Industrial "Hoover" Fishing
	A Policy Vacuum
	



	Q Greenpeace International
	Keizergracht 176
	1016 DW Amsterdam
	The Netherlands
	January 1997


ISBN 1 871532 09 4

Written and researched
by Phil Aikman (consultant to Greenpeace)
	

Contents


	The Summary					05
1.	The causes and consequences of industrial fishing	
	1.1	Introduction				09
	1.2	International overview			10
	1.3	Poor data, poor management		11
	1.4	The consequence of going hungry		12
	1.5	Lessons learned?				14

2.	Products based on industrial fisheries	
	2.1	Introduction				16
	2.2	The fish meal and fish oil industry		17
	2.3	European markets			18
	2.4	Fish oil in margarines			19
	2.5	Fish meal in animal feeds			20
	2.6	Fish farming				21
	2.7	Retailers precautionary approach		23
	2.8	Future markets for industrial fisheries	24

3. 	Crisis in the North Sea	
	3.1	Overview				26
	3.2	The significant sandeel			28
	3.3	The impact of industrial fisheries		31
	3.4	Interactions with wildlife			44
	3.5	Closed areas				52
	3.6	The science vacuum			56
	3.7	The political vacuum			62

4.	Conclusions	
	4.1	Overview				66
	4.2	Recommendations			67

	Explanation of abbreviations and terms		68
	Bibliography					69

	
The Summary

The Problem

1.	Almost a third of the annual world marine catch comes from industrial 
fisheries. Since the Second World War, landings from the North Sea have dramatically 
changed from being totally for human consumption to a situation where half the catch is 
landed for industrial raw materials, fish meal and fish oil (pages 10,26). 

2.	There is a clear threat of serious or irreversible damage to fish stocks, seabirds 
and marine mammals caused by the North Sea industrial fisheries if they remain largely 
unregulated. Over the last 30 to 40 years industrial fishing has doubled the yield from 
the North Sea, largely through the development of the sandeel fishery. At first the 
fishery was local, but it soon spread to become the largest, landing a third of all fish. In 
recent years the sandeel fishery has further shifted into areas close to seabird 
populations despite numerous warnings, including work published by the Danish 
Fishing Ministry. The fishery has known no boundaries and has so far escaped any 
catch limits (pages 26, 47).

3.	Two decades ago the industrial fisheries became a text book example of the 
need for effective conservation and management policies. Since then the practices that 
caused the collapse of Peruvian anchovy in the 1970s have been repeated in other 
industrial fisheries around the world. In each case, short-term economic and political 
considerations dominated decisions regarding overfishing of the stocks, to the point of 
commercial collapse. Biological and related environmental data about the relevant 
resource was inadequate or, if data was available, fishery managers ignored it (pages 11-
15).

4.	Fisheries management has not been effective in controlling fishing effort or the 
industrial fishing fleets. Fisheries scientists, and the European Commission, actually 
predict that the long-term profitability of the fishing industry would increase if the 
fishing effort were halved in most fisheries (pages 56-65).

5.	On some occasions, EU fisheries scientists and managers believe that they can 
predict and effectively manage food stocks with no adverse effects on predators. On 
other occasions, they contradict themselves by saying that they don"t have enough 
information to accurately predict. In the case of sandeels, fisheries managers themselves 
admit they do not have the resources to collect information on the scale required to 
assess these stocks (pages 56-58).


The Evidence

6.	A report published by the International Council for Exploration of the Seas 
(ICES) suggests that, with the exception of Norway pout, the amount of industrial fish 
species taken by fishermen and predatory fish appears to leave little for seabirds and 
marine mammals (pages 32-33).

7.	Sandeels are shown in various studies to be a significant food source for the 
survival of North Sea cod. A recent Unilever funded report estimated that up to 60% of 
the diet of species fished for human consumption consists of the industrial fish species, 
sandeels and Norway pout (pages 35-37).

8.	The overfishing of sandeel stocks on a local scale may represent the greatest 
threat to seabirds in the North Sea, especially in the breeding season when seabirds 
forage close to their colonies (pages 44-48). 

9.	Studies into the diet of harbour porpoises, grey seals and common dolphins in 
Scottish waters have shown that they mainly feed on sandeels during the summer. The 
striped dolphin is known to feed on Norway pout (pages 48-52).

10.	The problem of bycatch in the industrial fisheries is clearly seen in the Norway 
pout and sprat fisheries. The Commission admit that resolving this problem has proven 
to be unmanageable and only partial solutions have been found (pages 38-44). 

11.	The distribution of Norway pout extends into the same areas of small haddock 
and small whiting. The Norway pout industrial fishery now accounts for over 80% of 
the mortality of immature (less than one-year old) haddock and whiting caught as 
bycatch (pages 40-42).

12.	In the case of the sprat industrial fishery, approximately 6.9 billion immature 
herring were landed in this fishery in 1995. Despite herring being banned as an 
industrial catch in the 1970s, substantial quantities of herring, almost equal to that 
landed by the human consumption fishery, will have been legally landed as bycatch in 
1996 mainly in the industrial sprat fishery (pages 38-44).


The Solutions

13.	The Maastricht Treaty requires that all EU fisheries policies, including the 
Common Fisheries Policy, must comply with the precautionary principle (pages 61-62).

14.	In contrast to the EU, Norwegian fisheries managers have adopted 
precautionary measures by closing the industrial fisheries for capelin recognising its 
vital importance as food for cod. One prominent Norwegian scientist has criticised EU 
scientists for not advocating similar policies on North Sea fisheries to politicians and 
fishermen (pages 14, 15, 56).

15.	European fisheries scientists (ICES) have recently advised the Commission that 
a 'precautionary' Total Allowable Catch for sandeels of 1,100,000 tonnes might be 
established which would be greater than any recorded catch. ICES advise that a lower 
limit would require more data, in direct contradiction of the precautionary principle 
(pages 56-58). 

16.	The precautionary principle applied to the industrial fisheries, including those 
for industrial purposes, requires that the less robust the data, the more stringent the 
controls. The principle would also require the protection of important nursery areas for 
human consumption fish, and prevent the capture of juveniles by small-mesh trawls 
used by the industrial fisheries. A precautionary approach to the industrial fisheries 
requires giving the benefit of the doubt to the marine ecosystem and the more important 
human consumption fisheries industry.
(pages 61-67).


Recommendations

Sandeel fishery

Recommendation 1: Full closure of all North Sea sandeel fisheries in areas sensitive for 
wildlife. These include areas where sandeel fisheries extend onto critical wildlife 
feeding grounds, such as those foraged by seabirds (pages 66-67).

Recommendation 2: Full closure of sandeel fisheries which extend into spawning, 
nursery grounds and important feeding grounds for human consumption fish (pages 66-
67). 

Recommendation 3: Elsewhere, catch limits should be regulated according to a strictly 
precautionary approach (pages 66-67).


Sprat fishery

Recommendation 4: A protected area closed to the industrial sprat fishery is needed to 
protect the herring nursery grounds in the North Sea, Skagerrak and Kattegat (pages 66-
67). 


Norway pout fishery

Recommendation 5: Stringent bycatch limits must be introduced to effectively 
discourage the
industrial fishing for pout in areas where there is likely to be a high bycatch mortality on 
other fish, such as cod, haddock, herring, saithe and whiting (pages 66-67). 













Third Fisherman:	Master, I marvel how the fishes
		live in the Sea.

First Fisherman:	Why as men do a-land,
		the great ones eat up
		the little ones.


		William Shakespeare - Pericles.










The causes and consequences of
industrial fishing


1.1	Introduction

Since the Second World War the development of intensified systems of livestock 
farming which require large supplies of cheap protein has led to the development of 
some very large- scale fisheries directed at species specifically for reduction to fish meal 
and oil. In other words, these fish are used as an industrial raw material, not eaten as 
table fish. These fisheries are generally known as industrial fisheries (Coull 1993; ICES 
1993b).

The world"s industrial fisheries first developed along the coasts of North America, 
South America and Africa, where strong upwellings bring to the surface of the sea cold, 
nutrient-rich water. This upwelling allows the proliferation of plant and animal plankton 
which supports large stocks of small pelagic species. These species include various 
types of anchovies, sardines and pilchards, which industrial fisheries target 
(Commission 1992a).

Industrial fisheries have also developed in areas such as the east coast of Canada, 
Iceland, the seas between Iceland and Norway, the Norwegian coast and the North Sea. 
Various target species are exploited in these fisheries; some are predominantly bottom-
dwelling (Norway pout), others are semi-demersal/semi-pelagic (sandeels, blue whiting, 
capelin) while others are totally pelagic (sprat, herring) (Commission 1992a). 

Sophisticated electronic fish-finding devices now allow more successful tracking and 
netting of fish schools. While powerful fish pumps, 'hoovers', speed up operations by 
allowing the direct pumping of the fish from the net directly into the fish holds and, 
when unloading, from the fish hold directly into the conveying system of fish meal 
factories (Commission 1992a).

Although poorly equipped vessels may also participate in some smaller-scale fisheries, 
the general perception of these industrial fisheries is of large, well-equipped vessels 
returning overloaded with fish, which are discharged into the storage tanks of a highly 
mechanised industry (Commission 1992a).

An issue that is contentious in a number of areas is how far catches should be used for 
the low-value outlet of reduction to meal and oil, rather than human consumption. 
While some species, like the capelin, sandeels and Norway pout are not thought fit for 
human consumption, these species are part of the diet of more valuable species in the 
food web, and in some cases more valuable species may constitute a high proportion of 
the bycatch (Coull 1993). While Danish fish landings for human consumption 
represented only 24% of the weight of the country"s total landings they generated œ250 
million (73%) in value. The industrial fisheries, on the other hand,  made up over 76% 
of the weight but only accounted for about œ95 million (27%) of the value (WorldFish 
1996b). The average weight of bycatch of human consumption species in the North Sea 
industrial fisheries would be worth in the region of œ62 to œ75 million, assuming 
inflexible market price (MAFF 1994b; ICES 1995b; WorldFish 1996b).


1.2	International overview

The world"s marine catch has increased more than five fold in the past 50 years Ñ from 
20 million tonnes in 1947 to 106 million tonnes in 1994. The development of industrial 
fisheries is responsible for around 40% of this increase (Coull 1993; Fishing News Int. 
1996b) and almost a third of the annual world marine catch comes from industrial 
fisheries. This produces around seven million tonnes of fish meal and one million 
tonnes of fish oil (Nicolson 1996; Fishing News Int. 1996a, 1996b). Over half of world 
production of fish meal and oil is generated in South America (Fishing News Int. 
1996a).

50% of the fish meal produced is sold to poultry producers and 25% to pig farmers with 
aquaculture taking 15% of sales, but this sector of the market is set to grow fast. 70% of 
fish oil production now goes to margarine manufacturers in Western Europe and 
industrial uses, paints, varnishes, etc, account for a further 5 % (Fishing News Int. 
1996a).

Traditionally, fish meal and oil are used as an alternative to vegetable-based sources, 
such as soya meal and oil, mainly because of their relative cheapness. Some producers 
and companies are now moving away from products derived from industrial fisheries, 
substituting vegetable protein for fish based protein, and vegetable oil for fish oil 
(Commission 1992a; COWI 1994; Brown 1996; BBC 1996).
However, markets for fish meal and oil in countries like China are rapidly expanding. 
Sergei Garcia of the UN Food and Agriculture Organisation (FAO) warns that 
regulation of the North Sea industrial fisheries must be introduced before these new 
markets pressurise the stocks to the point of collapse, as has happened in the past with 
other industrial fisheries around the world (Garcia 1996). The Peruvian anchovy being 
the best example of how increasing market demand can drive a stock to collapse.


1.3	Poor data, poor management

Of all the world"s fisheries the experience of the Peruvian anchovy fishery is the most 
salutary reminder of the need for effective conservation policies and management 
(Coull 1974).

Before 1950, anchovies in Peru were harvested mainly for human consumption, and the 
fishery was small. In 1950, the total catch in Peru was 86,723 tonnes. In 1953, the first 
anchovy fish meal processing plants were built in Peru, a speculative venture designed 
to make use of a large and relatively untapped resource to produce meal and oil for 
export (Muck 1989).

Over the next several years, anchovy fishing rose in response to growing demand from 
the plants and more fish meal plants were constructed. During the seven-month fishing 
season in 1969-70 the 1,700 strong fleet brought in 11 million tonnes of anchovies 
(Muck 1989; Thompson 1981). Peru became the number one fishing nation in the world 
by volume, with anchovies making up roughly 94% of its total catch (Muck 1989) and 
accounted for one fifth of the world"s fish catch. At one point it supplied over half the 
fish meal entering international trade (Lean et al. 1990; Coull 1993). 

A group representing FAO and the Peruvian Government"s Ocean Institute (Instituto 
del Mar del Peru) estimated the maximum sustainable yield at 9.5 million tonnes per 
year which included nearly two million tonnes estimated to be consumed by sea birds. 
In 1970, the group issued a warning: if the catch remained above 9.5 million tonnes, the 
fishery would be in danger of imminent collapse (Instito del Mar 1970).

The Peruvian Government turned a deaf ear. Fish meal prices were high in 1969 
because overfishing had recently led to a collapse of herring fisheries in Norway and 
Iceland. Peruvian officials, hoping to earn more hard currency through the export of fish 
meal, were also unwilling to risk declining employment in the industry. In 1970, a 
harvest of 12.4 million tonnes was allowed, followed by 10.5 million tonnes in 1971 
(Paulik 1981). Yet by 1973 it had fallen to two million tonnes and by 1983 to a mere 
100,000 tonnes, as overfishing compounded the effect of changes in the oceanic 
currents known as El Ni-o (Lean et al. 1990). 

Today, the misconception persists that the El Ni-o was responsible for the demise of 
Peru"s anchovy fishery (Evans 1989). El Ni-o, which occurs at irregular intervals every 
three to seven years is characterised by the appearance of warm surface water off Peru 
and Ecuador, resulting in fewer nutrients rising to the ocean"s surface. This reduction in 
phytoplankton causes decline of all the other elements along the food chain: 
zooplankton, fish, birds and mammals (Paulik 1981). Most research, however, supports 
the idea that although El Ni-o contributed to the collapse, it was unrestricted fishing that 
placed the resource in inevitable jeopardy (Evans 1989).

