• The deep outflow of cold water from the Nordic seas over the Greenland - Scotland ridge has fallen by 20% since 1950, suggesting comparable reduced surface inflow from the Gulf Stream and the North Atlantic Current.
• Maximum flow conditions in the North Atlantic Current and the Subtropical Gyre occurred in 1995 and 2000 and minimum circulation conditions between 1996 and 1998.
• Two pulses of inflow into the North Sea in 1988/89 and 1998 coincided with unusually strong northward transport of anomalously warm water through the Rockall Trough.
• Coastal flow conditions from the Irish Sea to Scottish coastal waters changed considerably after 1977, with a further change in Irish Sea outflow during 1980 to 1981, after which the flow pattern returned to that of 1977-1980.

1. Introduction

1.1 Types of current

Tidal currents, or ‘streams’, are generated by astronomical forcing due to the varying gravitational attraction of the Moon and the Sun. UK waters respond strongly to tidal forcing at the Atlantic Ocean boundary - the general response is to amplify the semi-diurnal (two tides a day) component of the tide. Particularly strong responses occur in the Irish Sea and the Bristol Channel.

Meteorologically forced ‘surge’ currents are due to variations in wind stress and atmospheric pressure. The former depends upon water depth and increases in importance as the depth decreases whereas the pressure effect is independent of depth. Surge currents have time scales of hours to days according to storm duration, water depth and the extent of the storm.

Density currents are driven by density gradients due to changes in temperature and/or salinity, arising from the net flux of heat through the sea surface and freshwater inputs from rivers and the atmosphere respectively.

1.2 Circulation

The net movement of water, the circulation, is driven by ‘residual’ currents due to the net tides, mean meteorological forcing and the mean density distribution. (Currents due to upwelling contribute to exchanges but do not contribute significantly to the net movement.) The idea that circulation is a smooth, wide constant flow tends to be supported because it’s difficult to measure accurately (see Section 2) and so it has only been measured where it’s strong and persistent. In reality, circulation is variable in space and time, especially on short term (daily and monthly), seasonal and inter-annual timescales.

1.2.1 Short-term mean circulation

As tidal currents are primarily oscillatory, they usually contribute little to daily mean circulation, although exceptions can occur where the water depth is shallow or in regions near to a headland or island. However, the whole flow pattern may reverse over a tidal cycle, particularly in estuaries.

Over a few days, the net movement is likely to be determined by the last storm, because then the surge currents are likely to exceed both the tidal and density currents in strength, and so the daily or monthly mean circulation may even be the reverse of the long-term pattern.

1.2.2 Seasonal mean circulation

This is mainly due to the strong seasonality in surge and density currents (storms mainly occur in winter, river discharge has an annual cycle and solar input varies seasonally). In particular, residual currents are generated by seasonally occurring ‘fronts’, the sharp boundary between well-mixed and stratified (layered) regions, with flow tending to be along the front. Some regions tend to remain well mixed throughout the year where depths are shallow and tidal currents are strong enough to provide the energy for mixing, but other regions exhibit seasonal stratification when mixing is insufficient to mix down lighter water at the sea surface (the water may be lighter either because of solar heat input during summer or because of freshwater river discharge).

Usually, a frontal system includes a narrow (typically a few kilometres wide) jet-like current driven by the horizontal density difference (Rodhe, 1998). In particular, jets are associated with the margins of cold (or salty) dense bottom-water pools that remain trapped in deep basins during the summer months after the onset of summer stratification. Although relatively narrow, they can transport water over many hundreds of kilometres in areas of the North, Celtic and Irish Seas (see Section 4). The timing of the onset of this seasonal circulation is dependent on wind mixing, surface heat fluxes and freshwater input and may vary by up to a month (Brown et al., 1999; 2003).

There is also a vertical variation of the seasonal mean circulation because jets are stronger near the surface and the relative contributions to the residual current from surge or density currents may vary with depth. This is illustrated in Figures 1 and 2, showing the circulation at a site in Liverpool Bay over a six-week period in winter and over a year respectively. (During the latter period, the residual currents would have transported water over 1 000 km from the point of measurement.) Both figures show the typical variability in the vertical of an ‘estuarine / coastal’ type circulation, with flow near the bed in towards the coast and away from the coast at the surface. The prime driving force is density for both the outflow and the inflow, although the near surface is more affected by the wind (Howarth, personal communication).

