J. K. PINNEGAR, S. JENNINGS, C. M. O’BRIEN and N. V. C

Journal of Applied Ecology 2002 39, 377–390

© 2002 British Ecological Society

Blackwell Science, LtdLong-term changes in the trophic level of the Celtic Sea fish community and fish market price distribution

J. K. PINNEGAR, S. JENNINGS, C. M. O’BRIEN and N. V. C. POLUNIN*Centre for Environment, Fisheries and Aquaculture Science, Lowestoft Laboratory, Pakefield Road, Lowestoft, Suffolk NR33 0HT, UK; and *Department of Marine Sciences and Coastal Management, University of Newcastle, Newcastle upon Tyne NE1 7RU, UK

Summary

1. The intensive exploitation of fish communities often leads to substantial reductionsin the abundance of target species, with ramifications for the structure and stability ofthe ecosystem as a whole.2. We explored changes in the mean trophic level of the Celtic Sea (ICES divisions VIIf–j) fish community using commercial landings, survey data and estimates of trophiclevel derived from the analysis of nitrogen stable isotopes.3. Our analyses showed that there has been a significant decline in the mean trophiclevel of survey catches from 1982 to 2000 and a decline in the trophic level of landingsfrom 1946 to 1998.4. The decline in mean trophic level through time resulted from a reduction in theabundance of large piscivorous fishes and an increase in smaller pelagic species whichfeed at a lower trophic level.5. Similar patterns of decline in the trophic level of both catches and landings implythat there have been substantial changes in the underlying structure of the Celtic Sea fishcommunity and not simply a change in fishery preferences.6. We suggest that the reported changes in trophic structure result from reductions inthe spawning stock biomass of traditional target species associated with intensivefishing, together with long-term climate variability.7. The relative distribution of fish market prices has changed significantly over the past22 years, with high trophic level species experiencing greater price rises than lowertrophic level species.8. Although decreased abundance of high trophic level species will ultimately havenegative economic consequences, the reduction in mean trophic level of the fishcommunity as a whole may allow the system to sustain higher fishery yields.9. Management objectives in this fishery will depend on the relative values that societyattaches to economic profit and protein production.

Key-words: climate, environment, exploitation, fisheries, nitrogen, stable isotope.

Journal of Applied Ecology (2002) 39, 377–390

Introduction

The intensive exploitation of fish communities oftenleads to substantial reductions in the abundance oftarget species and changes in species composition(Greenstreet & Hall 1996). Large or slow-growingspecies with late maturity decline in abundance morerapidly than their smaller, faster-growing, counterparts

(Jennings, Greenstreet & Reynolds 1999), and as largespecies typically feed at higher trophic levels, fishing isexpected to reduce the mean trophic level of exploitedfish communities (Pauly et al. 2001).

Changes in the composition of landings often reflectchanges in the structure of underlying fish communities(Gulland 1987). Pauly et al. (1998a) used aggregatedlanding statistics from the United Nations (UN) Foodand Agriculture Organization (FAO) and estimates oftrophic level for individual species derived from foodweb models, to demonstrate that the mean trophic levelsof global fisheries have declined significantly since thelate 1950s. This decline was found to be particularly

Correspondence: J. K. Pinnegar, Centre for Environment,Fisheries and Aquaculture Science, Lowestoft Laboratory,Pakefield Road, Lowestoft, Suffolk NR33 0HT, UK (fax + 441502 254229; e-mail [email protected]).

378J. K. Pinnegar et al.

© 2002 British Ecological Society, Journal of Applied Ecology, 39,377–390

marked in the north-east Atlantic (FAO area 27), andreflects a gradual transition from long-lived hightrophic level piscivorous fish towards short-lived inver-tebrates and planktivorous fishes at lower trophic levels(Pauly et al. 1998a, 2000a).

Several concerns have been expressed regardingthe approach and methodology used by Pauly et al.(1998a) (Caddy et al. 1998; Pauly, Froese & Chris-tensen 1998b; Caddy & Garibaldi 2000; Pauly &Palomares 2000). These focused principally on (i)errors in assigning trophic levels to individual spe-cies; (ii) the validity of FAO landings data; and (iii)the potential influence of environmental factors.