The events that caused the collapse of the Peruvian anchovy fishery form a scenario that 
has been repeated often around the world (Glantz 1981). The California sardine fishery 
was the world"s first industrial fishery to take off and the first to collapse. The capital 
raised from this fishery was used to finance the Peruvian operations. Later examples of 
collapses include the North Sea herring and mackerel, the Namibian pilchard, the 
Norwegian capelin, Atlanto-Scandic herring and the Japanese anchovy, pilchard and 
sardine fisheries (Coull 1993; World Resources Institute 1994; Oil & Fats 1995; 
WorldFish 1996a; Globefish 1996). In each case, economic and political considerations 
have dominated decisions regarding targeted fish stocks. Either biological and related 
environmental data about the relevant resource was scarce, or, if data was available, 
fishery managers decided to disregard it (World Resources Institute 1994).


1.4	The consequence of going hungry

Studies have demonstrated that declines in prey available to seabirds were related to 
fishery exploitation (ICES 1996d), such as the *recruitment overfishing of the Peruvian 
anchovy when the Peruvian brown pelican, Guanay cormorant and the Peruvian booby 
populations declined (Nelson 1978; Duffy 1983). Repeated, almost total, breeding 
failures of the puffins at R¨st on the Lofoten Islands (Lid 1981; Barrett and Vader 1984; 
Furness and Monaghan 1987) that coincided with the reduction in Norwegian herring 
stocks (Hamre 1988). Breeding success was also reduced in kittiwakes, razorbills and 
guillemots (Lid 1981; Barrett and Vader 1984). The same thing may have occurred in 
Newfoundland puffin colonies following the overfishing of "industrial" capelin (Brown 
and Nettleship 1984). In the Pacific, breeding success in the southern Californian brown 
pelican population has been influenced by the abundance of their local food supply, the 
anchovy, which is also industrially fished (Anderson et al. 1982; Schaffner 1986; 
Springer et al. 1986).

*'Recruitment overfishing' occurs when intensive fishing progressively reduces the stock
 until it no longer produces enough new fish to regenerate itself each year. Given the 
inherent variability in the survival of small fish before they 'recruit' to the stocks, this 
type of overfishing is very difficult to diagnose. From a conservation point of view it is 
the most serious threat to seabirds because, if not corrected, it may result in stock 
collapse (OSPAR and ICES 1993).

A 31-year study by Coulson and Thomas (1985) showed that at a kittiwake colony, at 
North Shields on the north-east coast of England, the population declined when an 
important source of the birds" food, herring, became steadily less abundant outside the 
breeding season.

In the north-eastern Atlantic, the Norwegian spring spawning herring and the Barents 
Sea capelin (the two largest fish stocks) have been depleted by overfishing. The herring 
collapsed in the late 1960s (Hamre 1988) and fishing was banned for almost 25 years 
(IFOMA 1996b). The capelin was depleted in the middle of the 1980s. Crowds of 
underfed seals invaded Norwegian coastal waters and thousands of dead seabirds have 
drifted ashore on the north Norwegian coast. The dramatic events demonstrate that the 
industrial fishery for capelin in the area has thrown the normal food-web out of balance 
(Hamre 1988). 

In the late 1970s, total catches of fish in the Barents Sea reached 4.5 million tonnes. 
Then, due to a decreasing abundance of cod, catches dropped sharply in the early 1980s. 
In the second half of the 1980s, after the industrial fishery of capelin ended, the total 
catches were about 300,000 tonnes. Fishing was very heavy in the 1970s and the first 
half of the 1980s, despite a declining biomass (Kovtsova et al. 1991; Zilanov et al. 
1991). By the second half of the 1980s, herring, capelin, and polar cod -  the main food 
for cod and haddock - were very scarce. This led to slow growth, starvation, and 
increased mortality (Kovtsova et al. 1991).

The collapse of capelin stocks in the Atlantic has also been associated with changes in 
the distribution of humpback whales around Newfoundland (Whitehead 1987), with the 
drastic decline in the guillemot population around Bear Island in the Barents Sea 
(Meehlum and Bakken 1994), and reductions in puffin populations off south-east 
Newfoundland in the late 1970s (Brown et al. 1984).

1.5	Lessons learned?

Between 1972 and 1975, the Barent Sea capelin stocks increased to around 7 million 
tonnes. However, after 1975 there was a steady decline in the stocks until 1986-87 when 
they fell to 20,000 tonnes (ICES 1993c). The capelin industrial catch in 1983 was 
around one million tonnes, but is now zero. 

Norway has now severely limited the industrial fishery for capelin because it has been 
realised how vital this "industrial" species is to the Barent Sea cod stock. Capelin is an 
important food component in the Barent Sea ecosystem and there has been no industrial 
fishery for the Barent Sea capelin since 1993. The annual predation from cod on capelin 
is estimated to have varied between 0.6 and 3.7 million tonnes in the period 1989-1995. 
The latest management advice given by the International Council for the Exploration of 
the Seas (ICES) is that 'no fishing should take place on this stock in 1997' (Fishing 
News Int. 1996a; ICES 1996c).

Odd Naken (a prominent scientist at Norway"s Institute of Marine Research) was 
recently quoted in Fishing News International as stating (Fishing News Int. 1996a): 'We 
want to protect the smaller fish which cod eat, especially herring and capelin. Only if 
there is enough capelin for the cod will, perhaps fishing be allowed again.'

Growth of cod in the Barent Sea depends on availability of prey such as capelin, and 
variability in cod growth has had major impacts on the cod fishery. Cod are able to 
compensate only partially for low capelin abundance, by switching to other prey 
species. This may lead to periods of high cannibalism on young cod (ICES 1996c).

The UN Food and Agriculture Organisation (FAO) has recently recommended that the 
Pacific fishery for sardine and anchovy, exploited mainly by Peru, be cut by between 
10% and 20% and that a higher share of the catch be channelled toward human 
consumption . 90% of the catch is currently ground into fish meal (Fishing News Int. 
1996d). Peruvian catches have been reduced from 10.9 million tonnes in 1994, to 8.2 
million tonnes in 1995 (Globefish 1996). The new Peruvian fisheries minister, Ing 
Pandolti, has introduced a number of FAO recommendations, including tighter controls 
on fishing bans and quotas, and suspending the issue of new licences for fish meal 
plants and their fleets (WorldFish 1996c). 

But despite these shifts towards conservation of fish stocks, reports of fish stock 
collapses or declines further highlight the need to adopt stringent conservation measures 
throughout the world"s industrial fisheries:

'A shortage of supplies in recent months has caused fish meal prices to sky-rocket since 
October 1995.' World Fish Report, Feb 96.

'The crisis of the South African pilchard resource was reflected in a further reduction of 
South African fish meal production.' Globefish, Jan 96.
'The drastic decrease of fish oil production after 1990, due to the declining haul of 
sardine, is severely affecting consumption.' Oil & Fats International, 1995.

In the North Sea, industrial fishing has doubled its yield in the last 30 to 40 years 
(AFMFOM 1993) largely due to the expansion of the largely unregulated sandeel 
fisheries. Odd Naken, has also suggested that: 'There should be quotas in everything Ñ 
especially the [North Sea] sandeel where the Danish industrial fleet catches without any 
limits.' Industrial fisheries in European waters (especially for sandeels) are not subject 
to catch controls to the same extent as most of the human consumption fisheries (COWI 
1994), nor to the same extent as they are now regulated in Norwegian waters.
Products based on industrial fisheries
	

2.1	Introduction
	
Fish meal and fish oil are used to produce a range of products, including oil for 
margarines and solid cooking fats Ñ both for domestic consumption and for the 
commercial production of cakes and biscuits; meal and oil for aquaculture (fish farm) 
foods; and fish meal for the production of feed for livestock, particularly poultry and 
pigs. .

On a global scale, some 50 per cent of fish meal production is sold to poultry producers 
and 25 per cent to pig farmers, amounting to 17.5 million tonnes and 8.75 million 
tonnes respectively  of raw fish (Fishing News Int. 1996a). 

Consumption of fish makes a very significant contribution to nutrition to a large number 
of people worldwide (James 1995). The conversion of fish to fish meal and oil makes 
only an indirect contribution to human nutrition (Moen 1983). Fish used as feed 
produces far less nutrition for humans than would be obtained by eating the fish directly 
(Kent 1994). If only a relatively small proportion were diverted to direct human 
consumption, it could make a significant impact to world nutrition (James 1995). 

In 1990, the fish meal trade amounted to about 16 million tonnes of fish, of which 11 
million was from developing countries. Of this trade, some of the raw material that is 
used for making fish meal could be used by populations in developing countries for 
their own consumption (Kent 1994). The substantial profits to be made from raising 
pigs and poultry assures that it is the animals which win that competition. (Kent 1994). 
Moving the use of fish meal and oil away from pigs/poultry demands in more developed 
countries would assist in the amount of fish available to those countries at source.
At the World Food Summit in Rome, November 1996, a resolution was passed to halve 
the number of undernourished people in the world by 2015. Jorge Csirke, senior fishery 
resources officer at the UN Food and Agriculture Organisation (FAO) pointed out that 
his native Peru turns between eight and ten million tonnes of anchovies and sardines 
into fish meal a year. He points out that 'just diverting a small fraction of what is now 
being used for fish meal [and fish oil] into human consumption will make a huge 
difference.' (WorldFish 1996e).


2.2	The fish meal and fish oil industry
	
The largest producer of fish oil until recently has been Japan, though since the collapse 
of its pilchard industrial fishery, Peru and Chile now dominate the market. The largest 
producer of fish oil and meal in Europe, and the sixth largest in the world, is Denmark. 

In 1995 Denmark produced 106,000 tonnes of fish oil and 374,000 tonnes of fish meal 
from approximately 1.5 million tonnes of fish (FAO 1996). Ninety per cent of this raw 
material comes not from the offal or unsold fish of commercial (human consumption) 
fisheries but directly and predominantly from industrial fisheries of the North Sea 
(COWI 1994). Fish oil amounts to approximately 7% of the body weight of a fish 
(depending on the species), and fish meal to approximately 22% (COWI 1994).

With the total fish landings from the North Sea standing at 2.4 million tonnes for 1994, 
one processing plant in Denmark has the capacity to consume 30% of all biomass of fish 
removed from the North Sea. Esbjerg Fiskeindustri, also known as the "999" plant, 
promotes itself as 'The World"s Largest' fish meal and oil plant. This one plant 
processes about 750,000 tonnes of industrial fish annually and thus processes half of the 
total Danish production of fish meal and fish oil. 4,000 tonnes of raw materials can be 
processed every day (Esbjerg Fiskeindustri undated).

There are five fish meal plants in Denmark located along the west and north coast of 
Jutland. Esbjerg is the largest followed by Skagen, Thyboran, Hanstholm, Hvide Sande 
and Hanstholm. These plants produce around 300,000 tonnes of meal and 85,000 tonnes 
of fish oil. The Danish Government announced a subsidy of DK 13 million (œ1.4m) in 
1995 to refurbish a redundant fish meal plant, the "Hugo Mogeberg" (COWI 1994, 
Fishing News 1995a), with the intention of expanding the Danish industrial fisheries.

Until recently fish oil has been used for municipal power stations in Denmark but the 
Danish Government has not issued any new contacts for burning fish oil in power 
stations since 1990. The current contracts were honoured until the end of 1996. 

In 1992, Denmark exported 76% of fish meal production and 82% of fish oil: the rest 
was consumed domestically. The Far East (fish/prawn farming) and southern Europe are 
the biggest expanding markets for the Danish fish meal industry, while exports to 
Northern Europe have been declining (COWI 1994).

The UK is a relatively insignificant producer of these products, with average annual 
production of fish oil approximately 10,000 tonnes and meal approximately 50,000 
tonnes, most of which is largely derived from offal from the white-fish filleting industry 
(MAFF 1996a). However, on the Shetland Isles, the Bressey fish meal plant takes its 
raw material supplies from the regulated sandeel fishery around the Shetlands. This 
plant has however taken sandeels caught by Scottish fishermen prosecuting the Wee 
Bankie fishery (Shetland Times 1996).


2.3	European markets
	
The EU and Norway consume 49% of the world"s fish oil and 19% of the world"s fish 
meal (FAO 1993a; Trade Bulletin 1995). Britain imports both meal and oil from a 
number of countries including Peru, Iceland, Denmark, Norway and Chile.

About 80% of the EC"s overall production of fish meal is currently used as a feed 
supplement, mostly for poultry, cattle, piglets, pigs and fur-bearing animals (minks), the 
remaining 20% being used in aquaculture. 70% of the EC production of fish oil is used 
in the margarine industry, 25% in aquaculture, and the remaining 5% in the manufacture 
of paint and varnish (AFMFOM 1993).

The UK is the largest consumer of industrial fish products in Europe, accounting for 
around one-third of total EC consumption of fish meal and oil. In 1995 this accounted 
for 260,000 tonnes of meal and 128,000 tonnes of oil (MAFF 1996a). See Figure 4. 
Imports of fish meal and oil amount for around œ115 million per year, with a trade 
deficit of œ76 million for 1995 (Fishing News Int. 1996b).

The Norwegian company Norsk Hydro, through its subsidiary BioMar Ltd, is a 
significant buyer of fish meal and fish oil in the UK. BioMar recently started production 
of fish-feed at Grangemouth in Scotland. It is Europe"s second largest producer of fish-
food, depending for its raw materials on the industrial fisheries of the North Sea. 
BioMar is supplied with fish meal and oil derived from the North Sea sandeel fishery. 
The company says 'we agree that it should be effectively controlled' and it supports 'a 
precautionary approach which permits controlled industrial fishing.' Norsk Hydro also 
have substantial interests in UK salmon farms in and around Scotland (BioMar 1996; 
Shepherd 1996b, 1996c).


2.4	Fish oil in margarines
	
As far as fish oil is concerned, the UK has imported more than any other country, 
chiefly for the manufacture of margarine. The UK currently consumes around 11% of 
global fish oil production, amounting to 1.8 million tonnes of industrial fish (COWI 
1994; FAO 1993a, Trade Bulletin 1995).

In 1995, 99,000 tonnes of fish oil went into producing both domestic and commercial 
margarines and solid cooking fats for the UK market. This represents around 70% of 
fish oil imported into the UK. In total, approximately 1.4 million tonnes of industrial 
fish were required by the UK for the production of margarines and solid cooking fats 
alone. A quarter of all margarines produced in the UK in 1995 were produced using fish 
oil (MAFF 1996b). See
Figure 5.
	