1.3 The significance of circulation

The high density and specific heat of water means that it can store and transport large amounts of heat, so the role of the world’s ocean circulation is critical in the global climate system. A meridional (tropics to poles) transport of energy is required for the Earth system to be in global radiative balance, with some 30-50% of the energy carried by ocean currents at mid latitudes and a higher proportion at lower latitudes (Bryden and Imawaki 2001).

The overall movement and distribution of passive objects like eggs, larvae, nutrients, contaminants, flotsam and sediments are controlled by the circulation patterns. For example, the circulation flow off the north east coast of England provides a direct pathway for material and fish larvae from coastal regions to the northern Dogger Bank and central North Sea (Brown et al., 1999). On a smaller scale, the dispersal of herring larvae in the Blackwater estuary is dependent on the circulation in the area (Fox and Aldridge, 2000). In general, the movement will depend on the object’s density – if neutrally buoyant or dissolved it will move with the water circulation; if it is particulate or heavier than water it will tend to sink and move less far; if it is floating it will be driven by the wind as well as the water circulation (see the chapter on Sediments for further details).

2. Measuring circulation

The lack of spatially diverse and good quality long time series of observed currents makes the definition of long-term circulation and its variability difficult. Most long-term circulation patterns in UK waters have been inferred from the distribution of tracers like salinity or radionuclides or from numerical hydrodynamic models, optimised with any available observations.

The use of models has to be treated with caution because experience from European seas projects such as ESODAE, NOMADS, NOWESP and PROMISE suggest that different models can give the closest reproduction of observations, hind-casting, at different times. In fact, occasionally, the outlier of an ensemble of hind casts from different models may be the closest to reality (Jones, 2002).

However, although there may be large variability in the hind-casting of day to day or month to month currents from different models, the models are more consistent when used to determine the long-term circulation. For example, as part of the NOWESP project, Smith et al. (1996) found that three different models from the Institute of Marine Research, the Institut für Meereskunde and the Proudman Oceanographic Laboratory showed similar and persistent patterns of variability in the water volume transports calculated across sections in the North Sea and in the English Channel when run for periods up to 39 years with long term meteorological forcing. They therefore indicated consistent long term or ‘climatological’ residual currents in broad agreement with the generally accepted circulation patterns inferred from observations (see sections 4.1 and 4.2). Although the transport calculations agreed very well in well-mixed water regions, there was poorer agreement in the deeper water regions of the northern North Sea where baroclinic effects due to density changes are important and were not well modelled. Also, agreement was poor in the Irish Sea because of model limitations, i.e. low resolution, between approximately 20 and 35 km, and unsuitable advection schemes which add additional structure such as eddies and gyres which are exaggerated when model resolution is poor. (In fact, the models gave a climatological residual flow direction from the north, contrary to the northerly transport indicated by observations (see section 4.3)).

The difficulty in obtaining consistent circulation patterns from models is illustrated from the results of the NOMADS2 project (Delhez et al., 2004), which compared nine 3-dimensional advection-dispersion models of the southern North Sea (to 57 °N) run from November 1988 to October 1989. All the models used the same bathymetric, meteorological and hydrological data sets; the same initial and time-varying boundary conditions for water elevation, salinity, temperature and velocity) and the same prescription of the heat flux. However, the models varied with respect to horizontal and vertical resolution, the representation of surface wind stress and turbulence and the interpolation schemes used, i.e. the size of space and time steps. This led to large differences in some output parameters, e.g. there was a factor of 2.5 to 3 variation between models in the year-long volume residual fluxes across North Sea sections.

Delhez et al. (2004) concluded that much more development of 3-dimensional advection-dispersion models was needed before they are capable of delivering robust estimates of long term circulation patterns. Recently, POL’s 3.5km ‘ecosystem’ model of the Irish and Celtic Sea (49 - 56°N and 2 - 10°W) has been run over a 40 year period from 1960 – 2000 using realistic meteorological and hydrological forcing, including inflows from 106 rivers. It has reproduced well the time series and interannual variation of bottom temperature and salinity at the Cypris station off the Isle of Man (and has also successfully reproduced the observed doubling of nutrients between 1950 and 1990 and the subsequent levelling off) (Proctor, personal communication). Such models indicate the potential to deliver ‘state of the art’ physics simulation, from which realistic circulation patterns could be derived, and point to the need for an eddy-resolving 50 year run of such models over the rest of the European shelf seas.

