Are coastal power stations affecting Northern European inshore fish populations?

by P.A. Henderson
Animal Behaviour Research Group, Dept. of Zoology, University of Oxford,
and Pisces Conservation Ltd.

(PLEASE NOTE: details on operational power stations were correct at the time of writing this article; plainly some stations will now have closed, and other new stations come online)

At present there are 45 large, direct-cooled power stations bordering Britain and its adjacent continental coast. For the 119 species of fish killed during cooling water extraction it is estimated that 3 to 5 x 108 individuals > 3cm long are killed per annum. For eggs, larvae and fish < 3cm long the mortality rate is circa 1014 individuals per annum. For some species these losses may significantly reduce abundance or undermine conservation efforts. Fish are particularly vulnerable as power stations are often sited in estuarine nursery areas or on migratory paths. The 17 power stations sited in the southern North Sea are estimated to kill sole and herring equivalent to circa 50 % of British commercial landings for the region. In Northern Europe there is a need for international cooperation to determine the magnitude of fish kills and their effect on populations.

From the 1950s there has been a steady development of coastal direct cooled power stations which use large volumes of cooling water (about 50 m3 s-1 for a 1250 MW nuclear power station). The fate of animals sucked in with the water depends on their size. Fish > 30 mm in length are caught by screens of 2.5 - 10 mm mesh. These impinged fish are killed, even if a return system has been installed. Those which pass through the screens travel via the station condenser circuits to be discharged to sea. Few survive passage as they suffer mechanical, temperature, biocides and pressure damage.

Data on passage and impingement moralities were either obtained from the literature or estimated by extrapolation from adjacent power stations for which surveys had been undertaken. Most estimates refer to plant operation in the 1980s. Similar, if not larger losses are probably occurring today. Table 1 lists all the studies known to have been undertaken in N. Europe and gives the estimated total annual kill through impingement for each station under the assumption that all the installed capacity is in constant use. Almost all inshore fish species are vulnerable to capture by cooling water intakes. The stations listed in Table 1 have caught 119 of the 122 inshore species known from the region (Henderson 1989). For the 33 stations for which estimates could be obtained the annual kill is estimated as 5.7 x 108 individuals > 3 cm long. These stations comprise 83% of the total installed pumping capacity. Plant availability and demand are constantly changing, but, probably 50-75% of the total pumping capacity is used per annum giving a total annual mortality on the filter screens of between 3 and 5 x 108 individuals. The commonest species caught are Sprat Sprattus sprattus, whiting Merlangius merlangus, sand goby Pomatoschistus spp., and flounder Platichthys flesus. Table 1 also includes estimates of the number killed passing through the station. Because of lack of data for many stations, reliable estimates of total mortality cannot be given, but, indications are that total mortality for eggs and fish < 3 cm long are of the order of 1010 and 1014 individuals per annum respectively.

Power station catches are dominated by young fish and recruitment varies between years. At Hinkley Point B, in the Bristol Channel sampling over a 15 year period gave estimated total annual captures ranging from 6.0 to 17 x 105. At sites where large numbers of clupeid fish are caught between year variation is higher. At Paluel (N. France), for example, between year variation in the catch of sprat Sprattus sprattus and herring resulted in the annual catch changing by a factor of 10 between consecutive years (Table 1). At Dungeness (S. E. England) occasional massive captures of sprat result in complete blockage of the filter screens and station closure. The values used to estimate total annual captures were years when Paluel, Sizewell and Dungeness did not catch large herring or sprat shoals.

Dramatic as these numbers are they have little meaning unless compared against either population size or mortality rates from other causes. While isolated populations in the vicinity of intakes may be destroyed it is difficult to determine if mobile, widely distributed fish are also likely to be suffering significant losses. Only for commercial fish are there estimates of population size or mortality rate to compare power station kills against. To gain some appreciation of the significance of the numbers a series of different approaches are taken for 6 wide spread species each representing a different life style. Sole, Solea solea is an abundant, commercially fished, benthic flatfish with planktonic eggs and young that settle close inshore. Herring, Clupea harengus, is pelagic, has benthic eggs and has been commercially over-exploited. The eel, Anguilla anguilla is an abundant catadromous species breeding in the Sargasso sea. Bass, Dicentrarchus labrax ranges from Northern Scotland to the Mediterranean, and the juvenile fish use estuaries. Twaite shad, Alosa fallax, is an anadromous member of the herring family which was common and widely distributed in Northern European estuaries but is now considered endangered.