Margarines using fish oil include Tesco"s, Sainsbury"s and Iceland"s Super Soft "own 
brands": most of these are produced by Pura Foods in the UK. Commercial margarines 
and cooking fats containing fish oil are also used for the production of cakes and 
biscuits and in most commercial bakeries (including supermarkets). For instance, 
McVitie"s biscuits produced in the UK by United Biscuits require approximately 6,000 
tonnes of fish oil or approximately 85,000 tonnes of industrial fish a year (Hood 1996), 
although McVitie"s are now phasing out the use of industrial fish oil.


2.5	Fish meal in animal feeds
	
While most of the protein for animal feeds is derived from sources such as soya and 
sunflower cake, animal feeds have also contained roughly equal parts of animal (meat 
and bone) and fish meal. In 1995, the UK animal feed industry consumed 239,000 
tonnes of fish meal, predominantly in feed for chickens and pigs (MAFF 1996a).
	
The animal feed industry accounts for 65% of the UK"s consumption of fish meal. The 
remainder of fish meal production and imports is used in aquaculture feed and pet 
foods. In all, UK consumption of industrial fish meal represents approximately 1 million 
tonnes of industrial fish (MAFF 1995).

Dalgety Agriculture, whose agri-business activities include the production of animal 
feeds, holds a 21% share of the livestock feed market in the UK. Dalgety Agriculture 
would therefore  consume approximately 35,000 tonnes of fish meal a year in the UK in 
the production of these animal feeds. PIC, a subsidiary of Dalgety, is the world"s 
leading pig breeding company. Dalgety"s other interests include subsidiary company 
Spillers Pet Foods (Europe"s second largest pet food manufacturer) also uses fish meal 
from industrial fisheries. (Dalgety 1995).

Pedigree Petfoods also use fish meal, although this is solely sourced from offal ("trash" 
fish) from the UK white-fish filleting industry in Aberdeen, Scotland (Jenkins 1996).


2.6	Fish farming

Declines in fish stocks and burgeoning markets for seafood products have recently 
propelled aquaculture into a major investment focus for numerous governments and 
businesses around the world. Aquaculture currently accounts for more than 14 per cent 
of global fisheries production, an estimated 15.8 million tonnes of fish per year 
(Greenpeace 1995; FAO 1995).

Fish farming has long been promoted as a solution to overfishing. Many optimistically 
believe that supplies could one day replace declining resources from the sea (Holden 
1994). But it is often self-defeating, for the farmed fish are fed with fish caught at sea. 
In other words, most fish reared by fish farming are simply a reprocessed version of fish 
taken from the sea, notably industrial fisheries (Holden 1994; ICES 1996e: Lean 1995; 
Fishing News Int. 1996b).

A leading European producer of feed for salmon farms quotes in its company literature:  
'Fish have been captured by man from rivers, lakes and seas from early times and have 
always been a vital part of our diet. As the resources of the oceans dwindle, it has 
become necessary to find ways of ensuring supplies of a wide range of fish reach the 
table.' (BioMar 1996).

In its promotional literature, the Association of Fish Meal and Fish Oil Manufacturers in 
Denmark claims that it is 'reproducing the natural food chain' by 'an optimal use of the 
natural resources of the sea.' It remarks: 'You could take a small fish, use it as bait on a 
hook - and catch a big fish for your dinner. We do it on a more industrial scale by 
processing the small fish to fish meal and fish oil. In turn these products are used in the 
feed for the aquaculture industry growing fine fish and shrimp/prawn.' (AFMFOM 
undated).

It takes approximately five tonnes of pelagic "industrial" fish to produce one tonne of 
farmed salmon or shrimps. A raw material catch of about 2.75 million tonnes is being 
used in salmon feeds alone (pers. comm. Albert Tacon FAO). Salmon farming is, in 
effect, 'a net loss of fish' (pers. comm. Robin Welcomme FAO).

Globally, salmon farmers are now producing around 550,000 tonnes a year. Of the 
world consumption of salmon (around 1.5 million tonnes), 63% is wild salmon and 37% 
is farmed. The annual harvest of farmed salmon in Norway is in the range of 260,000-
320,000 tonnes, Chile more than 95,000 tonnes, Scotland approximately 70,000 tonnes, 
Canada 41,000 tonnes, Ireland 15,000 tonnes and other countries about 60,000 tonnes. 
(Int. Salmon Farmers Ass. 1996; WorldFish 1996f).

The annual world production of farmed shrimp in both sea and fresh water is about 
800,000 tonnes (AFMFOM undated). The growth of the shrimp farming industry has 
exploded throughout south-east Asia and Latin America in the past decade, mainly in 
countries like Thailand, Indonesia and India. Shrimp farming however, is designed for 
export markets and does not contribute to local food needs ( WorldFish 1996d).

Most aquaculture production is from developing countries and is freshwater fish. Asia 
supplied 84% of world aquaculture production in 1992. China alone supplies about half 
of the world production, and India is the second most important Asian producer (FAO 
1993c). Projections for the expected increase in marine shrimp feed demands in Asia 
alone by the turn of the century indicate a potential demand of some 776,000 tonnes, up 
from about 480,000 tonnes in 1990. It has been estimated that 20-25% of fish meal 
production by the year 2000 will be destined for aquafeeds, creating a "fish meal trap" 
(Jory 1996).

The Food and Agriculture Organisation estimates that aquaculture production will 
double over 1993 levels, from 15.8 to 31 million tonnes per year (FAO 1995). Many 
would argue that it is the demand for food that drives aquaculture development. Often 
the motivation to feed the hungry is cited, but very often aquaculture produces food for 
general and luxury consumption, not for the benefit of the underfed (Kent 1995). 

Some forms of aquaculture production may possess the potential, with proper 
environmental safeguards and in appropriate social and cultural conditions, to make a 
significant contribution to meeting important nutritional requirements. Unfortunately, 
the current emphasis is firmly fixed on increasing export-orientated production of 
luxury foods for wealthy markets, e.g. shrimp mariculture (Greenpeace 1995).


2.7	Retailer"s precautionary approach

Fish oil ban

In April 1996, Unilever undertook to cease using fish oil derived from industrial fishing 
in European waters within one year. This decision was reached in recognition of 
mounting evidence that industrial fishing in the North Sea is not sustainable as it 
threatens the marine ecosystem and puts at risk an important part of Unilever"s food 
business which relies on stocks of cod and haddock (Nash 1996; Unilever 1996).

Unilever"s use of fish oil has halved in the last three years from 20% to 10% of the 
world market, about 100,000 tonnes of fish oil a year. Of this 10%, the majority is 
represented by third-party sales (Unilever 1996). Unilever has stated that all fish oil 
markets within Europe will be discontinued, but it is reviewing, through the recently 
established Marine Stewardship Council (MSC) with the World Wide Fund for Nature 
(WWF), markets for fish oil in countries like Japan (pers. comm Unilever).

Sainsbury"s followed Unilever"s lead, announcing its intention to remove fish oil from 
some 120 of its product lines (Sainsbury"s 1996). On later review Sainsbury"s 
discovered that fish oil was present in significantly fewer products: more in the region 
of 20 (Shrimpton 1996). 

Since Unilever"s announcement, other companies such as the Co-operative, McVities 
(United Biscuits), Tesco"s, Heinz, Asda, and Pura Foods have also followed with 
statements to phase out fish oil derived from European waters.

McVitie"s have since reaffirmed their commitment by stating: 'The product is not of 
strategic importance' and they expect to phase out all fish oil by the end of 1997 with 
North Sea sourced oil phased out by March 1997. (Pers. Comm. 3/9/96 & 30/9/96, Gary 
Williams McVitie"s.) Tesco"s, Sainsbury"s and the Co-operative have adopted the same 
approach to the future use of fish oil and have stated that it will be replaced with 
vegetable oil as soon as possible (Shrimpton 1996; Sawday 1996; Henderson 1996).

'Sainsbury"s is actively eliminating from its products fish oil from all sources, and not 
just European waters.' (Shrimpton 1996).

In Holland, Cargill, Remia and Unimills (partly owned by Unilever) have announced 
their intention to phase out fish oil derived from the North Sea.

Fish meal ban

A number of animal feed manufacturers have decided to ban the use of fish meal in 
animal feeds. The main concern came from consumers and retailers that cows and sheep 
don"t eat animal products in nature and therefore should not eat fish (BBC 1996).

In Scotland, the Scottish Quality Beef and Lamb Association (SQBLA) have banned the 
use of fish meal in ruminant feeds due to concerns from retailers, such as Sainsbury"s 
and Marks & Spencer"s. Replacing the fish meal with vegetable based protein will not 
be more expensive (BBC 1996).

Earlier in 1996, J. Bibby Agriculture, one of the largest animal feed manufacturers, 
dropped the use of fish meal from its main brand of feed for dairy cattle. The ban now 
extends to cattle and lambs destined for the butcher (Brown 1996).


2.8	Future markets for industrial fisheries

There have been declines in several industrial fisheries witnessed around the world over 
the last few years. Some lessons have been learned. There are now cutbacks of between 
10% and 20% for the Peruvian industrial fisheries as well as greater restrictions in the 
imposed fishing bans and quotas. A larger share of the Peruvian catch will be 
channelled towards human consumption in the future. Currently only 5% is used for 
human consumption (WorldFish 1996c).

The Norwegian capelin industrial fishery catch has been reduced to zero in recent years 
in order to enhance cod stocks. Only if there is enough capelin for the cod will, perhaps, 
fishing be allowed again. (Fishing News Int. 1996b).

Consumers of fish meal and fish oil from the North Sea and others parts of the world are 
committing to withdraw their markets. Most users in the UK have made the decision to 
stop using fish oil all together. Others have taken fish meal out of animal feeds.

However, the Food and Agriculture Organisation predicted that fish farming will double 
by the year 2010. If these predictions are correct, or underestimated, a huge demand 
could be made on the world"s pelagic "industrial" fish stocks. Dr Barlow of the 
International Fish meal and Oil Manufacturers Association (IFOMA), believes that: 
'The future of the world"s fish meal and oil industry will become even more tied to the 
aquaculture industry' (Fishing News Int. 1996b).



Crisis in the North Sea

	
3.1	Overview

The North Sea has been one of the most prolific fishing grounds in the world, both in 
the quantity and the variety of edible species of fish (Wood 1956; Levieil 1996; Furness 
and Monaghan 1987). However, it is also one of the most exploited (Furness et al. 
1987). 

Total landings of fish from the North Sea reached an all time high at some three and a 
half million tonnes a year in 1974 (OSPAR and ICES 1993), primarily due to the rapid 
development of the Danish and Norwegian fleets designed to supply factories reducing 
fish to meal and oil (ICES 1971). The composition of the catch from the North Sea has 
dramatically changed from being totally for human consumption before the Second 
World War, to a situation where half the catch is reduced into industrial raw materials, 
fish meal and oil (OSPAR and ICES 1993; ICES 1995a). 

The North Sea industrial species comprise a significant part of the diet of cod, haddock, 
whiting, mackerel and saithe. Thus, one potential explanation for the decline in human 
consumption stocks over and above the effect of directed fisheries, is the alteration in 
food availability caused by the extraction of large amounts of fish biomass (Robertson 
et al. 1996).

North Sea industrial fisheries were founded in the early 1950s on mackerel and herring. 
They grew dramatically in the 1960s. In the 1970s the total landings of industrial 
fisheries in the North Sea exceeded landings for human consumption, averaging over 
two million tonnes per year (OSPAR and ICES 1993; Furness and Monaghan 1987; 
Gauld et al. 1986; ICES 1991b, 1995a). Since then, overall landings from the North Sea 
have fallen to a current average of between 2.3 and 2.5 million tonnes a year (OSPAR 
and ICES 1993). 

North Sea industrial fisheries land over a million tonnes of fish every year with 769,000 
coming from sandeels, 281,000 from sprats, 172,000 from Norway pout in 1994. An 
average bycatch of around 170,000 tonnes of other species such as haddock, whiting, 
herring and saithe are also landed for fish meal and oil. These are mainly juvenile fish 
(OSPAR and ICES 1993; Furness and Monaghan 1987; ICES 1995a). 

Industrial overfishing

Denmark"s share of the herring catch in the North Sea and Skagerrak rose from 21% in 
1960 to 34% in 1970 to 43% in 1977. Norway"s share dramatically increased from 2% 
in 1960 to a massive 42% in the peak year of 1965. The catches of most other countries 
fell both in absolute and relative terms (ICES 1971).

Symptoms of overfishing were apparent before the post-1965 decline in catch levels of 
herring. The international team of scientists working within ICES pointed to falling 
catches per unit of fishing effort. Such deteriorations in yield were first most apparent in 
the southern North Sea, where the Down Herring became severely depleted. As effort 
was diverted northwards, so stocks in other parts of the North Sea yielded lower and 
lower catches. The diversion of fishing activity is typical of overfishing. By moving to 
new grounds fishermen can, in the short run, keep up, or even increase total catches. 
But, as the eventual collapse of North Sea herring demonstrates, the consequences of 
excessive effort cannot be avoided in the long run (Wise 1984).

Another classic symptom of overfishing is the capture of an increasing proportion of 
immature fish, perfectly illustrates the case of the North Sea herring. An ICES working 
group calculated that, in 1968, 81% of the total catch consisted of immature herring 
(ICES 1971). Clearly, the North Sea herring provides an exceptionally severe example 
of overfishing (Wise 1984). 

Prior to the mid-sixties the North Sea mackerel fishery was relatively stable with 
landings of between 60,000 and 100,000 tonnes. After 1964 the North Sea landings 
increased very rapidly from 200,000 tonnes in 1965 to almost 1 million in 1967. It was 
more than the stock could support and landings fell as fast as they had risen (Lockwood 
1978). In a few years of fishing on immature fish for industrial purposes, stocks 
declined by 90% (Furness et al. 1987). The North Sea mackerel stock collapsed in the 
late 1960s, and has never recovered (ICES 1991b). 

As a result of the collapse in herring and mackerel stocks, the industrial fishery has 
shifted its focus from herring and mackerel to relying more on other, lower value 
species, notably sandeel, Norway pout and sprat (ICES 1991b; COWI 1994).

Norway pout and sprat were mainly important in the 1970s, after which catches 
declined. Following this, sandeel catches increased dramatically and from 1985 
onwards, sandeels have constituted approximately two-thirds of the total catches 
(Furness and Monaghan 1987; ICES 1992a). Industrial fisheries landings declined from 
1.5 million tonnes in 1989 to a minimum of 1 million tonnes in 1990, largely due to a 
steep decline in sandeel landings during the same period from 1,035, 000 to 590,000 
tonnes (ICES 1991b; ICES 1995a).