Observational data on circulation comes from current meter measurements, drifting buoys and floats, submarine and telephone cables (to measure induced voltages across channels) and the concentration distribution of ‘tracers’ like salinity and radionuclides (e.g. Caesium 137 and Technetium 99). However circulation is difficult to measure accurately and can only be measured where it is strong and persistent. There are problems with current meters and ADCPs because the circulation is a weak signal in the presence of much stronger signals (typically with a signal to noise ratio of about 1%). The motion of floats is often difficult to interpret in continental shelf seas because of the usual short time of deployment and observation; and also because surface floats are affected by ‘windage’, the direct effect of the wind, so that their motion is not solely due to the current.) Circulation patterns can be determined well with tracers but the determination of current speed is difficult.

Descriptions of the monitoring networks which regularly measure currents and circulation are given in the chapter on Monitoring networks, including details of how to access near real-time data.

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

3. Circulation in the North Atlantic and along the continental shelf edge

3.1 North Atlantic

The North Atlantic Meridional Overturning Circulation (Namoc) is part of the current system that transports heat around the world. Surface currents, including the Gulf Stream and North Atlantic Current, transport (relatively) warm salty water into the Arctic. There the water loses heat and is diluted with fresh water from river inputs and the melting of ice and hence becomes denser; with deep colder fresher currents carrying the return flow southwards into the Atlantic. This ‘circulation conveyor belt’ helps drives Namoc and maintains the mild climate of northern Europe by warming the prevailing westerly winds blowing over the ocean surface.

The overflow and descent of cold dense water from the sills of the Denmark Strait and Faroe Shetland Channel is the principal means by which the deep Atlantic Ocean is ventilated and so is a key element of the Namoc. There is evidence (Dickson et al., 1999; 2002) that this system has steadily changed in character over the past four decades, resulting in a sustained and wide spread freshening of the deep waters south of the Greenland-Scotland ridge. Hansen et al. (2001) have monitored the deep outflow of cold water from the Nordic seas as it passes over the Greenland-Scotland ridge and show that the outflow has fallen by 20% since 1950, suggesting comparable reduced inflow from the Gulf Stream and the North Atlantic Current.

The North Atlantic Oscillation (NAO) (see the chapter on Weather and climate) controls or modifies three of the main parameters that drive ocean circulation - wind speed, air/sea heat exchange and evaporation/precipitation. Pingree (2002) used satellite altimeter data from 1992 to 2002 to calculate sea level anomalies (sla) and thus determine the changes in North Atlantic circulation over that period. He showed that the long-term changes in the North Atlantic Current and the Subtropical Gyre transport during this period correlate with the winter NAO Index, with maximum flow conditions occurring in 1995 and 2000 and minimum circulation conditions occurred between 1996-1998. Years of extreme negative winter NAO Index resulted in enhanced poleward flow along the eastern boundary and anomalous winter warming along the west European Continental slope, as was measured in 1990, 1996, 1998 and 2001.

3.2 Continental shelf edge, including Rockall Trough and Faroe Shetland Channel

Observations at the continental shelf edge indicate a poleward along-slope current, the European Slope Current (ESC), flowing along the entire length of the ocean-shelf boundary from the Goban Spur to north of Shetland, a distance of some 1600 km. The flow is forced by the combined effect of the steep topography and the mutual adjustment of shelf and oceanic regimes to meridional density gradients - the Joint Effects of Baroclinicity and Relief (JEBAR) effect (Simpson, 1998).

Currents and transports along the continental slope from the Celtic Sea to the Faroe Shetland Channel are summarised by Huthnance (1986). Estimated transports between the shelf break and the 2000m depth contour (probably the great majority) were fairly consistently poleward in the range 1 - 2 Sv (1 Sv = 10**6 m**3/s) from the Celtic Sea to the Wyville-Thomsen ridge. Mean current speeds were quoted are typically 0.05 to 0.2 m/s, but more variable than the transport as the flow may be locally "squeezed" between depth contours.

More recent information about the ESC mean currents near the Celtic Sea (Pingree and le Cann, 1989; Pingree et al., 1999; Huthnance et al., 2001) indicates some evidence of seasonality with weaker flow in spring and stronger flow in autumn but does not change the overall transport estimate given by Huthnance (1986).

Holliday et al. (2000) and Holliday (2003) have calculated the mean transport though the Rockall Trough as 3.7 Sv, but the flow fluctuates on interannual timescales. There was unusually strong northward transport in the Trough during 1988/89 and 1998, peaking at 7.9 Sv in 1989 and 7.5 Sv in 1998.