For sole, power station and commercial fishing mortality were compared using the Equivalent Adult Value (EAV) method (Horst 1975), (Turnpenny 1988). Fish number are converted to the equivalent number of 3 year old fish, which is the age at which they enter the commercial fishery and start to reproduce. The EAV is defined as the proportion of the average lifetime fecundity of an adult that has just reached maturity which is required to replace a juvenile of known age, t:

EAV(t) = 1/S(t)F(a),

where: S(t) is the probability of survival to age t, and F(a) the average lifetime egg production.

F(a) was calculated as:

where: a is the age at which > 50% of the fish are mature, m the number of age classes in the population, P(j) the proportion of females mature at age j, S(j) the probability of survival from age a to j, E(j) average fecundity at age j, and R(j) proportion of females in age class j. Thus an old juvenile, many of whose siblings have already died, will have a higher EAV than a recently born larva. The EAVs used are tabulated in (Turnpenny 1989).

Sole are abundant in the Southern North Sea (ICES area IVc) where there is a commercial fishery. Seventeen large power stations were operating in this region in the 1980s, if all were working at full capacity the calculated EAV of the sole they would kill by number and weight was approximately 1.73 x 106 and 3.37 x 105 Kg respectively. These estimates are only for the egg and post-metamorphosis stages as there are no quantitative data for the capture of larval sole on these power stations. The EA weight is 45.9% and 10.5% of average UK and International sole landings for area IVc for the years 1989-91 and 1986-88 respectively (MAFF fisheries statistics). If account is taken of intermittent power station operation the number killed for the years 1985-1992 would probably be reduced to between 50 - 75% of the maximum. However, it is known that larval sole are caught at some stations (Dempsey 1983) and it is certain that not all small sole killed were observed as some penetrate the screens.

Previous studies of UK east coast herring estimated an EAV of 435 tonnes per annum which was 50% of the average annual UK commercial landings for ICES area IVc between 1989-91 (Turnpenny & Henderson 1992). The EAV for total power station mortality would be higher as this estimate did not include passage mortality. However, the largest captures of North Sea herring probably occur at power stations in the English Channel including Dungeness A and B, Paluel and Graveline. Single filter screens at Paluel and Graveline were estimated to catch 1.55 x 107 and 1.16 x 107 O-group herring per annum respectively (Unpublished Electricite de France report). Running at full capacity both stations catch in the order of 108 O-group herring per year . Scottish stations are also situated in important herring nursery areas resulting in kills of 106-107 post-larval herring per annum.

The eel, Anguilla anguilla, supports a small commercial fishery for which no landing statistics are available. This common and widely distributed fish has declined in abundance in many European rivers including the Thames (Naismith & Knights 1988). Unlike most fish it is caught both by stations sited on rivers and the coast. Elvers and glass eels are caught but passage mortality is unknown. The number of large individuals caught is known. For the 18 power stations included in Table 1 for which reliable data exist eel impingement was available, the number killed between 1980 and 1990 was about 2.4 x 105 per annum. By extrapolation, the total annual catch of maturing eels by coastal European stations in N Europe is of the order of 106 individuals. They are also killed by freshwater power stations. A study of fish impingement at six Dutch freshwater stations gave an estimated total annual mortality of 1.17 x 105 (Hadderingh et al. 1983).