The trophic level estimates used by Pauly et al.(1998a) for north-east Atlantic fishes were largelyderived from the North Sea Ecopath model of Chris-tensen (1995). Christensen’s (1995) estimates werebased on gut contents analyses, yet these tend toprovide only snapshots of diets in time and space andthus offer a poor basis for establishing trophic level. Inaddition, gut content analysis neglects certain dietarymaterials (e.g. gelatinous plankton and detritus)that are hard to identify but which are an importantconstituent in the diet of some fish (Deb 1997; Polunin& Pinnegar 2002). Furthermore, this tool may beparticularly unsuitable for predicting trophic levels oftop predators that feed intermittently and frequentlyregurgitate food upon capture (Bowman 1986).

The landings data used by Pauly et al. (1998a), forthe north-east Atlantic, were derived by the FAO frommore detailed International Council for the Explorationof the Sea (ICES) assessments based on 15 distinctsubareas. The large-scale geographical aggregation ofsuch data makes it very difficult to discern the speciesand fleets responsible for any observed changes inmean trophic level. There are distinct differences in thefisheries and fleets that operate in each of the ICESsubareas (MAFF 1999), which were not taken intoaccount by Pauly et al. (1998a).

The Celtic Sea is an intensively fished ecosystem andis unusual among north-east Atlantic seas in that itsmajor fisheries were developed relatively recently, at atime when good fishery monitoring and survey pro-grammes were in place. As a consequence the Celtic Seaoffers an ideal situation for study of the indirect effectsof intensive fishing on trophic structure and to com-pare the changes in trophic level of survey catches andcommercial landings. Christensen (1998) has demon-strated apparent declines in trophic level of an eco-system may be even more marked in survey data than infishery landing data from the same region.

The Celtic Sea comprises ICES statistical divisionsVII f–j, and is mainly fished by France, Ireland, theUK, Spain and Belgium. The Celtic Sea supportsvaluable fisheries for several species, notably forMerluccius merluccius (Linnaeus), Lepidorhombuswhiffiagonis (Walbaum) and Trachurus trachurus(Linnaeus) (MAFF 1999). The fisheries of the CelticSea are harvested by several distinct fleets (métiers),

characterized by the use of different fishing gear typesand different target species (Laurec, Biseau & Charuau1991; Marchal & Horwood 1996). The recent expansionof Celtic Sea fisheries has prompted concerns aboutthe present and future state of fish stocks (Horwood1993), the scale of fishery discards/by-catch (Perez,Trujillo & Pereda 1996; Treganza et al. 1997; Morizuret al. 1999; Rochet, Trenkel & Péronnet 2001) andpossible implications for the ecosystem as a whole.

Changes in fishery preferences are often driven byeconomic forces, and the value of a species willdetermine the investment that fishermen are willing tomake in order to catch it, and thus how heavily it isfished at low abundance. Existing evidence suggeststhat the average market price of a species will increaseas it becomes scarce (Murawski & Serchuk 1989;OECD 1997), and Sumaila (1998) has demonstratedthat markets are good at giving value to previouslyundesirable fish when target species become unavail-able. On a world-wide basis, between 1952 and 1994 theaverage price of low trophic level species apparentlyincreased relative to the price of high trophic levelspecies (Sumaila 1998). Long-term price data areavailable for the Celtic Sea fishery, and these provide anopportunity to examine how market prices might relateto changes in the composition of landings at a local scale.

Stable isotopes of nitrogen have been widely usedfor establishing trophic level in aquatic organisms(Cabana & Rasmussen 1996; Vander Zanden, Cabana& Rasmussen 1997; Post, Pace & Hairston 2000). Theisotope method relies on the general observation thatwith every trophic step, there is bioaccumulation of theheavier isotope 15N (Minagawa & Wada 1984). Trophiclevel estimates resulting from stable isotope analysis(SIA) largely corroborate trophic level data derivedfrom steady-state modelling (Kline & Pauly 1998;Pinnegar 2000) and/or gut contents analysis (VanderZanden, Cabana & Rasmussen 1997), but offer advan-tages because they integrate diet over a year or more inmany fish species (Hesslein, Hallard & Ramal 1993)and thus they are much less subject to seasonal bias.Moreover, isotopes in tissues of a consumer reflectmaterials that are assimilated and not merely ingested.