Based on detailed year-round measurements during 1995-1996, Souza et al. (2001) found that the ESC at the latitude of the Malin Shelf (~56°N) west of the Hebrides had a maximum mean flow of ~ 0.15 m/s, with greater flow variability in winter. In summer there was a maximum flow at about 200m depth whereas in winter the flow was more nearly uniform in depth. The fastest mean flow was in 500m depth or more, but in winter the mean flow was broader and extended onto the shelf. A mean transport of about 2 Sv is suggested by combining these measurements with tracked drogues (Huthnance, personal communication).

At the Wyville-Thomsen Ridge (near 60°N with typical depth 400-500 m) there is a complex exchange of flow. Some of the deeper slope current from the Hebrides slope is probably diverted by the Ridge to the north-west. However, the upper-slope current continues to the west Shetland slope (the Faroe Shetland Channel). Here it is joined by a broader flow of warm, saline North Atlantic water across the Ridge from the Rockall Trough. Further on, it is also joined by water that has circulated clockwise around the Faroe Islands to the Faroes side of the Faroe Shetland Channel (ICES 2003a). These additions result in an increased transport along the west Shetland slope, on average about 4.5 Sv with a spring minimum and autumn maximum (Huthnance, personal communication).

The concentrated flow at the shelf edge and the effective separation of the shelf and oceanic regimes by the topographically steered flow is illustrated by the behaviour of drifting floats. Released into the narrow slope current, floats have a strong tendency to remain in it and move rapidly along the slope, in contrast to those released on the shelf or in the oceanic regime, which show much more variable behaviour unless they are entrained into the slope current (Simpson, 1998).

Click here to see an animation of the along-slope current, as measured by drifting floats (STEMgis).

4. Circulation in UK Waters

As discussed in section 1.2, circulation is variable in time and space and therefore it is difficult to describe any generally persistent circulation patterns in UK waters. There are only a few regions where the long-term circulation has been convincingly measured (usually from the distribution of tracers) e.g. the north-easterly flow of the North Atlantic to the west of Ireland and Scotland, some aspects of flow in the North Sea, the north-easterly flow from the Dover Straits into the North Sea and the mean flow northwards through the Irish Sea. For this reason authors are reluctant to produce over-simplified maps of general circulation patterns.

The models discussed in section 2.1 show some consistencies in the pattern of long-term climatological circulation of the North Sea and English Channel, in broad agreement with those inferred from observations, (but not in the Irish Sea).

Click here to see the results from the POL model.

4.1 North Sea

The dominant motion in the western and southern parts of the North Sea is tidal, whereas the wind is the dominant source of energy in the northern and eastern parts (Rodhe, 1998). The tides enter the North Sea from the Atlantic Ocean north of Scotland and sweep around it in an anticlockwise direction. Surges travel anti-clockwise: southwards along the eastern UK coast and then northeastwards along the coast of continental Europe. Tidally generated residual currents are generally small compared with density-driven currents and wind-driven currents, but are responsible for a significant part of the residual currents in the western and southern parts. The wind-driven currents are induced by mostly south-westerly and westerly winds, but easterly winds, which occur mostly in spring and summer, can reverse the broadly anti-clockwise circulation.

Most of the central and northern North Sea becomes thermally stratified during April/May, due to increasing solar heat input; with a well-mixed layer of about 30 to 40m deep (Howarth, 2001). In autumn, heat loss at the surface leads to the surface mixed layer deepening and cooling until the bottom is reached in October/December. The tidal energy in the southern and western regions is strong enough to keep the water column well mixed most of the year, but some coastal regions stratify because of freshwater river discharge, with the fresher water tending to form a thin surface layer about 30km wide which stays close to the coast (Howarth, 2001).

The fronts in the northeast of the North Sea (outside UK waters) are related to the low-saline water in the Norwegian Coastal Current. The main front in the central North Sea separating the thermally stratified water in summer to the north from the well-mixed water from the south starts from Flamborough Head, bifurcates around the Dogger Bank and passes to the north of the Frisian Islands (Howarth, 2001). Some of the fronts in the southern North Sea are related to freshwater outflow from rivers, but most are tidal fronts (Rodhe, 1998).

Click here to see a schematic diagram of frontal zones and stratification of the North Sea (Figure 5.11).
Link to http://www.offshore-sea.org.uk/sea/dev/html_file/sea2_consult.cgi?sectionID=43

A major contribution to the seasonal circulation of the central North Sea is the existence of a persistent and narrow (10 to 15 km) near-surface flow extending continuously for ~ 500 km along the 40 m contour between the Firth of Forth and the Dogger Bank, associated with strong bottom fronts bounding a pool of cold, dense bottom water isolated below the seasonal thermocline (Brown et al., 1999).