Following concern about declining bass catches during the 1980s, inshore estuarine nursery areas were defined where fishing for bass is restricted. No restrictions were placed on the operation of power stations in nursery areas. For Kingsnorth power station on the Medway Estuary it is estimated that during the winter of 1987/88 2-5% of the local O-group population was killed on the filter screens (Pickett & Pawson 1994). Grain, W. Thurrock, Littlebrook and Tilbury power stations are also situated within the Thames basin. Kingsnorth and W. Thurrock kill an estimated 7.7 x 103 and 43.5 x 103 bass per annum respectively, indicating that mortality at W. Thurrock and the adjacent Littlebrook may have had more important impact on the bass population. Another important nursery is the Severn estuary. During the 1980's, Berkeley and Oldbury power stations, operated in the centre of distribution of the young bass. Berkeley is now closed. The estimated annual catch at Oldbury was 1.49 x 105 individuals, Berkeley was probably catching similar numbers. Other power stations operating in the Severn estuary / Bristol Channel and capturing juvenile bass were Hinkley Point A & B ( 2.7 x 103per annum), Uskmouth, Aberthaw and Pembroke. It can be argued that the MAFF policy to protect nursery areas was flawed, perhaps of negligible benefit, because it imposed no regulation upon the power companies, the single largest killer of juvenile bass. Similar arguments could be made for continental countries. Paluel for example catches an estimated 2.16 x 105 per annum at full capacity. For the 18 power stations included in Table 1 for which reliable quantitative data on bass impingement was available, the number killed between 1980 and 1990 was about 4.7 x 105 per annum.

Power stations also impact on endangered species such as twaite shad, Alosa fallax, which is listed under schedule 5 of the Wildlife Conservation Act and restricted in Britain to only four breeding populations. The largest British population is in the Bristol Channel - River Severn where Hinkley Point A , B and Oldbury power stations killed an average of 4.17 x 104 individuals per annum during the 1980s. In this region smaller kills also occurred at Berkeley, Uskmouth, Aberthaw and Pembroke.

We still have insufficient knowledge to determine if cooling water intakes are having a measurable effect on inshore fish abundance. The full extent of the problem is larger than presented as no estimates are given for other cooling water intakes such as those on petrochemical plants. Fish deaths can be reduced by measures such as fish deterrent and return systems, modified intake design and seasonal switching of production. This problem is not unique to Europe and high fish kills have been a concern in countries such as the USA (eg the Hudson river (Barnthouse et al. 1988)).

Table 1 - Estimated number of fish killed on the filter screens of marine and estuarine large power stations situated on the European coasts of the Southern North Sea and North East Atlantic.