Because the average transfer efficiency betweentrophic levels in marine systems is c. 10% (Pauly &Christensen 1995), Pauly, Christensen & Walters(2000b) predicted that a fall of 1 in the level at which afishery operates would lead to a 10-fold increase inpotential catches. To study this effect Pauly, Chris-tensen & Walters (2000b) and Christensen (2000)introduced the fishing-in-balance (FIB) index:

eqn 1

where y is the year of the time-series, TL is the trophiclevel of the catch, TE is the mean energy-transfer effi-ciency between trophic levels (assumed to be 10%), and1 refers to the first year in a time-series that is used

FIB log

=

××

−

−

Catch TE

Catch TEy y

TL

TL

1

1 11

379Trophic level of the Celtic Sea fish community


as a baseline. If catches increase 10-fold for everyfull trophic level decline, the FIB index will remainconstant and fishing can be deemed ‘in balance’.

The objectives of the present study were to: (i)describe changes in the mean trophic level of surveycatches and fishery landings using stable isotope-derived estimates of trophic level; and (ii) examinewhether changes are reflected in the relative marketprice distribution of fish species.

Methods

Individual species were sampled in the Celtic Sea fromthe research vessel Cirolana using the standard bottomtrawl gear utilized for annual ground-fish surveys. Thisgear consisted of a modified Portuguese High-HeadlineTrawl (Warnes & Jones 1995), and tows of 30-min dura-tion were made at a speed of approximately 4 knots.Sixty-one standard survey stations were fished through-out the Celtic Sea (Fig. 1) during February and March2000. Where possible, three fish of each species, between60% and 80% of their maximum recorded size, wereselected from the survey hauls (Jennings et al. 2001a),and these animals were dissected aboard ship to obtaintissue samples for nitrogen stable isotope analysis.Approximately 2 g of white muscle was dissected fromthe dorsal musculature of each fish, placed in a vial andimmediately frozen at −30 °C. On return to the labor-atory, the frozen tissue was freeze-dried and ground toa fine powder (particles < 60 µm). This was thoroughlymixed and a 1-mg sample was weighed into a tin capsulefor stable isotope analysis (Pinnegar & Polunin 1999).

The 15N composition of the samples was determinedusing continuous-flow isotope ratio mass spectro-metry (CF-IRMS). The weighed samples were oxidizedand the resulting N2 passed to a single-inlet dual col-lector mass spectrometer (automated nitrogen carbonanalysis, ANCA; SL 20–20 system, PDZ Europa,

Crewe, UK). Two samples of an internal referencematerial (homogenized cod white muscle) were ana-lysed after every six tissue samples in order to calibratethe system and compensate for drift with time. Theconventional delta notation was used to express stableisotope ratios and these are reported (in ‰) relativeto an international standard (atmospheric nitrogen):

eqn 2

where R is the ratio 15N:14N. Experimental precision(based on the standard deviation of replicates of theinternal standard) was < 0·1‰.

Stable isotope-based estimates of trophic level werecalculated assuming a constant per trophic level frac-tionation of 3·4‰ (Minagawa & Wada 1984). Theadductor muscle of Celtic Sea scallops Pecten maximus(Linnaeus) (shell length 11·37 cm, SD 0·06; n = 6) wasused as a reference material. Scallops were caught bydredge from the research vessel Corystes in November2000 (Fig. 1). The mean δ15N of adductor muscle was7·21‰ (SD 0·18) and this value was used to predict thetrophic levels of other species whereby:

eqn 3

TLNi is the trophic level of species i, δ15Ni is the meanδ15N of species i, and δ15Nref is the mean δ15N of thescallops, which were assumed to be herbivorous/detritivorous and consequently feeding at trophic level2. Although in the present study we assigned specifictrophic levels to species, it equally would have beensufficient to report only their δ15N values and simplyassume linearity in the relationship between δ15Nand trophic level (Jennings et al. 2001b). We have

Fig. 1. ICES divisions of the seas surrounding the southern British Isles and the position of the CEFAS groundfish surveystations (crosses). The Celtic Sea consists of ICES divisions VII f–j. The star indicates the area from which scallop Pecten maximussamples were dredged, and the dotted line represents the 200-m depth contour.

δ15N = R

Rsample

standard

−

×1 1000

( )

TL = N N

Nref

iiδ δ15 15

3 42

−⋅

+



converted δ15N values to trophic level simply for easeof comparison with those of Pauly et al. (1998a).

