2. North Atlantic

Figures 1 and 2 show winter and summer sea surface salinity (SSS) data respectively, averaged over the eastern North Atlantic (55-60N, 25-15W). (The averaging of salinity should normally be done over small regions within a water mass, but this area lies in an area of weak salinity gradients, particularly in winter when deep convection produces a 600m deep homogeneous water column (Dooley, 2003)). The low values observed in both winter and summer during the 1970s are associated with the Great Salinity Anomaly (GSA). This is believed to have been caused by the arrival of ‘fresher’ water that was created in the Arctic in the 1960s and then had drifted across the Atlantic in the prevailing current, arriving in UK waters in the mid-1970s (Dickson et al., 1988). The GSA formed during the extreme negative index phase of the NAO in the late 1960s, when clockwise flow around anomalously high pressure over Greenland fed record amounts of freshwater from the Arctic Ocean through the Fram Strait into the Nordic Seas. From there some of the fresh water passed through the Denmark Strait into the sub polar North Atlantic Ocean gyre. There have been other similar events in the past, and statistical analyses have revealed that the generation and termination of these propagating salinity modes are closely connected to a pattern of atmospheric variability strongly resembling the NAO.

The somewhat higher interannual variability observed during the late 1980s is probably a manifestation of the very low number of observations at this time, rather than a reflection of actual conditions (Dooley, 2003).

During 2002 salinity was higher than the long-term average in most areas of the North Atlantic and increased to the highest levels observed in over a decade (ICES 2003a).

3. UK Waters

Only a few long-term time series exist for UK waters and these do not indicate any overall long-term trend. Records from the Atlantic waters around the UK (Rockall and the Faroe Shetland Channel) and the areas on the shelf which are most influenced directly by in-flowing oceanic waters (Fair Isle, E1) reveal a general pattern of low salinity in the mid-late 1970s. This period was influenced by the passage of the GSA (see section 2 above) as it circulated around the Atlantic and Nordic Seas, and was then followed by three decades of quite large inter-annual variability probably closely associated with changes in the NAO. In the shallower areas of the North Sea and Irish Sea, the salinity is much more dependent on local runoff from land and local evaporation / precipitation changes, and hence is much more variable.

3.1 North Sea

The salinity characteristics of the North Sea are strongly influenced by freshwater exchange via the atmosphere and rivers, and also by water inflow from the North Atlantic. The North Atlantic Current brings oceanic water of high salinity into the northern North Sea in two branches: an inflow through the Fair Isle channel off the north of Scotland and a more significant inflow along the western slope of the Norwegian Trench. The Norwegian Coastal Current forms a contrasting outflow along the eastern side of the Trench, carrying less saline surface water from fjords and rivers northward. In addition to the oceanic inflow to the northern North Sea, the saline water of Atlantic origin also penetrates into the southern North Sea through the Dover Straits (ICES, 2003b). The main input of fresh water into the North Sea is from rivers discharging along its southern coast, so the southern North Sea is less salty than the northern North Sea. The salinity of the North Sea reflects the influence of the NAO on the movement of Atlantic water into the North Sea and the meteorological forcing of the ocean-atmosphere heat exchange and resulting precipitation (ICES, 2003a). Strongly positive values of the NAO Index are linked to low surface summer salinity and strongly negative values to high surface summer salinity.

Only a few really long salinity time series exist for the North Sea but none in UK waters. A salinity series measured at Helgoland in the German Bight since 1873 shows no significant trend over the 120 years of observations. Relatively high salinities were observed in the 1920s, at the end of the 1960s, and from 1989-95. In the late 1970s, and for most of the 1980s, the salinity was relatively low (OSPAR, 2000).

Figures 3 and 4 show time series of SSS anomalies averaged over the northern North Sea from 1950 to 2002. The salinity is higher in winter, reflecting changes in atmospheric forcing; in particular, the precipitation-evaporation balance of these seasons, with evaporation over the sea much higher in winter than summer (Dooley, 2003). The averaging of salinity should be done over small regions within a water mass and this area is one of significant spatial salinity gradients, both in winter and summer. However, the apparent freshening (i.e. decreasing salinity) since the 1970s, shown especially in the summer time series, is reflected at observations at fixed locations - see figure 5 (Fair Isle Current) and figures 6 to 12 (North Sea fishing grounds).

North Sea salinities returned to the long-term average values in 2002 following the low values observed in 2001, which were probably due to stronger than normal run-off from the continental rivers (ICES, 2002; ICES, 2003a).