The familiar pattern of overfishing, seen in the diversion of fishing activity with  the 
North Sea herring, has been replicated in recent years in the sandeel fishery moving into 
the Wee Bankie area in 1990, which is now a major site for the industrial fishing fleets 
(ICES 1994a, b). The rapid expansion of this sandeel fishery has coincided with a 
decline in seabird breeding success at nearby colonies (Harris 1996). Similarly, 1996 
was the first time the fleet moved to the North-East Bank off the Scottish coast, a further 
diversion of fishing effort to new grounds (Baldry and Beith 1996).

The true impact of industrial fisheries on commercial fish or other wildlife in the North 
Sea is simply unknown. The lack of information on the ecological impact of industrial 
fisheries illustrates the difficulties of reconciling conservation and exploitation in the 
North Sea. The scientific advisors to the European Commission and Member States 
admit that industrial fisheries remain among the least known and least managed in the 
North Sea (ICES 1992a).


3.2	The significant sandeel

The sandeel fishery

The sandeel fishery is now the largest of the North Sea fisheries (Monaghan 1992), 
currently comprising some 70% of the total amount landed by the industrial fishery 
(COWI 1994).

The fishery for sandeels in the North Sea started in 1953 (Macer and Burd 1970). At 
first it was local but it soon spread to cover most of the southern North Sea (Hart 1973). 
It began in areas around the Dogger Bank and later expanded northwards (ICES 1995a). 
Between 1982 and 1992, the largest catches came from two main areas: one off the 
English east coast and one off the approaches to Skagerrak (ICES 1994a).

The sandeel fisheries are quite distinct in that they operate during April to September in 
relatively shallow water, using trawls with no minimum mesh size restriction (Gauld et 
al. 1986). Up to a third of the weight of all North Sea fisheries catches are landed during 
the height of the 10-12 week sandeel fishing season (ICES 1995a, 1995b).

Although sandeel catches are spread over the entire North Sea, the industrial fishery has 
favoured areas where aggregates are encountered to give high and sustained daily catch 
rates which are necessary for the economic viability of the vessel (Robertson et al. 
1996). The industry argues that this is a sustainable way to fish, claiming that catches 
have been maintained over the past forty years (Dunn 1996e; Madsen 1996). However, 
this "averaged-out" measure of sustainability ignores the intensity of fishing on 
individual sandeel grounds (Dunn 1996e).

There have also been significant changes in the distribution of exploitation, such as the 
development of fisheries around the Fisher Banks (Danish coast) in the 1980s and Wee 
Bankie and Marr Bank (off the Scottish east coast) in the 1990s (ICES 1994b). Away 
from the traditional areas, the grounds close to the Scottish coast have become 
increasingly important in recent years (ICES 1994a). 

Sandeel ecology

The sandeel is a small eel-like fish which is found swimming in vast shoals or 
burrowing into the sea-bed (McIntyre 1996). These pelagic, shoaling fish usually 
remain buried at night coming out to feed at dawn, but from October to March they also 
remain buried during the day in a dormant condition. During the winter their energy 
requirements are low and they live off fat accumulated during the summer (Warburton 
1982; ICES 1993b).

There are sandeel stocks dotted all around the North Sea and adjacent waters. Five 
different species occur regularly: the lesser sandeel (Ammodytes tobianus and A. 
marinus), the smooth sandeel (Gymnammodytes semisquamatus), the greater sandeel 
(Hyperplus lanceolatus) and Corbin"s sandeel (H. immaculatus) (Kunzlik 1989; ICES 
1995c). 
The lesser sandeel is by far the dominant species and plays a very important role in the 
North Sea ecosystem (ICES 1993b; Madsen 1994). See Figure 8. It is, however, subject 
to a high and variable natural mortality (ICES 1991b) and also constitutes 95% of the 
Danish sandeel catches (Madsen 1994). 

The geographical range of the lesser sandeel extends from the Barents Sea to Iceland, 
south to the English Channel (ICES 1995c). Because of the habitat requirements, the 
distribution is mainly limited to the shallower areas of the North Sea and dense 
concentrations are typically found along ridges and edges of banks where tidal currents 
provide continuous water renewal (ICES 1991b). They are found over sandy ground at 
depths of 20-70 metres (Kunzlik 1989).

Most sandeels reach maturity at two years, although in some years a high proportion of 
one year old fish may mature (ICES 1995c). All signs indicate that sandeels do not 
undertake a spawning migration (Madsen 1994). Sandeel eggs adhere to the substrate, 
using spawning grounds that are localized and that are determined by the availability of 
the appropriate habitat. In contrast, most other species of fish spawn their eggs in the 
water column and are widely distributed over the North Sea. The sandeel larvae emerge 
and are found with plankton two to three months later. By May to June the young fish 
have developed and during their first year they are known as the 0-group (Kunzlik 
1989).

The distribution of the lesser sandeel varies with age. Juvenile 0-group fish are widely 
distributed across the North Sea. Larger 0-group sandeels settle in areas of sandy 
substrate, usually in depths of less than 100m. These areas include many coastal regions 
around the northern UK and Denmark, and large sandbanks in the north, east and 
central North Sea. Once settled, sandeels appear to be relatively sedentary (ICES 
1996d).

Sandeel movements are poorly understood. Tagging studies and fisheries catch statistics 
suggest that although there may be movement of older fish or at least some separation 
according to age (Kunzlik 1989), populations are likely to be distinct with little 
movement over great distances (Warburton 1982). In effect, the sandeel is largely 
stationary and the North Sea sandeel must be considered as a complex of local stocks 
(ICES 1995a).

The Union of Fish Meal and Fish Oil Manufacturers in the EC (UFMFOM) claims that 
'there is no real risk of overfishing the populations of Norway pout, sprats or sandeels 
since these are ephemeral species of fish with a very rapid rate of reproduction' 
(UFMFOM 1991). However, sandeels are known to live up to eight years if they survive 
being taken by a predator (marine or human). Norway pout are generally shorter lived 
than sandeels.

In the case of the sandeel, Professor Alasdair McIntyre (Professor of Fisheries and 
Oceanography, Aberdeen University) points out that it is 'an unusually vulnerable 
species... the limited number of age groups in the stock means that over-fishing or 
sudden natural fluctuations in spawning success have a huge impact on numbers.' 
(McIntyre 1996). Reduced numbers of age groups also increases stock variability and 
the risk of collapse (Everett 1996).
Over-exploitation of these stocks at the base of the food chain is potentially highly 
damaging, with consequences not just for commercial fisheries but for the whole marine 
ecosystem (Harrison 1993; House of Commons 1993).

In a year of poor recruitment, it is particularly difficult to forecast the effects of fishing 
and also to predict what stock will be available the following year. For example, in 1991 
the sandeel fishing in Shetland had to be closed because numbers were so depleted, and 
it is only recently that the fisheries were reopened after it appeared that the stock was 
recovering (McIntyre 1996). In general, sandeel stocks are highly unstable and 
unpredictable and vary by several orders of magnitude due to infrequent climate 
changes. In contrast, stocks of bottom dwelling fish such as cod and flatfishes are 
relatively stable and predictable. (Caddy and Gulland 1983; Kawasaki 1980; Sharp 
1986).


The Shetland crash - stock questions?

The well documented collapse of seabird populations around the Shetland Islands was 
caused by a collapse in their food source Ñ sandeels. The most common theory is that 
changing currents played the greatest role in the Shetland sandeel collapse. The 
contribution made by the industrial fishery is uncertain (Wright and Bailey 1993). 
However, a Shetland study entitled 'Biology of Sandeels in the Vicinity of Seabird 
Colonies at Shetland', showed that in some years, even without a sandeel fishery, there 
will be years of poor recruitment to the sandeel stock (Dunn 1996b; Wright and Bailey 
1993). 

The Shetland sandeel crash represents a striking case of a stock failure with serious 
environmental implications and has very similar parallels to the 1970s Peruvian 
anchovy crash. Such events have been widely reported in seabird populations elsewhere 
and demonstrate the dependence of breeding seabirds on fish recruitment. There is 
general agreement that both fisheries and environmental factors play a part (OSPAR and 
ICES 1993).

The Shetland experience illustrates the difficulties of reconciling conservation and 
exploitation when fundamental ecological and behavioural knowledge is lacking, and 
also the need to obtain further information on the ecological impact of industrial 
fisheries (Monaghan 1992). 

Most ecologists are aware that marine populations, like sandeels, have natural 
fluctuations in stock size. The more successful years are an evolutionary response 
serving as a cushion against leaner times. Removing large numbers may destroy that 
population"s insurance against disease, poor weather or short rations (Earle 1996b).

The Shetland sandeel fishery grew from 8,000 tonnes in 1974 to 52,600 in 1982. The 
sandeel boats kept fishing for diminishing returns: landings fell by an average of around 
30% a year from 1982 until it was finally closed in 1991. In Shetland, the sandeel 
fishery clearly removed the sandeel population"s insurance against poor recruitment. 
The fishery remained closed until 1995. 
The failure for six successive years (1984-89) of Arctic terns to produce young 
provoked alarm. Research showed that chicks were starving, with breeding of 
kittiwakes, puffins and other internationally important colonies of seabirds similarly 
affected (Dunn and Harrison 1995; ICES 1994a; ICES 1995a; Scottish Office 1995).

The EC co-funded report, "Study of the Danish Fish Meal and Fish Oil Industry", 
cleared industrial fisheries of the charge that they deplete seabird populations. The 
report states: 'Exploitation of industrial fish by seabirds in the North Sea is very small in 
relation to that by piscivorous fish and industrial fisheries'. It comments on the Shetland 
sandeel fishery that, 'although food shortage was identified as the main reason for the 
decline in seabird breeding success, there was no evidence to suggest that the local 
fishery was the primary cause of this phenomenon.' (COWI 1994).

Prior to the re-opening of the fishery in 1995, the Scottish Office gave advice that very 
strict controls would need to be imposed to ensure that fishing effort did not exceed the 
level which might deplete the stocks, even in periods of relatively poor recruitment 
(Dunn 1996b). In the past the Scottish Office has denied the existence of any link 
between the sandeel fishery and the effects on marine predators. It claims that sandeels 
are very susceptible to poor recruitment and are easily affected by environmental factors 
such as plankton distribution (Scottish Office 1990).

The current regulations for the Shetland sandeel fishery are: a Total Allowable Catch 
(TAC) set at 3,000 tonnes per year; a restriction on the number of boats allowed to 
prosecute the fishery; and a closure date for the fishery at the end of June. In 1995, a 
little over one-third of this TAC was actually met (Scottish Office 1995; pers. comm. 
Scottish Office).

The Shetland sandeel crash was a clear warning that the stock is vulnerable to change. 
Of all the sandeel fisheries in the North Sea, the Shetland fishery is currently the only 
one that is managed at all, acknowledging the importance of sandeel stocks to the 
marine ecosystem. The Shetland Fishermen"s Association, despite opposition by outside 
fishermen, supports this strict management.


3.3	The impact of industrial fisheries

The two main effects of industrial fisheries in the North Sea on the ecosystem are: 1) the 
removal of large quantities of small fish species which may lead to a shortage of food 
for the predators; 2) bycatch of human consumption fish which may reduce the yield in 
the human consumption fisheries and diminish the spawning stock size (Kirkegaard et 
al. 1996). 

If industrial fishing alters the abundance of populations, there will almost certainly be 
indirect consequences on other species or groups of species (ICES 1992a). The effects 
of removing too many fish can dramatically alter the populations of other predator 
species in the marine food web (Gulland 1971).

Food loss - predator or man?

Marine plants, mainly single celled plankton, usually have to pass along the food chain 
through at least one, and more usually two, mouths before they are converted to fish 
flesh. Where fish teem as a result of this process, predators can normally thrive. 
Sandeels feed on animal plankton (zooplankton) and are fed on in turn by many 
predatory fish, seabirds and seals (Monaghan 1992; McIntyre 1996).

The beginning of the marine food chain is of great importance as there is a heavy loss in 
biomass between the different trophic levels as primary and secondary production is 
effectively converted into fish, such as sandeels (Coull 1993). Fish which swim in large 
shoals, such as herring, mackerel, sprats, Norway pout and sandeels are one of the 
results of North Sea primary production. These small fish are consumed by seabirds, 
seals and by large predatory fish such as cod, haddock and whiting (Furness and 
Monaghan 1987). However, since industrial fisheries began in the 1950s most of these 
small fish stocks have been heavily exploited and there would appear to be a greater 
reliance on the North Sea sandeel stocks for primary production and provision of food 
abundance. This is evident through the shift in seabird diets during this period (Cushing 
1952; ICES 1993c, 1996a)

The total annual fish production in the North Sea has been estimated at 10-13 million 
tonnes (Bailey 1986; Sparholt 1990). Many top predators, including marine mammals 
and seabirds, depend on fish. But by far the most important natural predators on fish are 
other fish. Although 224 species of fish have been found in the North Sea, fewer than 
20 species make up over 95% of the total biomass (OSPAR and ICES 1993). 
Numerically, the lesser sandeel is one of the most abundant fish in the continental shelf 
area of north-west Europe (Bailey et al. 1991) and accounts for somewhere between 10 
and 15% of the total fish biomass of the North Sea (Yang 1982).

The five main North Sea human consumption fish predators Ñ mackerel, saithe, 
whiting, haddock and cod Ñ consume some 1.3 million tonnes of Norway pout. The 
main predators of pout are haddock and saithe, and the older fish of those species are 
concentrated in the north part of the North Sea where the main pout concentrations are 
found (Robertson et al. 1996). The consumption of Norway pout by fish predators far 
exceeds the take of the fishery (Greenstreet in ICES 1994c). The diet of haddock is 
dominated by sandeel and Norway pout (Robertson et al. 1996).

The same five species consume some 1.3 million tonnes of sandeels. This represents 
32% of the total tonnage of fish prey consumed. During the peak consumption of 
sandeels in the second quarter of the year some 56% of all fish diet consists of sandeels. 
This total consumption is now considered an underestimate (Robertson et al. 1996). 
Greenstreet suggests that, with the exception of Norway pout, industrial fisheries and 
predatory fish appear to leave little over for seabirds and marine mammals (Greenstreet 
in ICES 1994c). Apparently, total consumption of sandeel equates to the total stock 
available (Robertson et al. 1996). 

It is often argued by the industrial fishers that fishing for sandeels is just like "cutting 
grass", grass being the equivalent of primary production on the terrestrial ecosystem. 
This assumption is made on the basis of the annual harvesting of agricultural crops. 
However, some ecologists point out that taking sandeels, from the bottom of the food 
chain, makes as much sense as farmers exporting spring grass to feed foreign cows, 
leaving their own cattle to graze on empty pastures (White 1994).