The relative abundance of Celtic Sea demersal fisheshas been monitored since 1981 (Warnes & Jones 1995).The original purpose of the surveys was to investigatethe distribution and biology of mackerel, but subse-quently, with an increasing need for fishery-independentdata on western stocks, the objectives were widened toinclude the biology, distribution and abundance of allspecies that could be sampled representatively by bot-tom trawl. From 1982 onwards, catch numbers, weightand length compositions were recorded routinely, thusgiving a time-series of 18 years. The area of coverageextends from 47°30′N to 52°30′N and from 3°W to12°W, and includes ICES Divisions VII f,g,h,j and thenorthern part of VII e (Fig. 1). In the earlier years twosurveys were carried out each year, one in the spring(March–April) and one in winter (December). How-ever, from 1989 onwards only the spring survey hasbeen undertaken, and thus only data for the spring sur-veys were utilized in the present analysis. No springsurvey data were available for 1983.

Nominal landings of fish and shellfish are officiallysubmitted to ICES by each of the 19 member countrieson an annual basis. ICES has published these data inBulletin Statistique des Pêches Maritimes from 1903 to1987 (ICES 1906–90) and from 1988 onwards in ICESFisheries Statistics (in electronic form). We used aggre-gated data from ICES divisions VII j–k (Fig. 1). Datafor the period 1946–72 were obtained directly fromBulletin Statistique and data for 1972–98 were down-loaded from the ICES web site, http://www.ices.dk.Landings were expressed in tonnes live weight equival-ent. Bulletin Statistique describes the limitations of theannual data sets, and notable discrepancies over the time-series include the occasional aggregation of Limandalimanda (Linnaeus) with Microstomus kitt (Walbaum)in French reports and the separation of Micromesistiuspotassou (Risso) from non-specified gadoids in 1975.

The weight and value of fishery landings by English,Welsh and Northern Ireland vessels are recorded annu-ally by the Centre for Environment, Fisheries andAquaculture Sciences (CEFAS; formerly MAFF Fish-eries Directorate). We estimated the average price of 26fish species (in £ per tonne) landed from the Celtic Sea(ICES regions VII f–k) between 1979 and 2000, aslandings (in tonnes) divided by value (in UK £). Forthese 26 species (Table 1) there was a complete time-series of price estimates; species where a complete

time-series did not exist were excluded. Within eachindividual year, we examined the relationship betweenprice and trophic level using linear regression. The slopeof the regression (b) was taken as the relative priceindex (RPI) for that particular year. If the RPIdecreases, then the prices of lower trophic level speciescan be interpreted as having increased relative to highertrophic level species (sensu Sumaila 1998), and con-versely if RPI increases then the top predators haveincreased in value relative to the low trophic level spe-cies. If the RPI remains constant then the relationshipbetween the prices of low and high trophic level specieshas also remained constant, although in absolute termsprices may have increased due to inflation.

To test for significant long-term trends in time-seriesdata, non-parametric Mann–Kendall tests were per-formed (Gilbert 1987). The Mann–Kendall test is par-ticularly useful because data need not conform to anyparticular distribution. Where a significant linear trendwas indicated, the true slope (change per unit time) wasestimated using the procedure developed by Sen (1968).This non-parametric estimator is not greatly affectedby outliers and can be computed when data for individ-ual years (e.g. 1983 in Celtic Sea survey data) are missing(Gilbert 1987). Differences were judged significantwhen P < 0·05. For some analyses, species were groupedinto ISSCAAP (International Standard StatisticalClassification of Aquatic Animals and Plants) categories.

Results

Nitrogen stable isotope compositions were determinedfor 48 fish species, which represented 99·7% of the bio-mass in the year 2000 groundfish survey. δ15N valuesranged from 10·2‰ in Gadiculus argenteus Guichenotto 17·2‰ in Merluccius merluccius, and trophic levelranged from 2·88 to 4·94 (Table 1). The unweightedmean trophic level of all fish combined was 3·86.

There was a significant decline in the mean (weightedby species abundance) trophic level of the fish caught intrawl surveys from 1982 to 2000 (Mann–Kendall Z =−2·01, P = 0·04; Fig. 2a). Sen’s non-parametric estima-tor of slope indicated that the average rate of trophiclevel decline was around 0·04 year−1. This decline co-incided with substantial changes in the composition ofsurvey catches, whereby the proportion represented by‘cods and hakes’ and ‘sharks and rays’ (high trophiclevel species) seems to have decreased markedly(Fig. 2b), as did the proportion represented by Trachu-rus trachurus [trophic level (TL) = 3·94]. In contrast therelative abundance of Scomber scombrus Linnaeus(TL = 3·61) increased (Fig. 2b), as did the proportionrepresented by ‘seabasses, redfishes and congers’



(ISSCAAP group 33), largely due to increased catchesof the boarfish Capros aper (Linnaeus) (Fig. 2c), whichfeeds at a low trophic level (TL = 2·94). Among speciesof ‘seabasses, redfishes and congers’, the proportionrepresented by monkfish (Lophius spp.), which feed ata high trophic level (Lophius piscatorius LinnaeusTL = 4·09), declined markedly, from 22·8% of thegroup’s biomass in 1986 to only 0·7% in 2000 (Fig. 2c).The mean trophic level of the groundfish survey was

relatively stable until 1988 (Fig. 2a), but in subsequentyears it became more variable, largely due to variationin Scomber scombrus and Trachurus trachurus abundance,with very low catches of these species in 1993–95.