Name Type, cooling water volume m3s-1, screen mesh mm Present status Total number of fish killed on the filter screens per annum. Fish > 3cm Total number of species recorded on the filter screens Method of estimation of impingement data or source Total number of fish eggs passing via the condenser circuits per annum Total number of young fish passing via the condenser circuits per annum. Fish < 3 cm Method of estimation of passage data and source
Inverkip Conventional, 11.5, 10 Shut            
Peterhead Conventional, 28, 10 Working            
Hunterston B Nuclear, 30, 10 Working            
Methil Conventional, 3.4, 10 Working insignificant          
Longannet Conventional, 90, 10 Working            
Kincardine Conventional, 11.5, 10 Shut            
Cockenzie Conventional, 38, 10 Working            
Torness Nuclear, 45, 8 Working 3.0 x 104 39 1 year survey 1991-1992 (Anon. 1993)      
Blyth A & B Conventional, 55, 10 Working 8.6 x 106 49 3 year survey 1979-81- qualitative data only. (Davis & Dunn 1982) Quantitative estimate extrapolated from Hartlepool      
Hartlepool Nuclear, 26, 9..5 Working 4.0x 106 35 3 year survey (Peaty 1993)      
Sizewell A Nuclear, 34.4, 9..5 Working 3.7 x 106 73 1 year survey 1981-82 (Turnpenny et al. 1983) 2 x 1010 4.9 x 107 Unpublished survey by Fawley Aquatic Research Laboraties 1992-93
Sizewell B Nuclear, 48, 9..5 Working 5.2 x 106   Extrapolation Sizewell A - plus survey and comparative data for 1994 3.3 x 1010 8.0 x 107 Extrapolation Sizewell A
Bradwell Nuclear, 26, n/a Working 3.20 x 106   Extrapolation Sizewell A 1.7 x 1010 4.2 x 107 Extrapolation Sizewell A
Tilbury C Conventional, 50, 10 Shut 4,7 x 106   Extrapolation from W. Thurrock   3.7 x 1013 Extrapolation from Kingsnorth
W. Thurrock Conventional, 58.4, 10 Shut 4.9 x 106 68 10 year survey   3.7 x 1013 Extrapolation from Kingsnorth
Littlebrook D Conventional, 50, 10 Working 4,7 x 106   Extrapolation from W. Thurrock   3.7 x 1013 Extrapolation from Kingsnorth
Kingsnorth Conventional, 64, 9.8 Working 9.9 x 105 59 2 year survey 1977-79 (van den Broek 1980)   4.7 x 1013 2 year survey (Dempsey 1983)
Isle of Grain (half operating) Conventional, 57.6, 9.8 Working 1/5 cap. 9.9 x 105   Extrapolation from Kingsnorth   4.7 x 1013 Extrapolation from Kingsnorth
Dungeness A Nuclear, 27, 9..5 Working 7.4 x 105   Extrapolation from Dungeness B >3.5 x 108 >4.3 x 108 Extrapolation from Graveline
Dungeness B Nuclear, 40, 9..5 Working 1.1 x 106 79 1 year survey unpublished >5.6 x 108 >7 x 108 Extrapolation from Graveline
Fawley Conventional, 60, 10 Working 1/4 cap. 6.0 x 105 80 1 year survey 1973-74 (Holmes 1975) 1.23 x 106 2.4 x 107 1 year survey 1986-87 (Dempsey 1988)
Hinkley Point A Nuclear, 40, 10 Working 1.3 x 106   Extrapolation from Hinkley B      
Hinkley Point B Nuclear, 30, 10 Working 9.9 x 105 73 16 year survey unpublished      
Oldbury -Upon Severn Nuclear, 26.5, 10 Working 2.5 x 105 75 5 year survey 1971-76 (Claridge et al. 1986)      
Berkeley Nuclear, 26.5, 10 Shut 2.5 x 105 71 Qualitative data (Claridge et al. 1986) ,quantitative estimation by extrapolation from Oldbury-Upon-Severn      
Uskmouth Conventional, 30.3, 5 Shut 2.9 x 105 35 Qualitative data (Claridge et al. 1986), quantitative estimation by extrapolation from Oldbury-Upon-Severn      
Aberthaw B Conventional, 67, 10 Working 2.2 x 106   Extrapolation from Hinkley B      
Pembroke Conventional, 50, 10 Working 1.6 x 106   Qualitative data (Claridge et al. 1986), quantitative extrapolation from Hinkley B      
Wylfa Nuclear, 68.3, 8 Working 4 x 104 59 1 year survey unpublished 9.66 x 107 1.08 x 108 2 year survey 1986-87 (Dempsey & Rogers 1989)
Heysham 1 Nuclear, 33.3, 10 Working 7.7 x 105 51 1 year survey unpublished      
Heysham 2 Nuclear,50, 10 Working 1.6 x 106   Extrapolation Heysham 1      
Coolkeeragh Conventional, 11.5, 10 Working 17 x 105 28 1 year survey, 1989-90(Moorehead & Service 1992)      
Ballylumford Conventional, 29.4, 8 Working 1.06 x 105 41 1 year survey, 1989-90(Moorehead & Service 1992)      
Belfast West Conventional, 9.1, 10 Working 1.5 x 104 27 1 year survey, 1989-90(Moorehead & Service 1992)      
Kilroot Conventional, 16.6, 5 Working 1.1 x 105 37 1 year survey, 1989-90(Moorehead & Service 1992)      
Moneypenny Conventional, 40, 10 Working            
Graveline Nuclear, 240, 2.5 Working 2.14 x 108 49 Estimated from 2 year survey 1981-82 (Blanpied-Wohrer 1984) >3.22 x 109 >4 x 109 Estimated from 2 year survey 1981-82 (Blanpied-Wohrer 1984)
Dunkirk Conventional, 21, 3 Working 6.2x 105 32 Electricite de France Unpublished data >2.8 x 108 >3.5 x 108 Extrapolation from Graveline
Paluel Nuclear, 160, 3 Working 2.04 x 109 (1984) 2.7 x 108 (1985) 46 2 year survey unpublished study by Electricite de France >2.14 x 109 >2.66 x 109 Extrapolation from Graveline
Flamanville Nuclear, 80, 3 Working            
Maasvlakte Conventional, 11.5, 10 Working 1.0x 107   1+ years survey      
Eems Conventional, 55,? Working            
Borssele Conventional, 34.5,?              
Doel Nuclear, 50, 3 Working 2.5x 107   1 year survey (Maes et al. 1996)      


Anon. 1993 Fish impingement studies at Torness Nuclear Power Station summer 1991 - summer 1992. Hull: Institute of Estuarine and Coastal Studies, University of Hull.