The overall pattern of the mean circulation in the North Sea is broadly anti-clockwise around the coasts, with weak and varied circulation in the centre. The mean coastal flow is southward past Scotland and England and into the Southern Bight, where there are inputs of salty water through the Dover Straits and of fresh water down the main rivers, and on into the German Bight, flowing northward past Denmark in the Jutland current to join the Norwegian Coastal Current in the Skagerrak (Rodhe, 1998; Howarth, 2001).

There are major inflows in excess of 1 Sv of water of Atlantic origin across the northern boundary but very little penetrates far into the North Sea. The larger portion flows along the western slope of the Norwegian Trench and recirculates in the Skagerrak, flowing out along the eastern side of the Trench underneath the Norwegian Coastal Current (NCC).
A smaller inflow of mixed Atlantic and shelf water (including some from the Scottish Coastal Current, see section 4.4) flows in east of Shetland and between Shetland and the Orkney Islands. However, most of the flow is guided eastwards to the trench by the topography along the 100m-depth contour, and only a small part flows southward along the coast of Scotland and England. Less than 10% of the inflow to the North Sea enters through the English Channel. The only major outflow from the North Sea is along the eastern side of the Norwegian Trench and is approximately 1.3 to 1.8 Sv. The bulk of the transport in the circulation is concentrated in the northern part of the North Sea and in the region of the Norwegian Trench, with the main outflow along the Norwegian coast in the NCC (Howarth, 2001).

The flushing time, for the complete renewal of the water, is about one to three years (Simpson, 1998).

Click here to see a schematic diagram of the general circulation in the northern North Sea (Figure 5.10).
Link to http://www.offshore-sea.org.uk/sea/dev/html_file/sea2_consult.cgi?sectionID=43

Click here to see a tabulation of mean transport (Sv) across sections in the North Sea over 1987-1993 from three numerical models. Courtesy of Jane Williams, POL

Holliday et al. (2001) conclude that two pulses of oceanic inflow into the North Sea in 1988 and 1998 coincided with unusually strong northward transport of anomalously warm water at the edge of the continental shelf through Rockall Trough (see section 3.2). However they point out that factors other than the strength of the shelf edge current may be important for timing of inflow events, including the influence of local wind-driven advection. For example, they report that while high flows were measured in the Norwegian shelf edge current in 1996 (Mork and Blindheim, 2000), the inflow to the North Sea in that year was low and southerly warm-water plankton did not penetrate into the basin. This reduction in flow is thought to be a consequence of the pronounced reversal of the NAO and its effect on local winds in the winter of 1995/96. In contrast, they point out that in the winter of 1997/98, when the NAO was positive, the warm waters of the shelf edge again contributed southerly oceanic plankton to the North Sea (Reid et al., 1998).

4.2 English Channel and Celtic Sea, including the Bristol Channel

The residual flow along the English Channel is from west to east, driven by non-linear tides (due to strong tidal forcing from the Atlantic), predominantly south-westerly prevailing winds and density currents (primarily due to freshwater discharge from the rivers draining the south coast of the UK and the continental coast - most of the regions of strong tidal flow are continuously mixed). Prandle et al. (1993) estimated the net flux north-eastward through the Dover Straits as 0.11 Sv, but subsequent measurements by Prandle and Player (1993) revealed a complex flow pattern including the existence of an anticlockwise gyre off Cap Gris Nez; thus emphasising the difficulty in quantifying the long-term net flow.

Analysis by Pingree and le Cann (1989) of an extensive compilation of current meter data and observations of the distribution of the radionuclide Caesium 137 released from Sellafield and Cap de la Hague show a generally weak mean circulation in the Celtic Sea.

During winter (November to April) the Celtic Sea is vertically mixed and residual circulation is largely controlled by wind forcing. In summer, most of the Celtic Sea experiences strong thermal stratification, occurring where tidally-generated turbulent energy is insufficient to mix the increased surface heat input from solar heating throughout the water column. The summer seasonal circulation is dominated by strong anti-clockwise jets associated with bottom fronts bounding a cold saline pool (Brown et al., 2003) – the northward flowing jet on the eastern side of St George’s Channel transports water rapidly from the mouth of the Bristol Channel towards the Irish Sea.