The mean trophic level of fish landed from theCeltic Sea declined significantly from 1946 to 1998

Table 1. Fish species collected in the Celtic Sea during the spring 2000 groundfish survey, mean total length, δ15N and calculatedtrophic level of samples. Three individuals were sampled from each species except * where n = 1; trophic levels were calculatedassuming a fractionation step of 3·4‰ and against a baseline of Pecten maximus (7·21 ± 0·18‰). †Species included in the relativeprice analysis

Scientific name Common nameFish length (mm) mean ± SD

δ15N (‰) mean ± SD

Trophic level mean ± SD

Scyliorhinus canicula Lesser spotted dogfish 647 ± 18 15·0 ± 0·91 4·29 ± 0·27Galeorhinus galeus† Tope 1090 ± 164 17·0 ± 0·82 4·88 ± 0·24Mustelus asterias Starry smooth hound 1115 ± 15 14·0 ± 0·22 4·00 ± 0·06Squalus acanthias† Spurdog 628 ± 40 12·0 ± 0·57 3·41 ± 0·17Raja montagui† Spotted ray 626 ± 18 13·7 ± 0·93 3·91 ± 0·27Leucoraja naevus† Cuckoo ray 641 ± 30 13·6 ± 0·44 3·88 ± 0·13Clupea harengus† Herring 246 ± 1 13·3 ± 0·23 3·79 ± 0·07Sardina pilchardus† Pilchard 190 ± 17 12·6 ± 0·59 3·59 ± 0·17Sprattus sprattus Sprat 126 ± 6 14·3 ± 1·26 4·09 ± 0·37Engraulis encrasicolus Anchovy 139 ± 9 14·6 ± 0·95 4·17 ± 0·28Argentina sphyraena Lesser argentine 198 ± 7 12·1 ± 0·22 3·44 ± 0·06Lophius piscatorius† Anglerfish 653 ± 45 14·3 ± 0·72 4·09 ± 0·21Gadiculus argenteus Silvery pout 118 ± 7 10·2 ± 0·26 2·88 ± 0·08Gadus morhua† Cod 735 ± 99 15·2 ± 0·77 4·35 ± 0·23Melanogrammus aeglefinus† Haddock 490 ± 30 12·6 ± 0·85 3·59 ± 0·25Merlangius merlangus† Whiting 306 ± 15 17·2 ± 1·24 4·94 ± 0·36Micromesistius poutassou Blue whiting 240 ± 10 11·1 ± 0·64 3·14 ± 0·19Molva molva† Ling 1107* 15·3 4·38Phycis blennoides Forkbeard 426 ± 31 13·3 ± 0·50 3·79 ± 0·15Pollachius virens† Saithe 1054 ± 28 14·4 ± 0·60 4·11 ± 0·18Trisopterus esmarki Norway pout 168 ± 12 13·7 ± 1·03 3·91 ± 0·30Trisopterus minutus Poor cod 146 ± 5 12·9 ± 1·12 3·67 ± 0·33Merluccius merluccius† Hake 863 ± 120 13·5 ± 0·29 3·85 ± 0·09Beryx splendens Beryx 241 ± 2 10·8 ± 0·85 3·06 ± 0·25Zeus faber† John dory 356 ± 40 14·7 ± 0·74 4·20 ± 0·22Capros aper Boarfish 116 ± 5 10·4 ± 0·20 2·94 ± 0·06Helicolenus dactylopterus Bluemouth 311 ± 17 13·5 ± 0·26 3·85 ± 0·08Aspitrigla cuculus† Red gurnard 258 ± 15 13·2 ± 0·11 3·76 ± 0·03Eutrigla gurnardus† Grey gurnard 325 ± 27 12·8 ± 0·23 3·64 ± 0·07Dicentrarchus labrax† Bass 639 ± 3 15·8 ± 0·42 4·53 ± 0·12Trachurus trachurus† Horse-mackerel 347 ± 22 13·8 ± 0·41 3·94 ± 0·12Sponyliosoma canthurus† Black sea bream 302 ± 22 15·2 ± 0·85 4·35 ± 0·25Mullus surmuletus† Red mullet 280 ± 20 15·3 ± 0·82 4·38 ± 0·24Echiichthys vipera Lesser weaver 106 ± 8 15·1 ± 0·21 4·32 ± 0·06Hyperoplus immaculatus Corbins sandeel 267 ± 5 12·2 ± 0·87 3·47 ± 0·26Callionymus lyra Dragonet 158 ± 14 12·7 ± 0·40 3·61 ± 0·12Scomber scombrus† Mackerel 289 ± 4 12·7 ± 1·26 3·61 ± 0·37Lepidorhombus boscii Four-spot megrim 284 ± 17 11·9 ± 0·12 3·38 ± 0·04Lepidorhombus whiffiagonis† Megrim 394 ± 38 12·5 ± 0·74 3·56 ± 0·22Arnoglossus imperialis Imperial scaldfish 165 ± 8 10·9 ± 0·14 3·09 ± 0·04Glyptocephalus cyanoglossus† Witch 299 ± 24 13·6 ± 0·51 3·88 ± 0·15Hippoglossoides platessoides Long rough dab 180 ± 10 14·1 ± 0·16 4·03 ± 0·05Limanda limanda† Dab 193 ± 17 14·7 ± 1·66 4·20 ± 0·49Microstomus kitt† Lemon sole 239 ± 14 12·9 ± 2·00 3·67 ± 0·59Platichthys flesus† Flounder 315 ± 3 13·5 ± 1·17 3·85 ± 0·34Pleuronectes platessa† Plaice 336 ± 13 12·9 ± 0·97 3·67 ± 0·29Microchirus variegatus Thick back sole 202 ± 3 13·5 ± 0·73 3·85 ± 0·21Solea solea† Sole 329 ± 38 14·6 ± 0·71 4·17 ± 0·21