Impact of sandeel fisheries

Long before the sandeel fishery began, fishermen used to find whitefish feeding on the 
shoals around the sandbanks of the North Sea. For instance, Scottish fishermen call the 
sandeel "sile" and the sandbanks "banquets" because they are known to be whitefish 
feeding grounds. Where there was sile there were mackerel and codling (young cod). 
Also, the Dogger Bank has in the past been a prime site for around 22 human 
consumption fisheries including herring, haddock, hake, sole and whiting Ñ most of 
which feed on sandeels  The Dogger Bank has also been a prime site for the Danish 
sandeel fisheries. See pages 55-56. (Fishing Year Book undated; OSPAR and ICES 
1993; Whyte 1996).

Although sandeels are believed to bury themselves in the sand during winter many are 
found in the stomachs of predators during this period which may suggest that they are 
available as food in the winter. Sandeels appear in the diet of cod during the winter 
months probably because cod dig for them in the sand (Robertson et al. 1996)

'[The North Sea sandeel fishery] is removing a very large animal biomass from the 
North Sea which would otherwise be food for other fish, birds and so on, and we do not 
know enough about that situation to be confident that it is not having a more damaging 
effect than we presently think'
(Shepherd 1996a).

The largest fishermen"s organisation in England, the National Federation of 
Fishermen"s Organisation (NFFO), has in the past condemned industrial fisheries since 
they diminish important feed stock for fish and birds. They also catch immature human 
consumption fish restricting the number of new adults in whitefish fisheries, such as 
haddock (Banks 1994).

Billy Hughes, who represents fishermen in the port of Pittenweem in the Firth of Forth, 
has been campaigning for around 20 years to stop industrial fishing in the area as they 
believe that it 'has seriously affected the annual movement of whitefish into our once 
highly productive fishing grounds.' (Hughes 1994).

During the last 20 years the sandeel spawning stock biomass has fluctuated between 
500,000 and 1,200,000 tonnes, with a peak of 1,700,000 tonnes in 1987 and 1988, due 
to one very strong year class (Dunn 1996d). Something concealed in these trends is that 
not all the fishing grounds are equally productive in any one year and the fleet prospects 
until it finds the best grounds and then concentrates its effort there (Dunn 1996e). For 
example, 1993 was a bad year for fishing sandeel on the usual grounds in the southern 
and eastern parts of the North Sea. As a result, the area off the Scottish coast was fished 
more intensively and catches were twice as large compared to 1992. In 1993 the catch in 
this area was approximately 115,000 tonnes, making up between one-quarter and one-
fifth of the total Danish North Sea catch of sandeel. This is a considerable amount given 
that the fishery takes place in a very limited area (ICES 1994a; Madsen 1994).

Local impact

After four decades of exploitation of the North Sea sandeel, it was only determined 
within the last few years that the North Sea sandeel may be largely sedentary and 'must 
be considered as a complex of local stocks.' (ICES 1995a). Industrial fishermen still 
believe otherwise.

The sandeel population, within a certain area, will be influenced more by local rather 
than the overall fishing effort in the North Sea (ICES 1991b). In some areas stocks are 
inaccessible to the fishery because the ground is too rough for the sandeel trawl (Popp 
Madsen in ICES 1991b). The total extent of these areas is unclear and these "protected" 
components may be of importance for the local stock, but their actual size is unknown 
(ICES 1991b). As ICES states, 'it is alarming that the information on stock size and 
distribution (of sandeel, sprat and Norway pout) is particularly poor' (ICES 1995d), and 
that 'the present regime of assessment may not....detect the impact of local fishing 
pressure and the depletion of small localised stocks.' (ICES 1994b).

Cook (1996a) explains that whereas in the case of North Sea cod the catch is in the 
region of 100,000 tonnes and spread over the whole North Sea, a fishery for sandeel 
might be 100,000 tonnes taken out of only an area of 30 miles by 60 miles. In 1993, 
around 115,000 tonnes of sandeels were removed from the Wee Bankie. The sandeel 
fishery started in 1990 with a relatively small catch of 3,000 tonnes. The 1993 catch has 
not been equalled since. In 1996, it was estimated that 40,000 tonnes were taken from 
this same area, with 31 vessels involved (ICES 1994a; Mathews 1996).

Despite this intense fishing effort on concentrated sandeel populations ICES continue to 
assume that the sandeel stock (North Sea) 'appears to be within safe biological limits,' 
and can, 'sustain the present level of fishing mortality in the short term' (ICES 1995a). 
However, it should be pointed out that this is based on an average estimate for the entire 
North Sea stock of sandeels, whereas in fact the sandeel fishery depends on effort 
directed at particular areas (Dunn 1996d).

Growing problems

Unlike mammals, fish continue to grow throughout their lives. In the North Sea fish 
growth is most rapid during the summer months and slows down or even stops during 
the winter (ICES 1993b). Based on recent stomach sampling surveys it appears that 
sandeel and Norway pout are the two most important food items for the human 
consumption fish species of the North Sea. Up to 60% of the fishing component of the 
diet of human consumption fish species consists of sandeels and Norway pout 
(Robertson et al. 1996).

While it has been concluded that fishing is the main cause of mortality in the larger 
species used for human consumption, natural predation is the most important factor in 
the industrial species and for juvenile fish in general (ICES 1992a; COWI 1994). 
Robertson et al. 1996 concluded that between the years 1977-1986 and 1987-1994 a 
"trophic trickle" had occurred in the North Sea ecosystem, in which prey switching and 
shifts in the suitabilities of the predatory fish species had occurred, i.e. more juvenile 
fish (human consumption) had been eaten rather than the industrial species. The general 
result is that if the exploitation of sandeel and Norway pout reduces their abundance, the 
predators will eat more of the remaining species to make up their food ration, such as 
juvenile fish. This effectively increases the predation pressure on other fish species, 
such as cod, haddock, whiting, etc. (ICES 1992a; COWI 1994).
Changing predation of juvenile fish is one of the major causes of long-term fluctuations 
of the biomass and results in profound changes in recruitment. Juvenile fish have a fast 
growth rate in order to try to avoid predation and eat more during the months of June to 
September. Vulnerability to predation decreases with increasing size of fish (Laevastu 
and Larkins 1981; ICES 1989b). There can also be areas and times when not enough 
food is available, thus starvation will occur (Laevastu and Larkins 1981). The reduction 
of prey species by the industrial fishery will inevitably inhibit the recruitment of those 
juvenile fish that are eaten by other predators instead of sandeels, Norway pout, etc.

The management of the industrial fisheries involved has, therefore, an important effect 
on the human consumption fisheries. In effect, the industrial and human consumption 
fisheries are competing for the same stock because most industrial species are potential 
prey available to predators (COWI 1994). Therefore, the strict management of these 
industrial fisheries should include the protection of areas where the species (e.g. cod, 
haddock, seabirds, etc) rely heavily on sandeels , Norway pout, herring and sprats.

The Union of Fish Meal and Fish Oil Manufacturers in the European Community ' 
actually agrees that it is necessary to control both industrial fishery and fishing for 
human consumption in order to maintain the equilibrium of the food chain.' (UFMFOM 
1991). Increased catches of fish for human consumption could be the result of a 
reduction in industrial fishery leading to increases in stocks of sandeel, pout and sprats. 
The result will be that predators eat comparatively more industrial fish species and thus 
fewer cod, haddock and whiting (AFMFOM 1993). 

However, food web theorists and ecologists have shown that it is nearly impossible to 
predict what the consequences will be, and nearly impossible to attribute changes in an 
individual prey, predator, or competitor population to fishery-induced changes in a 
target stock. Therefore difficult to quantify (ICES 1992a).

Cod and sandeels

Madsen (1994) in the "Insignificant Sandeel", highlights the importance of sandeels in 
the diet of cod: 'When the traditional New Year cod are caught off Horns Reef, their 
stomachs are often full of sandeel.' Madsen, a Danish member of the ICES Working 
Group on the Assessment of Norway Pout and Sandeel, however implies in the same 
report quoted that on average sandeels only represents 7.9% of cod"s diet and 7.8% of 
haddock"s diet. Although such interactions may be simply indicated, the real world is 
much more complex. Other experts maintain that sandeel are far more important in their 
diet. See Figure 9.

The results of an "international study" carried out in 1981 and 1991 showed that 
sandeels were the most important fish prey for the younger cod, whilst for older cod the 
predominant species are haddock and whiting (ICES 1989b; Macer and Easey 1988). 
The results of this study are now thought to be underestimated due to a "computer 
coding error" (Hislop 1996). During the peak consumption of sandeels in the second 
quarter of the year some 56% of all fish diet consists of sandeels. The second quarter 
coincides with the requirement for post spawning human consumption fish species to 
concentrate on building body weight and therefore food intake is high. Sandeels are 
available in high numbers at this same period and the probability of a predator 
encountering them is much higher than during winter months (Robertson et al. 1996).

Another study published by ICES (1993d) shows that the diet of cod in the Firth of 
Forth (off the east coast of Scotland) consisted of between 49% and 100% of sandeels in 
the years 1985-87 during the months of September, October and November. See Figure 
9.

The importance of sandeels as a source of food for cod is highlighted by the fact that the 
ICES Working Group on the Assessment of Norway Pout and Sandeel regularly use the 
stomach contents of cod to produce 'worthwhile distributional information' on sandeels 
stocks in the North Sea (ICES 1993a).

Bycatch

A principal aim of the technical conservation measures of the Common Fisheries Policy 
is to protect the juveniles of protected species from being caught, either directly or as 
bycatch in the small mesh industrial fisheries. However, industrial fisheries take a 
bycatch of demersal species which are predominantly the targets of other human 
consumption fisheries (COWI 1994).

ICES estimates the average catch in the industrial trawl fishery for the five year period 
1987-1991 to be 1.3 million tonnes, including an average by-catch of 170,000 tonnes of 
species for human consumption Ñ herring, whiting, haddock, saithe, and others 
(OSPAR and ICES 1993).

In recent years, catches of whiting in the industrial fisheries have been of about the 
same magnitude as the human consumption discards and amount to about 35% of the 
total catch (ICES 1993d). The Study of the Danish Fish Meal Industry reported that: 'It 
is not entirely clear from which fishery the majority of the industrial bycatch of whiting 
is taken.' (COWI 1994). This lack of data illustrates the sort of problem facing any 
biological assessment of the impacts of industrial fisheries have on other stocks.

However, some data exists. For example, in 1991, about 28% of whiting were taken in 
the Norway pout fishery in the northern North Sea while 41% were taken in the 
southern North Sea by industrial fisheries with unspecified targets (ICES 1992b). 

'In practice it appears inevitable that in these fisheries, particularly for sprat and 
Norway pout, quantities of juvenile specimens of non-target species are caught. Both 
the Commission and the Member States have been aware of this state of affairs for a 
number of years. Handling this situation has proven to be refractory, but a number of 
partial solutions have been put into effect.' Commissioner Bonino (1996).

Bycatch limits have the objective of permitting industrial fisheries to operate while 
limiting the catch of human consumption species. Being based upon a fixed percentage 
of the total catch, which is never varied, they suffer from the disadvantage that the 
upper limit to the total bycatch is determined by the catch of the target species. The 
classic example of this is provided by the situation in the sprat fisheries. The regulations 
provide for a 10% bycatch of herring in catches of sprat. In the 1970s, one of the major 
impediments to the recovery of the stocks of North Sea herring was considered to be 
catches of juvenile herring in the industrial sprat fisheries (Holden 1994). The same 
situation is being repeated in recent years.

Bycatch limits also suffer from the practical disadvantage of having to be determined by 
sampling. This has two consequences: First, because sampling is very time-consuming, 
only a limited number of inspections can be carried out. Second, the results from 
sampling can be easily contested in court because it can always be questioned whether 
sufficient samples were taken and whether they were truly representative of the total 
catch. Consequently, prosecutions are brought only when the legal limit is so heavily 
exceeded that it is thought that the case will be successful (Holden 1994). Third, a 
salmon smolt bycatch in the region of 1,000 individuals in a sandeel fishery would be 
difficult to detect, but would represent a very significant catch of salmon.

'The Commission has also ascertained from ICES that the TACs for North Sea haddock 
and whiting recommended by ICES include the industrial bycatches.' (Commission 
1982)

One factor which has in the past been discussed between Member States and the 
Commission is whether or not bycatch (e.g. haddock and whiting caught when fishing 
for Norway pout) should be included in quota allocations. In June 1982, to the 
displeasure of the Danish, the Commission insisted that, in general, such catches should 
form part of the quota of a particular species allocated to a particular state (Commission 
1982; Council Wise 1984). One factor which has in the past been discussed between 
Member States and the Commission is whether or not bycatch (e.g. haddock and 
whiting caught when fishing for Norway pout) should be included in quota allocations. 

The Norway pout fishery

The problem of bycatch is more clearly seen in the North Sea industrial fisheries for 
demersal species, particularly in the case of the fishery for Norway pout. This is a small 
(less than 20cm) species found in the northern North Sea. Norway pout"s distribution 
overlaps that of small haddock and small whiting, and it is the youngest of these species 
which are killed as bycatch (Alverson et al. 1994). See Figure 11. Bycatches of 
protected species (e.g. cod, haddock, herring, saithe and whiting) may often be a 
problem in the Norway pout fisheries (COWI 1994; Council 1991; Robertson et al. 
1996).

Given that the majority of bycatch in the pout fishery comes from recruiting age classes, 
any geographical overlap between the industrial and human consumption fisheries will 
be likely to have an effect on the productivity of human consumption species 
(Robertson et al. 1996).

The industrial fishery for pout started in the mid-1950s in the North Sea, prosecuted 
largely by Danish and Norwegian vessels using small mesh trawls. The fishery 
expanded during the early 1970s, reaching a peak in 1974 with a total reported landing 
of 736,000 tonnes. The fishery declined during the 1980s following a decline in the 
stock size to around 102,000 tonnes in 1988 (ICES 1982, 1990a, 1995a).

In 1979, the Norway "Pout Box" was introduced Ñ an area of the North Sea off north-
east Scotland which excludes fishing for this species. The aim was to reduce the bycatch 
of juvenile fish protected species, such as haddock and whiting, in the northern North 
Sea (Wise 1984; ICES 1995a). The "Pout Box" originated in 1977 as a Community 
measure following pressure from the UK Government. At that time it covered the area 
between 567 North and 607 North, and 47 West and 07. But owing to Danish 
objections, the Council proved unable to agree on a continuation of the original measure 
at the end of 1977. The final "Pout Box" settlement was part of a package of 
conservation measures agreed in September 1980 (Wise 1984). See Figure 12.