Barnthouse, L. W., Klauda, R. J., Vaughan, D. S. & Kendall, R. L. (ed.) 1988 Science, Law and the Hudson River Power Plants. American Fisheries Society Monograph. Bethesda: American Fisheries Society.

Blanpied-Wohrer, C. 1984 Etude qualitative et quantitative des organismes entraines dans la centrale de Gravelines (Nord). Influence des facteurs du milieu sur la capture de ces organismes. In Biology, pp. 484. Paris: L'Universite Pierre et Marie Curie.

Claridge, P. N., Potter, I. C. & Hardisty, M. W. 1986 Seasonal changes in movements, abundance, size composition and diversity of the fish fauna of the Severn Estuary. Journal of the Marine Biological Association of the United Kingdom 66, 229-258.

Davis, P. S. & Dunn, J. L. 1982 The fish fauna of the estuary of the River Blyth. Transactions of the Natural History Society of Northumbria 49, 1-13.

Dempsey, C. H. 1983 Entrainment of young fish in cooling water system of Kingsnorth Power Station, Kent: CEGB.

Dempsey, C. H. 1988 Ichthyoplankton entrainment. Journal Fish Biology 33, 93-102.

Dempsey, C. H. & Rogers, S. I. 1989 Ichthyoplankton entrainment at Wylfa Power Station, Anglesey and implications for a further sitting proposal: CEGB.

Hadderingh, R. H., Van Aerssen, G. H. F. M., Groeneveld, L., Jenner, H. A. & Van der Stoep, J. W. 1983 Fish impingement at power stations situated along the Rivers Rhine and Meuse in the Netherlands. Hydrobiological Bulletin 17, 129-141.

Henderson, P. A. 1989 On the structure of the inshore fish community of England and Wales. Journal of the biological association of the United Kingdom 69, 145-163.

Holmes, R. H. A. 1975 Fish and weed on Fawley Generating station screens Feburary 1973 - January 1974: CEGB.

Horst, T. J. 1975 The assessment of impact due to entrainment of ichthyoplankton. In Fisheries and Energy Production (ed. S. B. Saila), pp. 107-118. Lexington, MA: D.C. Heath.

Maes, J., Taillieu, A., van Damme, P. & Ollevier, F. 1996 Impact van watercaptatie via het waterpompstation van de Kerncentrale van Doel 3/4 op de biota van de Beneden-Zeeschelde. Leven: Katholieke Universiteit Leuven.

Moorehead, P. W. & Service, M. 1992 Capture of fish on screens of power stations in Northern Ireland. Irish Naturalists Journal 24, 3-8.

Naismith, I. A. & Knights, B. 1988 Migrations of elvers and juvenile European eels, Anguilla anguilla L., in the river Thames. Journal of Fish Biology 33, 161-175.

Peaty, S. 1993 An assessment of Greatham Creek fish populations from Hartlepool Power Station cooling water screens.: National Rivers Authority.

Pickett, G. D. & Pawson, M. G. 1994 Sea Bass; biology and conservation. London: Chapman and Hall.

Turnpenny, A. W. H. 1988 Fish impingement at estuarine power stations and its significance to commercial fishing. Journal of Fish Biology 33, 103-110.

Turnpenny, A. W. H. 1989 The equivalent adult approach for assessing the value of juvenile fish kills, with reference to commercial species in British Water: CEGB.

Turnpenny, A. W. H. & Henderson, P. A. 1992 Sizewell: Reappraisal of fish impingement at the 'A' Station, and implications of multiple plant operation: Fawley.

Turnpenny, A. W. H., Utting, N. J., Millner, R. S. & Riley, J. D. 1983 The effect of fish impingement at Sizewell 'A' Power Station Suffolk on the North Sea fish stocks: CEGB/MAFF.

van den Broek, W. L. F. 1980 Aspects of the biology of estuarine fish populations sampled from power station trash screen. International Journal of Environmental Studies 15, 203-216.