(Mann–Kendall Z = −3·25, P = 0·01; Fig. 3a), and Sen’snon-parametric estimator of slope indicated a declineof around 0·03 TL each year. For the period 1982–98(the years covered by the groundfish survey), there wasno significant overall trend in the mean trophic level

of landings (Mann–Kendall Z = 0·48, P = 0·63;Fig. 3a).

There was a significant negative correlation (Spearmanrs = 0·49, P < 0·01) between mean trophic level and totallandings (in tonnes) (Fig. 3b). From 1946 to 1968 the

Fig. 2. Patterns of changing (a) trophic level, (b) catch composition and (c) catch composition of the ‘seabasses, redfishes andcongers’ in an 18-year time-series of fishery-independent survey data for the Celtic Sea (scientific names given in Table 1).



mean trophic level and landings varied relatively littlefrom year to year (TL 3·86–4·01, landings 51 235–201 494 tonnes); however, from 1969 to 1976 landingsincreased greatly (up to 489 776 tonnes in 1976) andthis was accompanied by a decline in mean trophic level(TL = 3·78 in 1976; Fig. 3b). Between 1976 and 1977landings fell dramatically (to only 155 131 tonnes) andthis coincided with a slight increase in the mean trophiclevel until 1985–88, when landings began to increaseagain and trophic level declined (to TL = 3·65, landings304 792 tonnes). In recent years (1989–98) landingshave varied relatively little from year to year (232 501–364 942 tonnes), and the mean trophic level of thelanded fish has ranged from 3·80 to 3·91 (Fig. 3b).

The trends apparent in the composition of fisherylandings were somewhat similar to those we havereported for survey catches (Fig. 4a). In both dataseries, there was a marked decline in the importance ofhigh trophic level groups such as the ‘cods and hakes’(78·1% of landings in 1946, 13·9% in 1998) and ‘sharksand rays’ (7·5% in 1946, 3·3% in 1998). However, infishery landings data the importance of small pelagic