An assessment carried out by ICES (1979) estimated that a 100% closure of the "Pout 
Box" would lead to a long-term gain in haddock stocks of 65% and whiting stocks of 
228% within the defined area through survival of juvenile and adult fish normally taken 
as bycatch (Wise 1984).

In 1985 the Scottish Fishermen"s Federation (SFF) carried out a number of trawls for 
Norway pout with small-mesh trawls around the Shetland Isles. One of the areas 
trawled, 40 miles east of Shetland, was within the original designated 1979 "Pout Box" 
and was being fished by a Danish trawler. The "Norway Pout Experimental Voyage" 
made 26 clear trawls over a 4 week period to illustrate the impact of this industrial 
fishery particularly on stocks of haddock and whiting. The study concluded that to 
remain within the legal bycatch limit, white-fish (mainly haddock and whiting) had to 
be discarded. The average bycatch was concluded as being 26%. A number of trawls 
were completely discarded as the bycatch was considerably excessive (SFF 1985).

The problem of bycatch in the Norway pout fishery is still unresolved. A Danish fishing 
skipper was fined at Stornoway Sherriff Court in 1996 for having an excess secondary 
catch while fishing for pout off Barra Head in the Hebrides. Eighteen tonnes of herring 
was kept on board, exceeding the permissible 5% bycatch (Times 1996). Another 
Danish industrial trawler, Helle Vickie, had fishing gear confiscated and was fined 
œ3,000 for exceeding the 15% whitefish bycatch limit (Shetland Times 1996). In the 
Norway pout fishery, a 15% legal bycatch is permitted, of which not more than 5% can 
be constituted of cod and haddock (Council 1991; COWI 1994).

In addition to the "Pout Box", a "precautionary" TAC covering the North Sea, the 
Norwegian Sea and the Skagerrak has been set in recent years for Norway pout, mainly 
because the stock has declined. For 1995 this amounted to 180,000 tonnes Ñ a quarter 
of the peak landings taken in 1974 (COWI 1994; ICES 1995a).

ICES has also assessed, through mathematical models, that a 40% reduction in industrial 
fisheries (mainly the pout fishery) would lead to a 20% improvement in whiting stocks, 
10% in haddock and 3% in cod (COWI 1994). These increases are mainly due to 
reduction in bycatch in the Norway pout fishery, giving a higher survival of white-fish 
juveniles (Kirkegaard et al. 1996).

The Herring ban?

Whilst landing herring directly for fish meal and oil was banned in the 1970s, 
substantial bycatches of herring are taken by industrial fisheries in the North Sea Ñ 
mainly in the sprat fishery but also in the fishery for Norway pout and sandeels 
(Commission 1996b). The overall total bycatch of mainly immature herring anticipated 
in 1992 was 300,000 tonnes, representing around 40% of the total North Sea herring 
catch (Commission 1992b). 

At the end of May 1996, scientific advice became available from ICES indicating that 
the North Sea herring stock was in serious difficulties. The situation was very similar to 
that of the mid-1970s when the size of the adult stock plummeted to such low levels that 
directed fishing for herring had to be prohibited for 4 years to allow the stock to recover 
(Commission 1996b). The Advisory Committee of Fisheries Management of ICES 
reiterated its 'concern expressed previously about the negative impact of juvenile 
bycatch on loss of yield in directed fisheries and on the spawning stock biomass.' An 
ICES report states that in 1995 alone, 6.9 billion North Sea herring were caught as 
bycatch from industrial fishing (ICES 1996b).

On 5 July 1996, the Commission decided to reduce the TAC for the directed fishery for 
herring (human consumption) from 313,000 to 156,000 tonnes. Bycatch of herring in 
industrial fisheries was limited to a total of 44,000 tonnes in the North Sea and 99,000 
tonnes in the Skagerrak and the Kattegat Ñ 12,000 tonnes in sprat fisheries and 87,000 
tonnes in other industrial fisheries. (Commission 1996b).

However, despite the fact there has been a ban on landing North Sea herring for fish 
meal and oil in place for 20 years, an estimated 143,000 tonnes of mainly juvenile 
herring will have been legally landed by the industrial fisheries throughout 1996 
(Commission 1996b).

The head of the Scottish Pelagic Fishermen"s Association, Jim Slater, who is involved 
in the industrial fishery for sandeels (together with some of his members), has opposed 
the sprat-herring fishery saying 'something has to be done urgently about this 
destructive fishery and imposing a protection "sprat box" along similar lines as the pout 
box to prevent fishing for juvenile herring on key nursery grounds is vital.' (Slater 
1996).
The Shetland Fishermen"s Association secretary, John Goodlad, has said that 'the 
Danish industrial fishery for sprats is the biggest culprit in the decline of the North Sea 
herring' and that these 'stocks will never recover unless the immature herring catch is 
eliminated.' (Goodlad 1996b).

The Danish fish meal report stated that 'choosing the "correct" balance between 
industrial and human consumption landings is only in part a scientific issue, to be 
associated with political consideration which needs to take into account the value of 
catching fish when they are small.' (COWI 1994).

Goodlad estimated that the 1995 catch of juvenile herring by the Danish (70,000 tonnes 
in the North Sea) meant the loss of what would probably have become half a million 
tonnes of fully grown herring, worth in the region of œ100 million if landed by the 
human consumption fishery. Landing the herring for fish meal and oil would only be 
worth around œ4 million (Goodlad 1996a; MAFF 1994b). 


3.4	Interactions with wildlife

Seabirds of the North Sea

Every summer millions of seabirds use the coasts around the North Sea for breeding. 
Some arrive after long migratory journeys from as far away as Newfoundland, Brazil 
and the Antarctic. Others spend their whole life within 50km of the colony where they 
were born. In June, seabird colonies on the cliffs, sand spits, rocky islets and coastal 
moorland around the North Sea shores teem with birds and echo to their cries (Lloyd et 
al. 1991). Some 10 million seabirds are present in the North Sea at most times of the 
year and in the summer more than 4 million seabirds of 28 species breed along the 
coasts of the North Sea (OSPAR and ICES 1993).

Seabirds can only survive and breed if they are able to find the right kind of food, of a 
suitable size and at an accessible depth. Populations are vulnerable to changes in their 
food stocks induced by fishing (Furness and Monaghan 1987). Breeding seabirds 
require marine feeding areas and land relatively nearby on which to nest and rear young. 
Most of us view seabirds mainly at their breeding sites, but it would be a serious 
omission to ignore the crucially important feeding areas of these birds (Pienkowski 
1991). 

Seabirds feed generally on fish and marine invertebrates which are high in energy in 
relation to volume and weight. This diet is a necessary adaptation for a lifestyle which 
requires long periods of flying when bulky or heavy food in the digestive system would 
be a disadvantage (Lloyd et al. 1991). Fatty fish are therefore of more importance rather 
than low energy fish such as cod. 

Owing to a high fat content, sandeel, sprat and herring are of high caloric value per unit 
mass. (Harris and Hislop, 1978; Massias and Becker, 1990; Hislop et al. 1991). Young 
gadoids, like cod, may be taken when more energy rich prey are in short supply (Harris 
and Hislop 1978; Hislop and Harris 1984). 

The most important fish for the nutrition of seabirds in the North Sea are sandeels, 
sprats and herrings, especially during the breeding season (Harris and Hislop, 1978; 
Massias and Becker, 1990; Hislop et al. 1991). In general, the information on diets of 
North Sea seabirds is poor for the non-breeding period (Blake 1983, 1984; Blake et al. 
1985), but moderate to very good for the breeding season (ICES 1996d).

The large shoals of herring which used to feed in the North Sea were once an important 
food source for seabirds, along with sandeels and sprats (Cushing 1952) until oceanic 
changes and fishing pressure reduced their numbers in the 1970s. Now sandeels 
represent the largest source of food for seabirds breeding along the coasts of the North 
Sea (ICES 1993c, 1996a).

During the 1980s a series of dead auks found in the southern North Sea clearly 
demonstrated that there was a fundamental problem with food supplies. Much of this 
wintering mortality may be related to availability of sprat, but healthy auks in the 
southern North Sea have a very mixed diet including scad, sprat, herring, bib, whiting, 
poor cod, sandeels, gobies, solenette, sticklebacks and others. 

Furness (1993) stated that the mass winter mortalities of auks each year in the 1980s 
were obviously due to food shortage rather than adverse weather conditions.

In December 1994, The Royal Society for the Protection of Birds (RSPB) called for a 
ban on all industrial fishing in areas where there are important colonies of seabirds or 
other wildlife. The RSPB endorses a precautionary approach to industrial fisheries 
which dictates that the total catch must be greatly reduced. Ian McCall of the RSPB said 
that 'the Wee Bankie off the Firth of Forth is a crucial feeding area for birds from the 
Bass Rock, the Isle of May and the Forth islands and we would like to see this resource 
safeguarded.' (Harrison 1993; House of Commons 1993; McCall 1994).

In a letter to the RSPB, Carston Krog of the Danish Fisherman"s Organisation, 
Danmarks Fiskeriforening, said that 'it is in my organisation"s interest to avoid closing 
areas even if this could be of significant benefit to specific colonies of birds.' (Krog 
1996).

Sandeels and seabirds

Sandeels and seabirds have always been found together and occupy the same areas, as 
sandeels are vital to the survival of seabirds. That is how fishermen found sandeel 
shoals and the fish feeding upon them long before the days of echo sounders 
(Greenpeace 1996d; White 1996).
Since most seabird colonies around the North Sea are concentrated along the northern 
UK coast, demands on industrial fish stocks are likely to be very high in some of these 
areas. This suggests that competition for fish resources between seabirds and industrial 
fisheries is possible at a local level, a point acknowledged by the Danish Fish meal 
industry study (COWI 1994). 

The work of the Joint Nature Conservation Committee (JNCC) also indicates that there 
is likely to be the largest overlap in the sandeel fishery and seabird/cetacean demand in 
the Wee Bankie, Marr Bank and Scalp Bank areas off the east-coast of Scotland 
(Vincent 1996).

In the North Sea, seabirds consume an estimated 600,000 tonnes of food per annum, 
which includes approximately 200,000 tonnes of sandeels, and 30,000 of sprats and 
small herring (ICES 1993c). As a consequence, North Sea seabirds are in potential 
competition with fisheries and at risk if the stocks of prey fish are depleted (Furness 
1987; Bailey et al. 1991). Moreover, seabird predation on sandeels is highly 
concentrated in a small proportion of the North Sea.

Seabirds prey on all age-classes of sandeels. However, the species of seabirds that have 
been found to be most vulnerable to declines in sandeels are those that feed 
predominantly on the young of the year (0-group), generally less than 0.5m from the sea 
surface. These include terns, kittiwakes and Arctic skuas. In addition, the diet and 
breeding success of puffins, a small shallow-diving species which mainly feeds on 0-
group sandeels, appears sensitive to changes in sandeel availability (ICES 1994b; 
Greenpeace 1996d).

Throughout the breeding season sandeels are a major food for most seabirds in the 
North Sea, especially the guillemot, red throated diver, shag, kittiwake, arctic tern, 
arctic skua, razorbill, puffin and great skua (ICES 1996a). The species most at risk from 
excessive sandeel fishing are surface feeders and shallow-diving auks such as arctic 
skuas, kittiwakes, arctic terns and puffins (Dunn 1996c). In winter, when sandeels 
become less available, they represent about 20% of the total seabird diet (ICES 1993c). 
Sandeel fisheries close to major breeding areas of seabirds or on inshore banks which 
are wintering areas represent a potential threat to the shag, arctic tern, arctic skua, great 
skua, guillemot, razorbill and puffin colonies (ICES 1996a).

Fisheries always have greater effects on seabirds than vice versa (Montevecchi 1993) 
and the recruitment overfishing of local sandeel stocks may represent the greatest threat 
to seabirds in the North Sea. Thus, industrial fishing, especially for sandeels, may 
compete directly with seabirds for food, especially in the breeding season, when 
seabirds are constrained to foraging within a given radius of the colony (Furness and 
Ainley 1984; German MAF 1996).

Norwegian seabird crashes

There have been recent and severe changes in stocks of two of the preferred prey 
species of Norwegian seabirds, the Norwegian Atlanto-Scandian herring and the Barents 
Sea capelin. Attributed to these changes have been massive declines in the R¨st puffin 
population and the Barents Sea guillemot population respectively (ICES 1993c).
Anker-Nilssen and Barrett (1991) estimated that the puffin population at R¨st, was more 
than 1 million pairs at the end of the 1970s and was thus one of the most important 
concentrations of seabirds in the North Atlantic. After the collapse in the herring 
spawning stock from over 11 million tonnes in 1957 to 20,000 tonnes in 1971, there was 
virtually no production of juvenile (0-group) herring in coastal waters. Between 1979-
1989, there was a 64% decrease in the numbers of puffins occupying burrows on R¨st. 
The lack of food in the R¨st area also affected the guillemots. Much of the 95% 
decrease in the guillemot population on R¨st may also be attributed to the production of 
few and underweight young guillemot and subsequent recruitment failure (Bakken 
1989).

Since the collapse in the herring stocks in the Barents Sea, capelin have become the 
dominant pelagic schooling fish and, together with sandeels, the main food source for 
most seabirds in the region (Furness and Barret 1985; Erikstad and Vader 1989; Barrett 
and Furness 1990). Between 1972 and 1975, the capelin stocks increased to around 7 
million tonnes. However, after 1975 there was a steady decline in the stocks until 1986-
87 when they collapsed to 20,000 tonnes (ICES 1993c).

The effects of this rapid collapse were twofold. During the decline in 1980-83, capelin 
was a major part of the diet of many species on Horn¨y and the breeding success of 
kittiwakes, puffins, guillemots and shags were high. In 1986 and 1987, the situation was 
very different. Both breeding seasons were very poor along the south coast of the 
Barents Sea with several species producing no young at all. In 1986, the kittiwakes 
abandoned breeding on Syltefjord, the largest colony in Norway (around 140,000 pairs), 
and the guillemots on Hjelms¨y in West Finnmark (northern Norway) had a very poor 
season (Vader et al. 1987). Kittiwakes produced fewer than normal young (ICES 
1993c).