There is an overall weak eastward residual flow in the Bristol Channel and the estimated flushing time is from 150 to 300 days. Prevailing south-westerly winds drive a flow northward along the Cornish coast and density gradients also contribute to the weak circulation, with depth-averaged flow into the channel in deeper water, and return down-channel flow in shallower waters. However, during periods of high freshwater input these flows are significantly enhanced, although no direct measurements have been made. There is a complex residual circulation within the Bristol Channel comprising of a series of closed eddies, arising primarily as water flows past headlands, bays and islands; but they contribute little to the overall mean circulation (Defra, 2000).

Along the northern coast of the Bristol Channel, between Carmarthen Bay and Nash Point, flow is also eastward. However as water is piled up into the channel an adverse pressure gradient is created, and this drives a depth-mean flow westwards along the central axis of the channel. This flow is then steered northward around St David’s Head and into the Irish Sea. There is also an indication of a large-scale, but weak, anti-clockwise recirculation at the mouth of the channel, the northward-flowing arm of which causes flow across the mouth of the channel at about 5°W (Defra, 2000).

4.3 Irish Sea

Both surge and density driven currents contribute significantly to the overall long-term mean circulation of the Irish Sea. The latter are particularly important in the eastern Irish Sea where the differences between the saline oceanic inflows and freshwater input from the Rivers Dee, Mersey, Lune and Ribble cause horizontal and vertical density changes in Liverpool Bay. These flows are strongest in winter and spring but can be overwhelmed during periods of strong winds (Defra, 2000).

The distribution of Caesium 137 discharged from Sellafield has been used to infer the mean surface water circulation in the Irish Sea (Jefferies and Steel, 1989; Irish, 2003). The main input of water is from the Atlantic, flowing south to north through St. George’s Channel. The general shape of the isopleths suggests that the main flow veers towards the Welsh coast as it moves north, with a weaker flow, generally northward, to the west of the Isle of Man. A minor component of the flow enters the eastern Irish Sea to the north of Anglesey and moves anti-clockwise round the Isle of Man before rejoining the main flow to exit through the North Channel (Defra, 2000). The flushing time is more than one year (Simpson, 1998).

Most regions of the Irish Sea are continuously mixed, because of the strong tidal currents. However a deep basin region in the western Irish Sea (centred at 53° 40’N, 5°W) and part of Cardigan Bay experience strong seasonal stratification in the summer and are separated from the well-mixed areas by tidal mixing fronts. In the former, a dome shaped pool of cold water sits below the thermocline and is separated from surrounding waters by strong temperature fronts. These fronts drive strong narrow (~10 km) jets that dominate the circulation in the region during summer months, forming a closed-circulation that acts to retain material in the region (Hill et al., 1997) but which does not contribute substantially to the net circulation of water. Following the breakdown of stratification in autumn, the mean flow is then weakly northward until the following spring.

There is evidence that there has been a considerable change in flow conditions in the Irish Sea during the last thirty years. McKay and Baxter (1985) and Jefferies and Steele (1989) found that they could only obtain a reasonable fit between observations of the concentration of Caesium 137 and model predictions by changing the circulation pattern and strength in their numerical models. The former found that the coastal flow conditions from the northeast Irish Sea to western Scottish coastal waters had changed considerably since 1977, with a further change in Irish Sea outflow during 1981. The latter had to infer a factor of two change in the Irish Sea circulation in the mid 1970s, having to double the flow rate out of the North Channel from the end of September 1976. Also they inferred that a change in flow took place in the 1980-81 period, after which the flow pattern returned to that of 1977-1980.

A direct link with the circulation of the Irish Sea and the NAO has not been established but it is reasonable to expect a degree of correlation. For example, a positive Index results in a higher frequency of Atlantic storms, the centres of which track to the north of Britain and so promote northerly and westerly winds over the Irish Sea region. There will then be an increased incidence of storm surges in the eastern Irish Sea and Liverpool Bay, enhancing the contribution of surge currents to the overall circulation. Also, it is conceivable that changes in storm tracks may regulate the circulation and flushing of the region (Defra, 2000).

4.4 Minches, west Scotland and Scottish continental shelf

The mean flow through the North Channel is northwards, occurring as a series of pulses in response to the effects of the wind. However, overall outward flow is strongest on the eastern (Scottish) side of the channel, with a weaker surface return flow along the Irish coast, see Figure 3. Knight and Howarth (1999) give an estimate of the flow through the North Channel of 0.077 Sv, based on one year’s measurements from July 1993 to July 1994.