species has increased more markedly. In 1946, reportedlandings of Trachurus trachurus were only 0·03% of thetotal, but after 1967 the fishery expanded greatly andrepresented 50·4% of total landings in 1998. Similarlythe fishery for Scomber scombrus was very small in theearly years of the time-series (6·6% in 1946) butexpanded throughout the 1960s and 1970s, peaking in1976, after which its importance was eclipsed by that ofTrachurus trachurus (Fig. 4a). The proportion of flat-fishes (ISSCAAP group 31) and ‘seabreams, redfishesand congers’ (ISSCAAP group 33) remained relativelyunchanged over the course of the whole time-series.However, the proportion of ‘herrings and anchovies’(ISSCAAP group 35) was less consistent and peaked in1958–59, 1962–64 and 1966–70. Between 1984 and1990, the proportion of ‘cods and hakes’ in the land-ings (ISSCAAP group 32) rose slightly, largely due tocatches of low trophic level (TL = 3·14) blue whitingMicromesistius poutassou (Risso). This peak in bluewhiting landings was reflected in a decrease in themean trophic level of landings (Fig. 3a) for these years.Subsequently, the proportion of ‘cods and hakes’ in the

Fig. 3. A 53-year time-series of aggregated fisheries landings (fish only) for the Celtic Sea: (a) changes in mean trophic level;(b) relationship between mean trophic level and total international landings (individual years indicated).



Fig. 4. A 53-year time-series of aggregated fisheries landings for the Celtic Sea: (a) changes in relative composition of landings;(b) changes in the absolute quantities of fish landed; (c) changes in the fishing-in-balance index (FIB).



landings continued to decline (Fig. 4a), largely due toreduced catches of high trophic level species such ascod and hake (TL = 3·85).

Overall, between 1946 and 1998, there was a significantincrease (Mann–Kendall Z = 7·48, P < 0·0001) in fish-ery landings from the Celtic Sea (Fig. 4b), and landingsof most fish groups increased in absolute terms(Fig. 4b). Particularly high landings of Scomber scombrusand Trachurus trachurus were taken in 1976. A clearand strongly significant increase was observed (Mann–Kendall Z = 4·32, P < 0·0001) in the FIB index, basedon total landings and the mean trophic level (Fig. 4c).

A highly significant increase in the RPI was observedover the whole data set from 1979 to 2000 (Mann–Kendall Z = 3·16, P = 0·016), indicating that hightrophic level species had become more valuable inrelation to species feeding at lower trophic levels(Fig. 5). Sen’s non-parametric estimator of slope indi-cated that the RPI had increased by around 0·03 year−1.

From 1979 until 1984 the RPI changed little. From1984 onwards, however, the relative distribution ofprices began to change more markedly, and in 1985prices of many low trophic level species, e.g. Trachurustrachurus, Eutrigla gurnardus (Linnaeus), Lepidorhombuswhiffiagonis, Clupea harengus Linnaeus and Scomberscombrus, declined in absolute terms (by 35%, 28%,25%, 19% and 11% in relation to prices in the previousyear). Prices of many high trophic level species weremuch higher in 1985 compared with 1984. Thesespecies included Pollachius virens (Linnaeus), Mullussurmuletus Linnaeus, Merlangius merlangus Linnaeus,Galeorhinus galeus (Linnaeus), Zeus faber Linnaeus, Molvamolva (Linnaeus) and Gadus morhua Linnaeus, whichincreased by 105%, 42%, 32%, 28%, 29%, 20% and 15%,respectively (CEFAS, unpublished data).

Discussion

Our study clearly demonstrates that there have beensignificant changes in the structure of fish communitiesand fishery landings in the Celtic Sea over the past50 years, and these have coincided with a period ofconsiderable fishery expansion. The response ofmarkets has been a marked switch away from hightrophic level, high price, species, to greater demand forlow trophic level, low price, species.

Our results support the assertion of Pauly et al.(1998a) that there have been major changes in the fishcommunities of the north-east Atlantic generally. Thedecline we observed in the survey data (0·04 TL year−1)was slightly stronger than the decline we observed inthe landings data (0·03 TL year−1), and in both casesthis decline was more marked than that given by Paulyet al. (1998a) for the north-east Atlantic as a whole(c. 0·02 TL year−1). The fact that we found a significantdecline in the fishery-independent survey data overthe past 19 years, when there was no apparent trendin the landing data over this same period, suggests thatthe changes occurring in the underlying ecosystem maybe stronger than any changes observable in fisherypreferences, as demonstrated by Christensen (1998).