In 1987, a sharp fall in the numbers of guillemots breeding on Hjelms¨y, Horn¨y, 
Bolshoi Kharlov and Bear Island was recorded. The decline in numbers and the 
breeding failures in 1986-87 coincided with the collapse in the capelin stock and have 
been attributed to both winter starvation of adults and the problems of finding enough 
food for chicks during the summer (ICES 1993c).

Proof or precaution?

The best known examples of recruitment failures of pelagic fish stocks leading to 
collapses or declines in seabird populations, include the kittiwakes in Nova Scotia and 
the Barents Sea, and puffins at R¨st (Furness 1993, IFOMA 1996b).

A Danish study produced in the early 1990s suggested that many seabirds exploit 
sandeel concentrations that are not generally subject to fishing pressure, such as along 
the Scottish east coast. The study warned that: 'changes in the distribution of fishing 
activity resulting in the exploitation of sandeels concentrations close to seabirds 
colonies could have a serious effect on prey availability to seabirds' (COWI 1994). 
However, the North Sea sandeel fishery expanded into the Firth of Forth area, an 
important feeding area for seabirds on the Isle of May. In 1990, sandeel landings from 
the Wee Bankie were around 3,000 tonnes. Landings increased in 1993 to about 
115,000 tonnes Ñ twice the amount caught in 1992 (ICES 1995b).

The extent to which reductions in fish stock biomass can be attributed to fishing effort 
rather than to natural factors is largely unknown (Furness 1993). Bailey et al. (1991) 
argued there is no evidence to suggest that industrial fishing for sandeels has reduced 
sandeel recruitment even in the Shetland stock where recruitment "failed". This is an 
extreme view. There is undoubtedly evidence; the issue is whether it amounts to a 
convincing case. Furness (1993) argues that this possibility cannot be discounted and 
that the industrial fishing for sandeels, which takes large quantities of immature fish 
must present a hazard to the stock, and hence to seabirds. This hazard exists in the case 
of all the industrial fisheries in the North Sea. See Figure 14.

North Sea mammals

The main threats to seals and small cetaceans from the industrial fisheries include 
bycatch, habitat degradation, such as reduction of food supply and habitat disturbance 
(Cooke 1991).

The impact of fishing is uncertain in the case of cetaceans in the North Sea owing to 
ignorance of their distribution, abundance, and mortality rates (OSPAR and ICES 
1993). However, the greatest effect is likely to occur when the fishery and the cetaceans 
are competing for the same target species (Hammond et al. 1995). Fish species which 
are commonly eaten by cetaceans in Scottish waters, e.g. sandeels, whiting and Norway 
pout are of direct or indirect importance to fisheries.

Seals

Two seal species breed along the coasts of the North Sea: the grey seal (Halichoerus 
grypus) and the common or harbour seal (Phoca vitulina). Approximately 40% of the 
world"s population of grey seals breed in European waters, of which 90% breed mostly 
on remote islands around the coasts of the United Kingdom. The North Sea population 
of grey seals was estimated in 1991 to be 43,000. Although the harbour seal is one of 
the most widely distributed seal species in the world, the North Sea contains around 
10% of the world"s population Ñ approximately 28,000 seals (OSPAR and ICES 1993).

Grey seals are considered a nuisance by fishermen in Scotland and Norway, and it has 
been suggested they may cause loss to the fishing industries of these countries as quite a 
number of commercial species are included in their diet (Parrish and Shearer 1978; 
ICES 1981). However, recent work by the Sea Mammal Research Unit (SMRU) in the 
UK has suggested that this effect may have been overestimated, as much of the grey 
seal diet can consist of small species such as sandeels (SMRU 1984). Fatty fish, like 
sandeels, will provide grey seals with approximately 2.5 times as many calories as fish 
like cod (Altman and Dittmer 1968; Geraci 1975).

SMRU examined seal faeces from 9 haulout sites around the British Isles. Their results 
indicate that overall, 60% of the diet of British grey seals is made up of sandeels, 12.4% 
was tusk and ling, 6.7% Norway pout, 6.3% whiting, 5.8% flatfish, 5.1% haddock, 
saithe and pollack, 3% cod, and less than 1% blue whiting, blenny, goby, herring, sprat 
and horsemackerel (Northridge 1984). Although sandeels overwinter buried in the sand, 
they are nevertheless found in seal droppings during this season (ICES 1993b).

The Scottish Office have also shown that seals were eating about twice the weight of 
sandeels than the amount taken by the Shetland sandeel fishery prior to it being closed 
in 1991 (Scottish Office 1990).

Harbour porpoise

The North Sea is regarded as the most important habitat for porpoises (IWC 1991). The 
harbour porpoise (Phocoena phocoena) is the most common cetacean species in coastal 
waters, particularly in the north and west of the North Sea (OSPAR and ICES 1993) and 
is known to feed on small herring, sprat, sandeel and Norway pout (Rae 1973). 

The population of harbour porpoises in the North Sea is estimated to be between 
242,000 and 384,000 based on the Small Cetacean Abundance in the North Sea 
(SCANS) survey carried out in 1994 (Hammond et al. 1995). Reijnders (1992) earlier 
reported that the population in the North Sea is estimated to have fallen by between 
53,000 and 89,000 individuals, though not all of this decline is through being caught as 
bycatch (Woodley and Read 1991).
Tomilin (1957) found that on the European coasts herring and whiting were the main 
species taken, and that this seasonal movements were attributable to the pursuit of 
herring. However, recent investigations of the diet of harbour porpoises in UK waters 
found a low prevalence of herring in their stomachs (IWC 1990). Evans (1990 a, b) has 
suggested that the depletion of herring stocks may have resulted in the disappearance of 
harbour porpoises from British coastal waters during the last 50 years. 

In some parts of the North Sea, there has been a decline in sightings of harbour 
porpoises in inshore waters. Population sizes and trends in the North Sea and the 
Kattegat/Skagerrak are largely unknown. However, analysis of sightings indicate that 
the harbour porpoise no longer frequently inhabits the Wadden Sea and the southern 
North Sea (OSPAR and ICES 1993). It has also been suggested that the decline in 
harbour porpoises off the Dutch coast may have been caused by changes in the herring 
stocks through overfishing, particularly in the mid-1960s (Smeenk and Addink 1990). 
See pages 26-27. The increase in the herring stock in recent years has been followed by 
a slight increase in sightings of harbour porpoises in the south-eastern North Sea 
(Camphuysen 1994). In 1996 the herring stock has once again declined and has come 
close to collapse.

In a recent study involving Aberdeen University, diets of small cetaceans in Scottish 
waters were investigated by analysing stomach contents from strandings and bycatches. 
From 1993-1995 a total of 55 animals were studied, including 37 harbour porpoises. 
The harbour porpoises studied showed seasonal variation in their diets. In animals that 
were stranded in summer the main prey was sandeel, making up more than half of the 
weight of prey. In winter the diet changed and included whiting, haddock, saithe, 
pollack and herring (Santos et al. 1995). See Figure 15.

Dolphins

White-beaked dolphins (Lagenorhynchus albirostris) are known to feed on fish, 
including cod, herring and whiting. According to Tomilin (1957), white-beaked 
dolphins are most common in the North Sea along the Norwegian coast and the eastern 
coast of Britain. The occurrence of a number of commercial fish species in their diet 
may indicate some degree of competition for food resources (Northridge 1984). Also 
known to eat commercial fish species, including herring, is the Atlantic white-sided 
dolphin, of which the population size is unknown (Tomilin 1957; Leatherwood et al. 
1983). Tomilin (1957) states that this dolphin enters Norwegian fjords in vast numbers 
in pursuit of herring.

Studies of common dolphins (Delphinus delphis) in coastal Scottish waters have shown 
that they feed mainly on sandeels, which comprise half the total weight of prey eaten; 
the other half being whiting, haddock, saithe or pollack. The striped dolphin is known to 
feed on Norway pout (Santos et al. 1995).

Whales

The minke whale (Balaenoptera acutorostrata) has been known to include fish in its diet 
(Von Brandt 1972). Off West Greenland sandeel and krill are mainly taken (Stewart and 
Leatherhead 1985). In the North Sea, minke whales are thought to feed on sandeels and 
whiting (Santos et al. 1994).
The minke whale is thought to be less abundant now than in former years (Northridge 
1984).  Estimates of the size of the minke whale population in the Northeast Atlantic are 
heavily disputed by Norway because of its commercial interest in whaling for this 
species. The only area surveyed in great detail was the North Sea, by the independent 
SCANS multinational survey in 1994, which estimated that there were 3,600 minke 
whales in this area. The Norwegian survey carried out by an all whaler crew in 1995 
estimated 20,294 in the same area (Santos et al. 1994; Hammond et al. 1995; 
Greenpeace 1996b, 1996c).

Minke whales migrate and are frequently seen in the northern and central North Sea 
during the summer, especially the western part. The SCANS survey sightings were 
concentrated in the north-western North Sea, north of about 557 North and west of 
about 27 East (Hammond et al. 1995). See Figure 16. They are rarely seen anywhere in 
the area at other times of the year (îien 1991; Northridge et al. 1995; Evans 1992). 
Seasonal changes in distribution may be related to environmental factors such as food 
availability (IMM 1996). See Figure 16.

Killer whales (Orcinus orca) are distributed over the north-western North Sea, along the 
east coast of Norway and in the western Channel (Evans 1988; Hammond and Lockyer 
1988). Killer whales are known to eat fish, squid and other marine mammals 
(Leatherhead et al. 1983).

The Sowerby"s beaked whale in Scottish coastal waters is known to feed on sandeels 
and whiting (Santos et al. 1994).

Fin whales, like most other species of baleen whales in the Northeast Atlantic, were 
badly affected by whaling and have been slow to recover. Their diet includes herring 
and capelin, stocks of which have been heavily exploited in the area. Off the coast of 
northern Norway, fin whales have been observed following and feeding on spawning 
shoals of capelin. Similarly, they followed spawning herrings in this migration to the 
south west coast of Norway (Ingebrigtsen 1929). Fin whales may have been adversely 
affected by change in these fish stocks.

Humpback whales are also known to eat capelin and herring (Tomilin 1957; Gaskin 
1972) for which there has been an intensive industrial fishery in parts of the Icelandic 
and Barents Sea. The possibility of this fishery affecting the rate of recovery of the 
population of humpbacks, which is likely to be less than a few thousand, cannot be 
overlooked (Northridge 1984; Tomilin 1957).


3.5	Closed areas

Within EC legislation a number of geographic areas are defined within which fishing 
activities are limited. The limitations may be defined as lasting for the whole year or for 
only part of the year. All areas closed for the conservation of fish stocks established in 
Community regulations are based on scientific advice. The EC believes that the 
establishment of closed areas is an important tool for the conservation of fish stocks 
(Commission 1996c).
Area closures appear to be particularly suitable for reduction of some unwanted side 
effects, such as bycatch. Indeed, within the array of technical measures of the EC 
fisheries management scheme, area closures, in combination with closed seasons, have 
been applied efficiently to safeguard nursery areas against exploitation, to reduce 
catches of undersized fish and to improve exploitation patterns. Areas closed to all 
forms of fishing have also been proposed for reasons of nature conservation, with 
particular reference to benthic communities (ICES 1992a). 

Major features of areas "sensitive" to industrial fishing

Some major criteria already exist where there are regulations in place to protect juvenile 
fish stocks and marine habitats from fishing effort. These criteria for sensitive/critical 
areas include:-

A.	nursery areas, spawning areas, or areas with high abundance of juveniles, of 
	non target species and/or;
	
B.	areas where there may be a significant bycatch of other species and/or;

C. 	areas where there may be significant risk of damaging competition with natural 
	predators, e.g. other fish, birds or marine mammals and/or;
	
D. 	areas where the target species is particularly vulnerable, eg. discreet local 
	populations and/or relatively rare species.

The following sensitive areas in the North Sea are suggested by Greenpeace for  
consideration with a view to their being protected from the unwanted effects of 
industrial fishing, in particular the sandeel fisheries. In May 1996, Greenpeace called on 
North Sea Governments to formally identify sensitive areas in the North Sea in which 
no industrial fishing should take place as a matter of urgency.

The plaice box
	
The plaice box is an extension of the 12 mile zone north of the Netherlands, in the 
German Bight and along the coast of Jutland. The box was established in 1989 and has 
been subject to fishing restrictions because it is a nursery area for plaice and cod (Daan 
et al. 1990; Commission 1995). Currently, fishing vessels over 300 horse power are not 
allowed within the box.
	
In the Common Fisheries Policy special attention is given to the 12 mile zone. This is 
because the area has a special function as a nursery area and is also important for 
breeding Sandwich tern populations and for red and black throated divers, red-necked 
grebe, common scoter, little gull and common gull (Skov et al. 1995).

It is in the interest of the fishing industry to protect nursery areas. EC policy supports 
this objective. The plaice box could be used to increase protection of this nursery area 
from small-mesh trawls of the industrial fisheries. (Evenwicht 1993).

The Norway pout box

The aim of the Norway "Pout Box" is to protect juvenile stocks of haddock and whiting 
from the industrial fishing for Norway Pout. Massive exploitation of the latter with 
small mesh nets inevitably produces a bycatch of young whitefish (Wise 1984). 
Currently, industrial fishing with small-mesh trawls for Norway Pout is not allowed in 
the area. Refer to pages 40-42 for Norway pout fishery and "Pout Box".
	
The western margin
	
The western margin is an area of international importance for seabirds. The vast 
majority of this zone encompasses the second and third most important areas for 
seabirds in the North Sea. Many fish eating seabirds have important concentrations here 
(Skov et al. 1995).

Juvenile (0-group) haddock are widely distributed over the north-western North Sea but 
are more concentrated at ages 1 and 2 as well as 3+ in winter (ICES 1990a). Haddock 
consume large quantities of sandeel. Haddock are mainly concentrated from the middle 
to the west side of the North Sea south of 56.57 N latitude and across the entire North 
Sea north of this latitude. Heavy concentrations of haddock are associated with sandeel 
grounds at the Northumberland coast, Wee Bankie and northwards along the Scottish 
coast including Orkney and Shetland waters (Robertson et al. 1996).

UK north-east coastal zone

A UK north-east coastal zone extends seaward for 30 miles encompassing coastal fish 
nursery areas.  The zone extends from Cape Wrath in a direction east and south to 
Flamborough Head and also encircles the Orkney and Shetland Isles.

Coastal areas are particularly important as nursery areas for many juvenile fish species. 
Both saithe and herring have nursery grounds on the Scottish northern and eastern 
coasts (Daan et al. 1990). In addition, the region off the north and north-east coasts of 
Scotland is an important spawning area for both cod and haddock. (Daan et al. 1990; 
ICES 1974). 