There is considerable variability in the vertical structure of the flows through the North Channel. Figure 4 shows the relative magnitude and direction of mean flow over a 15-month period. The near-surface mean flow was directed towards the Irish Sea, depth-averaged mean flow was directed across the channel towards the Scottish coast and near-bottom flow was directed towards the Malin Shelf. However episodes of residual flow can be seen in the top two vector diagrams often in opposition to the direction of mean flow with larger temporal variations at the near-surface, while at the near-bottom the flow was more stable and showed less time variability. Strong winds from the southeast between 1st and 28th February caused the largest reversals of near-surface flow from the direction of near-surface mean flow (Knight and Howarth, 1999).

The salinity deficit of the water flowing out of the Irish Sea through the North Channel is enhanced further by the substantial discharge of freshwater from the Clyde Sea and other sources along the Scottish coast, including the Firth of Lorne. Under the influence of the earth’s rotation, this low-density water moves northward in a well-defined Scottish Coastal Current (SCC) (Simpson, 1998).

The flow of the SCC in the Tiree Passage has been measured since 1981 (Inall and Griffiths, 2003). The along-channel residual flow shows a seasonal variation and has a mean value of 10.8 cm/s directed towards the north, with very few, short duration periods of flow reversal. The mean volume flow through the passage is calculated as 0.067 Sv, a similar figure to that for the North Channel outflow, although Inall and Griffiths state that this is clearly not the same water. (Caesium 137 studies (McKay et al., 1986) indicate the average dilution ratio of North Channel water to Atlantic water in the Tiree Passage to be approximately 3:1.)

At the entrance to the Minch, the Scottish Coastal Current divides, with one branch flowing through the channel between the Outer Hebrides and the Scottish mainland and the other turning south and the west around Barra Head before flowing northward up the west coast of the Hebrides. The flow is forced to a significant degree by wind stress, with the pulsed nature of the flow associated with the passage of depressions to the north of the British Isles (Simpson, 1998).

5. References

Brown, J., Hill, A.E., Fernand, L., Horsburgh, K.J. (1999). Observations of a seasonal jet-like circulation at the central North Sea cold pool margin. Estuarine Coastal and Shelf Science, 48: 343-355.

Brown, J., Carillo, L., Fernand, L., Horsburgh, K.J., Hill, A.E., Young, E.F. (2003) Observations of the physical structure and seasonal jet-like circulation of the Celtic Sea and St. George’s Channel of the Irish Sea. Continental Shelf Research, 23: 533-561.

Bryden, H.L. and Imawaki, S. (2001) Ocean heat transport in: Siedler, G., Church, J. and Gould, J. eds. Ocean Circulation and Climate: Observing and Modelling the Global Ocean. San Diego CA, Academic Press, pages 455-474.

Defra (2000). Quality Status Report of the Marine and Coastal Areas of the Irish Sea and Bristol Channel. Retrieved 14th September from the World Wide Web: http://www.defra.gov.uk/environment/marine/quality/03.htm

Delhez, É.J.M., P. Damm, E. de Goede, J.M. de Kok, F. Dumas, H. Gerritsen, J.E. Jones, J. Ozer, T. Pohlmann, P.S. Rasch, M. Skogen and R. Proctor. (2004). Variability of shelf-seas hydrodynamic models: lessons from the NOMADS2 Project. In press, Journal of Marine Systems.

Dickson R.R., J. Meincke, I. Vassie, J. Jungclaus and S. Østerhus (1999). Possible predictability in overflow from the Denmark Strait. Nature, 397: 243-246.

Dickson R.R., I. Yashayaev, J. Meincke, W.R. Turrell, S.R. Dye and J. Holfort (2002). Rapid Freshening of the Deep North Atlantic over the past Four Decades. Nature, 416: 832-837.

Fox, C.J. and J.A. Aldridge (2000). Hydrographic circulation and the dispersal of herring larvae in the Blackwater estuary. Journal of the Marine Biological Association of the UK, 80(5): 921-928.

Hansen, B., W. R. Turrell and S. Østerhus (2001). Decreasing overflow from the Nordic seas into the Atlantic Ocean through the Faroe Bank channel since 1950. Nature, 411: 927 – 930.

Hill, A.E., J. Brown, and L. Fernand (1997). The summer gyre in the western Irish Sea: shelf sea paradigms and management implications. Estuarine Shelf and Coastal Sciences, 44: 83 - 95.

Holliday, N.P. and P.C. Reid (2001). Is there a connection between high transport of water through the Rockall Trough and ecological changes in the North Sea? ICES Journal of Marine Science, 58: 270-274.

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.