3.2 English Channel and Celtic Sea, including the Bristol Channel

Figure 13 shows that there is no discernible trend in mean sea surface salinity at MBA station E1 from 1900 to the early 1980s.

3.3 Irish Sea

Figures 14 and 15 show winter and summer SSS anomalies averaged over the Irish Sea (53-55N, 6-4W) from 1950 to 2002. There is an indication of freshening (i.e. decreasing salinity), especially in the summer time series, but the observation coverage decreased significantly from the early 1980s onwards – for example, the very low value of the mean in 1991 is based on only one value (Dooley, 2003).

Figure 16 shows the annual mean SSS since 1966 at the Port Erin site on the Isle of Man and Figure 17 shows an analysis of the monthly means of SSS for the years 2001 and 2002, with long-term minima, maxima and monthly means. Surface salinity data from CTD casts further west, at station 38a, is shown in figure 18. There is a positive correlation between lower salinity at the Cypris station and the positive phase of the NAO (Gaynor Evans, personal communication, 2004).

3.4 Minches, west Scotland, Scottish continental shelf and Faroe Shetland Channel

Figure 19 shows that there has been an overall increasing trend in salinity in the Faroe Shetland Channel since the low values caused by the passage of the GSA in the late 1970s. During 2001 there was a downward turn but a return to higher values in 2002 (ICES, 2003a).

3.5 Rockall Trough and Bank and Atlantic north west approaches

Figure 20 shows de-seasoned upper ocean (0-800m) salinity anomalies from the Rockall Trough, from 1975. The maximum salinity anomaly of the time series occurred at the end of the 1990s, in May 1998 (Holliday, 2003). In contrast the early part of the 1990s was characterised by relatively low salinity, although it remained higher than the low values caused by the passage of the GSA in the late 1970s. In 2001 the Rockall Trough began to show signs of freshening (and cooling), following a peak in salinity (and temperature) in 1998–2000. However salinity (and temperature) remained high compared to the long-term mean, with values similar to previous peaks in the early 1980s (ICES, 2002).

Holliday (2003) states that the NAO Winter Index shows no statistically significant correlation with the time-series of subsurface salinity (or temperature); the Rockall Trough lying in a region of low to zero correlation, between the high positive correlation to the south and east and high negative correlation to the west (Rodwell et al., 1999). Therefore she considers that the conditions in the Rockall Trough do not appear to be directly related to atmospheric conditions, as indicated by the NAO Index; or to variations in local net freshwater fluxes; but instead are caused by varying inputs of the water masses to the south of the region - central North Atlantic Water (ENAW), Mediterranean outflow Water (MEDW), Western North Atlantic Water (WNAW) and SubArctic Intermediate water (SAIW).

4. References

Dickson, R. R., J. Meincke, S. A. Malmberg, and A. J. Lee (1988). The “Great Salinity Anomaly” in the northern North Atlantic 1968–1982, Progress in Oceanography, 20: 103–151.

Dooley, H. (2003). Personal communication. ICES, Denmark, Copenhagen.

Holliday, N.P. (2003). Extremes of temperature and salinity during the 1990s in the northern Rockall Trough: results from the “Ellett line”. ICES Marine Science Symposia, 219: 95-101.

ICES (2002). The Annual ICES Ocean Climate Status Summary 2001/2002. Prepared by the Working Group on Oceanic Hydrography, ICES, Copenhagen, Denmark. (Editors: Bill Turrell and N. Penny Holliday.) Retrieved 20th May 2003 from the World Wide Web: http://www.ices.dk/status/clim0102/IOACSS01.PDF

ICES (2003a). The 2002/2003 ICES Annual Ocean Climate Status Summary. Prepared by the Working Group on Oceanic Hydrography, ICES, Copenhagen, Denmark. (Editors: Sarah L. Hughes & Alicia Lavín.) Retrieved 4th October 2003 from the World Wide Web: http://www.ices.dk/status/clim0203/IAOCSS2002.PDF

ICES (2003b). Environmental status of the European seas. Edited by Neil Fletcher.
Retrieved 26th September 2003 from the World Wide Web:
http://www.ices.dk/reports/germanqsr/23222_ICES_Report_samme.pdf

OSPAR (2000). Quality Status Report 2000 for the North-East Atlantic. OSPAR (Oslo-Paris) Commission, London. 108 + vii pp.