It is probable that a large proportion of the cod in the northern North Sea spend their 
first one or two years of life inshore along the UK coast where they cannot be caught by 
conventional survey gear (Daan et al. 1990).

The Wee Bankie area 

The Wee Bankie area (including the Wee Bankie, Scalp Bank, Marr Bank, Cockenzie 
Bank), off the Firth of Forth, is particularly important for seabird populations. Major 
seabird colonies in the vicinity include the Isle of May, Bass Rock and St Abb"s Head 
(Lloyd et al. 1991). The Wee Bankie is a crucial feeding area for birds from the Bass 
Rock, the Isle of May and the Forth islands (McCall 1994). See Figure 17.

The Isle of May (Outer Firth of Forth) is a National Nature Reserve (NNR) and one of 
the key sites in the JNCC"s Seabird Monitoring Programme, with one of the longest-
running studies in Europe of seabird population dynamics. It is a major breeding station 
for auks in the North Sea, with 20,000 puffins pairs in 1995 (representing 22% of the 
population), about 5% of the populations of guillemots and 12% of razorbills. It also 
supports around 5% of the North Sea population of shags and significant numbers of 
kittiwakes (Lloyd et al. 1991; Dunn 1994; Koks 1994; Stone et al. 1995; Meinger et al. 
1995; Camphuysen 1996; pers. comm. Nisbet 1996).

Recent figures for seabird populations on the Isle of May indicate that there has been a 
decline in the numbers of guillemots (-5%), razorbills (-20%) and kittiwakes (-18%) 
between 1995 and 1996 (pers. comm. Nisbet 1996).

The work of the JNCC, the UK Government"s advisors on nature conservation, 
'indicates that there is likely to be the largest overlap in occurrence of fisheries 
[sandeels] and seabird/ cetacean usage - the Wee Bankie, Marr Bank and Scalp Bank.' 
The JNCC has advised the UK Government to prohibit the fishing for sandeel in the 
area (Vincent 1996). The UK Government have so far disregarded this advice.

Flamborough Head & Dogger Bank

Flamborough Head is a Site of Special Scientific Interest (SSSI), and has been accepted 
as a candidate for a Special Area of Conservation (SAC). See Figure 17. The north cliffs 
of Flamborough Head and Bempton support one of England"s most important seabird 
colonies, including an internationally important colony of kittiwakes: 83,700 pairs (35% 
of the North Sea population, 4% of the western European population), and important 
colonies of guillemots with 32,300 birds (over 6% of the North Sea population), 
razorbills and puffins. The cliffs also support England"s only mainland gannetry (780 
pairs) (Lloyd et al. 1991; Stone et al. 1995; Koks 1994; Meininger et al. 1995; 
Camphuysen 1996; JNCC 1996).

It has been suggested that the cod of the north-east coast of England are a distinct 
population within the North Sea. If this is so then future catches on the coast, which are 
such an important resource to the English industry, could be managed independently 
from other cod fisheries in the North Sea. The Flamborough spawning area (bordered 
by Brucey"s Garden, Whitby Rough, Flamborough Off Ground, the Hills and the 
Dogger Bank) is where cod eggs can always be found at spawning time (Houghton 
1977). The coastal zone to the north and south of Flamborough Head is also a nursery 
area for plaice (ICES 1974).

A number of recent reports suggest that the Dogger Bank area may exhibit higher 
primary production and more energy transfer within the food chain than other regions of 
the North Sea (Heip et al. 1992). Transfer of energy within the water column takes place 
through consumption of phytoplankton by zooplankton and, in turn, planktivorous fish 
like sandeels, by carnivorous fish (OSPAR and ICES 1993). 

The Dogger Bank has in the past been a prime site for commercial fishing with around 
22 human consumption fisheries including herring, haddock, hake, sole and whiting 
(Fishing Year Book undated). This site has been a primary site for the industrial fishery 
for sandeels.


3.6	Science vacuum

Scientific evaluation

The International Council for the Exploration of the Seas (ICES), which is responsible 
for providing recommendations for catch levels and assessing industrial species, have so 
far chosen not to advise the EC of any catch limits (e.g. Total Allowable Catch) for 
sandeels. The traditional argument against setting any catch limits is that sandeels 'are 
viable due to short life span and variable recruitment.' (ICES 1995a). However, this 
advice is contradicted by the fact that species like cod are now mostly caught at two 
years, the same age as some sandeels, and also have variable recruitment. Yet, ICES still 
advise on catch limits for cod.

'ICES has failed to get the message on sustainable fisheries through to EU politicians 
and fishermen strongly enough...Politicians are also guilty of failing to implement ICES 
advice when they get it.' Odd Naken, senior scientist at Norway"s Institute of Marine 
Research, 1996.

As early as 1982, the ICES Industrial Fishing Working Group reported that: 'It must be 
pointed out that even if the answer is that no direct management of these [industrial] 
fisheries is feasible, the importance of the species and their role in the North Sea 
ecosystem are obvious.'

The same ICES Group (1991b): stated that 'there is no apparent need for introducing 
management measures or precautionary TACs in the main sandeel fisheries in the North 
Sea.' In 1992 it restated that 'at present, there are no recommended TACs (on biological 
grounds) for these species. To some extent, the fisheries are self-regulating.' However, 
later on the same page, the Group recognised that, 'if the objective of a viable 
ecosystem shall be achieved, one must also keep these stocks at levels which ensure 
sufficient food supplies for their predators. At present, the multi-species assessment 
programme for the North Sea (MSVPA) is insufficient as a tool for elucidating these 
aspects of management. In particular...it does not take into account the area distribution 
of prey and predators.' (ICES 1992b). Faced with this uncertainty ICES have done 
nothing.

The same report acknowledges that: 'For sandeel in particular, the area distribution 
poses special problems. This fish is likely to be very stationary, and is found on many 
separate grounds. Since there are so many problems associated with industrial stocks, 
which are different from those encountered for most other stocks, the Working Group 
suggests that ACFM consider alternative guidelines for developing management 
advice...to these stocks.'

The ICES Ecosystem Working Group, which is made up of a larger representation of 
fisheries scientists from different countries than the industrial fisheries Working Group, 
states that: 'it is important that fisheries for those species on which seabirds (and other 
marine predators) depend are managed with other components of ecosystem in mind", 
and, "if fisheries are to be managed while taking other wildlife into consideration, 
perhaps these known areas and stocks should be the first to be considered.' (ICES 
1992a).

The Group also recognised their limitations to assess the North Sea sandeel stocks in the 
same way the assessment was carried out on the Shetland sandeel fishery by the Scottish 
Office. The Group stated that: 'Fisheries managers do not have the resources to collect 
information on a small scale.' (ICES 1992a).

This Working Group notes that, bearing in mind the potential impact of industrial 
fisheries, 'it is alarming that the information on stock size and distribution [of sandeel, 
sprat and Norway pout] is particularly poor.' (ICES 1995d). A report by ICES Study 
Group on Seabird-Fish Interactions confirms that 'the present regime of assessment may 
not, therefore, detect the impact of local fishing pressure and the depletion of small 
localised stocks' (ICES 1994b).

From 1996, the assessment of sandeels and Norway pout stocks are dealt with by the 
Working Group on Demersal Stocks (ICES 1996c; pers. comm. SOAEFD). In a 
response to the UK Government to advise on a "precautionary TAC" for sandeels, this 
Working Group responded that an upper catch limit with a "precautionary TAC" for the 
North Sea sandeel would be in the region of 1,100,000 tonnes . 

However, they also stated that to set a lower TAC they would need 'reliable predictions 
or real-time monitoring of recruitment to the fishery'. The Group goes on to admit that  
their 'ability to predict recruiting year-classes for sandeel is even worse than average 
because sandeel is a complex of local populations and the fleets change their area of 
operation from year to year and over the season'.

ICES has failed to recognise what Holden (1994) clearly understood as one of the main 
issues relating to conservation of fish stocks: 'We cannot manage either growth or 
natural mortality rates and we cannot manage the number of young fish which enter the 
fishery each year. The only factor which we can manage is the rate at which we catch 
fish.' (Holden 1994).

Scientific advice or censorship?

The scientific advice on which management measures (e.g. total allowable catches, 
minimum mesh-sizes, etc.) for the North Sea are based are provided by ICES through its 
Advisory Committee on Fisheries Management (ACFM) (FAO 1993b).

However, ICES revised its procedure and form of fishery management advice in 1991-
92. It only makes recommendations in cases where stocks are exploited outside safe 
biological limits, or when it is specifically requested to by a Member State. Where 
stocks are exploited within safe biological limits, ACFM will provide options without 
indicating a preference Ñ but will indicate the biological consequences and risks 
associated with each option. The choice of a particular option is left to the managers 
(Baldry 1996a; FAO 1993b), i.e. the European Commission and Member States.

The meetings of ACFM are restricted to its members, the ICES statistician and, in recent 
years, two scientifically qualified observers, one from the Commission and another from 
the Faroe Islands (Denmark). Membership is on a national basis, one from each 
contracting party to ICES, candidates being selected by the national authorities (Holden 
1994).

The publication of their assessment of stocks, and their recommendations concerning 
catch levels, are not subject to normal scientific peer review. Published ICES reports 
often attempt to exercise censorship, demanding that reports 'should not be quoted 
without consultation with the General Secretary.' (Holden 1994; ICES 1994b).

Holden noted in 1994 that the peculiar make-up of ICES means that membership 'does 
not consist of the most able scientists'. They are predominantly from national fisheries 
laboratories, and their attitudes may well be coloured by a perceived need to fight for 
the interests of their national fisheries (Holden 1994). The industrial fisheries Working 
Groups within ICES have largely been made up of participants from Denmark and 
Norway, more so in recent years (ICES 1991b, 1992b, 1995b). Independent fisheries 
experts, who would bring different attitudes and perspectives are excluded, making 
ICES, in the words of Holden 'a closed society.' (Holden 1994). Politicians often 
overlook or ignore this yet have a great reliance on ICES"s advice on stock management 
rather than seeking advice from independent fisheries experts. In 1996, the UK 
Government requested ICES to advise 'on appropriate levels for precautionary TACs 
for sandeels' in the North Sea (Baldry 1996a).

Scientific  management or failure?

The most obvious flaw in fisheries management policies is the way investigators 
typically regard each exploited species independently, and treat each commercially 
exploited population as though it lived separate from the complex communities of other 
fish, invertebrates, plants and an ever-changing physical and chemical environment. 
Such an approach is convenient for statisticians, but is far from realistic (Earle 1996a).

Scientific advice on fishery management is typically based upon "single species" 
mathematical models into which are fed all the factors which determine the catch from a 
fishery for any species (Holden 1994). The "single-stock" model cannot deal effectively 
with mathematically untidy factors (Earle 1996a).

Because consumption of one fish by another fish is a major cause of death for many 
species, ICES have attempted to develop multispecies models for the North Sea for 10-
15 years (FAO 1993b). Presently, the predators modelled are cod, whiting, mackerel, 
saithe and haddock: prey include cod, whiting, haddock, herring, sprat, sandeel and 
Norway Pout Ñ nine species of fish (ICES 1992a). Although sandeels are included in 
this model, none of the North Sea seabirds, cetaceans or other fish species that eat 
sandeels are included. 

The annual production of fish in the North Sea by these nine species included in the 
model is assumed to be around 7.5 million tonnes. Of this, fishing takes between 2.5 
and 3 million tonnes, predators take 2 million tonnes and other sources of mortality are 
presumed to account for about 3 million tonnes. Of the "residual" mortality, it has been 
estimated that seabirds take in the region of 0.35 million tonnes (Bailey 1986). Entirely 
unsurprisingly, these figures indicate an annual deficit of around 0.35 to 0.85 million 
tonnes.
Multispecies models, if they are to be used to make quantitative recommendations, 
require the collection of enormous amounts of detailed information because the 
quantities of prey which are eaten each year by predators change with the availability of 
food and detailed understanding of ecological interaction. Because of the enormous 
resources needed to collect the necessary data and to analyse it, only one multispecies 
model exists for Community waters, that for the North Sea which includes only nine 
species (Holden 1994). However this model is so far insufficiently developed to use in 
annual assessments of North Sea stocks. Norwegian fisheries scientists have stated that 
it will use multispecies in future assessments but only when 'ICES finds the models 
good enough.' (Commission 1996c). The enormous data requirements and wide range 
of analytical issues arising with the MSVPA model make it an inappropriate tool for 
annual assessments and day-to-day management (Robertson et al. 1996).

More recently the Danish Institute for Fisheries and Marine Research modified the ICES 
multispecies models (MSVPA) to allow inclusion of other fish predators, for which data 
on biomass, food intake and food composition are available. The inclusion of starry ray 
and western mackerel as additional predators increased predation mortality for herring, 
sprat, Norway pout and sandeel (Gislason and Lewy 1995). Not surprisingly, the more 
predators included in these models would lead to more and more mortality on industrial 
species, as shown by the Danish work.

'models routinely used by biologists....induce overfishing.' Dr Daniel Pauly,
International Centre for Aquatic Resources Management, Philippines.

Often, models simply reflect the modelmakers preconceptions. Given our lack of 
understanding of how complex ecosystems work, great caution should be exercised in 
disrupting them, as nature rarely performs on cue. At worst, trends reflected by models 
based on incomplete data may be dangerously misleading and raise false hopes, for 
example, about "surplus" numbers of one species when a known predator is depleted 
(Earle 1996a; MacIntyre 1996). At an ICES seabird symposium in November 1996 the 
Chairman, George Hunt, referred to the 'gross inaccuracies in multi-species 
assessments.' (Hunt 1996).

Other biologists have been extremely critical of what fisheries scientists attempt to do. 
One summed up the feeling of many when he said: 'Fisheries management is so 
accustomed to inaccuracy in its basic models that striking differences between model 
and observation are scarcely noted...fisheries biologists fit data to models that are 
clearly inaccurate and make management decisions on that basis.' (Peters 1991).

Fisheries science is an inexact science and advice must, therefore, reflect its 
uncertainties if that advice is to retain its integrity (Holden 1994). The House of Lords 
Select Committee on Fish Stock Conservation and Management recommended that: 
'Where the scientific evidence offers a range of options, decision-takers should be left 
in no doubt as to which option accords with the precautionary principle. Scientists will 
always want to draw attention to any degree of uncertainty in their assessments. But 
scientific professionalism i