Howarth, M.J. (2001). North Sea Circulation. pp 1912 – 1921 in Encyclopaedia of Ocean Science volume IV, (Steele, J.H. Editor in Chief.) Academic Press.

Huthnance, J.M. (1986). The Rockall slope current and shelf-edge processes. Proceedings of the Royal Society of Edinburgh, 88B: 83 - 101.

Huthnance, J.M., H. Coelho, C.R. Griffiths, P.J. Knight, A.P. Rees, B. Sinha, A. Vangriesheim, M. White and P.G. Chatwin (2001). Physical structures, advection and mixing in the region of Goban Spur. Deep-Sea Research II, 48: 2979-3021.

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

Inall, M. and C. Griffiths (2003). The Tiree Passage Time Series: 1981 – 2003.
Retrieved 11th January 2004 from the World Wide Web: Retrieved 10th September 2003 from the World Wide Web: http://www.sams.ac.uk/dml/projects/physics/tiree/tiree.pdf

Irish, R. (2003). Monitoring of the quality of the marine environment, 1999-2000. Scientific Series, Aquatic Environment Monitoring Report, CEFAS, Lowestoft, 54, 98pp.

Jefferies, D.F. and A.K. Steele (1989). Observed and Predicted Concentrations of Caesium-137 in Seawater of the Irish Sea 1970-1985. Journal of Environmental Radioactivity, 10: 173 – 189.

Jones, J.E. (2002) Coastal and Shelf–Sea modelling in the European context. Pp 37-141 in Oceanography and Marine Biology: an Annual Review 2002 (Gibson, R.N., M. Barnes and R.J.A. Atkinson, Editors). Taylor & Francis.

Knight, P.J. and M.J. Howarth (1999). Flow through the North Channel of the Irish Sea. Continental Shelf Research, 19: 693-716.

Levitus, S., J.I. Antonov, T.P. Boyer and S. Stephens (2000). Warming of the World Ocean. Science, 287: 2225-2229.

McKay, W.A. and M.S. Baxter (1985). Water transport from the north-east Irish Sea to western Scottish coastal waters. Estuarine and Coastal Shelf Science, 21: 471-480.

McKay, W.A., J.M. Baxter, D.J. Ellett, and D.T. Meldrum (1986). Radiocaesium and Circulation Patterns West of Scotland, Journal of Environmental Radioactivity, 4: 205-232.

Mork, K. A. and J. Blindheim (2000). Variations in the Atlantic inflow to the Nordic Seas, 1955-1996. Deep Sea Research, 47: 1035-1057.

Pingree, R.D., and B. Le Cann (1989). Celtic and Armorican shelf and slope residual currents. Progress in Oceanography, 23: 303-338.

Pingree, R.D., B. Sinha and C.R. Griffiths (1999). Seasonality of the European slope current (Goban Spur) and ocean margin exchange. Continental Shelf Research, 19: 929-975.

Pingree, R.D. (2002). Ocean structure and climate (Eastern North Atlantic): in situ measurement and remote sensing (altimeter). Journal of the Marine Biological Association of the United Kingdom, 82: 681-705.

Prandle, D., C. F. Jago, S. E. Jones, D. A. Purdie and A. Tappin (1993). The influence of horizontal circulation on the supply and distribution of tracers. Philosophical Transactions of the Royal Society of London, A, 343: 405 - 421.

Prandle, D. and R. J. Player (1993). Residual currents through the Dover Straits measured by H.F. radar. Estuarine, Coastal and Shelf Science, 37: 635 – 653.

Reid, P. C., Edwards, M., Hunt, H. G., & Warner, A. J. (1998). Phytoplankton change in the North Atlantic. Nature 391: 546.

Rodhe, J. (1998). The Baltic and North Seas. Pp 699 – 732 in The Sea, Volume 11, edited by A. R. Robison and K.H. Brink, John Wiley & Sons. Inc.

Simpson, J.H. (1998). The Celtic Seas. Pp 659-582 in The Sea, Volume 11, edited by A.R. Robinson and K.H. Brink, John Wiley & Sons, Inc.

Smith, J.A., P.E. Damm, M.D. Skogen, R.A. Flather and J. Pätsch (1996). An investigation into the variability of circulation and transport on the north-west European shelf using three hydrodynamic models. Deutsche Hydrographische Zeitschrift: 48, 325 – 348.

Souza, A.J., J.H. Simpson, M. Harikrishnan and J. Malarkey (2001). Flow structure and seasonality in the Hebridean slope current. Oceanologica Acta, 24: S63-S76.