Our trophic level estimates for the 48 fish species(Table 1) in many cases differed from those used byPauly et al. (1998a). There are several possible explana-tions for this phenomenon, including the use of‘default’ values by Pauly et al. (1998a) to representcertain species. Of the 48 species for which we haveisotope-derived trophic level estimates (Table 1), 18estimates in Pauly et al. (1998a) were allocated atrophic level of 3·5, the default for ‘demersal per-comorphs’, five were given the default value for‘gadiformes’ (TL = 3·8), and two were given thedefault for ‘clupeiformes’ (TL = 2·8). Most of thesedefault values were obtained from Pauly & Christensen

Fig. 5. Pattern of change in the price distribution of 26 species of fish over a 22-year time-series for the Celtic Sea, expressed asa relative price index (RPI).



(1995). The use of default values was necessary becauseno locally derived estimates for the particular specieswere available, thus values for related species fromaround the world were averaged and utilized as asubstitute. Feeding behaviour, and therefore trophiclevel, can range markedly within most fish families, forexample among the gadoids where blue whiting areplanktivorous but cod and hake are primarily pisci-vorous. Thus the use of default values by Pauly et al.(1998a) to represent species groups may lead to unre-alistic conclusions concerning the trophic structure ofthe fish assemblage. We have provided species-specificand geographically relevant trophic level estimates for99·7% (by biomass) of the Celtic Sea fish community(based on the 2000 survey data).

Another possible explanation for the discrepanciesbetween our trophic level estimates and those of Paulyet al. (1998a) stems from the different abilities of gutcontents analysis vs. stable isotope analyses. In thepresent study, several species of flatfishes and mullet(notably Solea solea (Linnaeus), Limanda limanda,Hippoglossoides platessoides Fabricius and Mullussurmuletus) possessed higher trophic levels than mighthave been expected for species feeding primarily onbenthic invertebrates (Table 1). Soft bottom ecosystemscan be very complex, such that many polychaete species,for example, feed exclusively on other polychaetes,thus giving them a high trophic level (Jennings et al.2001). Some soft bottom fish species are known toselect these carnivorous polychaetes in particular(Badalamenti et al. 2000), and such subtleties arerarely incorporated in ecosystem models such as thoseused by Pauly et al. (1998a), or appreciated from gutcontents studies. Similarly, the anchovy Engraulisencrasicholus (Linnaeus) was suggested to feed at ahigh trophic level (4·17) according to isotope analysis(Table 1). This may be related to this species’ preferencefor smaller zooplankton species in comparison withlarger zooplanktivorous fish species (e.g. boarfish,mackerel and horse-mackerel). Microzooplanktontend to be part of complex food webs involving manymicrobial trophic levels (Azam et al. 1983), whereaslarge calanoid copepods typically feed directly ondiatoms and dinoflagellates (plants); again these sub-tleties are rarely appreciated from gut contents studies.Both our analysis and that of Pauly et al. (1998a) makethe assumption that trophic levels of species are stablefrom year to year. In the absence of time-series isotopedata, we are unable to assess whether this is an acceptableassumption, and it is possible that the diets of somespecies may have changed markedly over the course ofthe various time-series (Wainright et al. 1993).

A possible explanation for the more rapid decline introphic level of survey catches than in fishery landingsover recent years might be that some of the moreabundant species in the Celtic Sea are not the targetsof fisheries. Capros aper, for example, is a small plank-tivorous species that accounts for much of the biomassof low trophic level species in survey catches but has no

commercial value. Discards and by-catch are reportedto be substantial in the Celtic Sea, and may represent asmuch as one-third of total catches in some years(Rochet, Trenkel & Péronnet 2001).

In general, total landings from the Celtic Sea havecontinued to increase over the past 50 years. This is notthe case elsewhere in the north-east Atlantic (e.g.North Sea), where most stocks have been exploited fora considerable period of time. After the Second WorldWar, Celtic Sea fisheries targeted mainly gadoids, butthroughout the 1960s and 1970s new fisheries openedup that targeted pelagic species. Mackerel stocks wereexploited at a low level before the Second World War;however, in the mid-1960s a major winter fishery beganadjacent to the Cornish peninsula, operated by theUSSR (Lockwood & Shepherd 1984). The fishery forhorse-mackerel developed as a potential alternative tomackerel in the late 1960s (Lockwood & Johnson1977), again largely operated by the USSR, with smallcatches by UK, French, Norwegian, Polish, Spanishand Portuguese vessels. With the exclusion of the East-ern Bloc countries in 1977, the dominant pelagic fleetbecame that of the Netherlands, and their landings ofhorse-mackerel rose from 2000 tones to more than40 000 tonnes in 1981 (Eaton 1983). Prior to 1974,there was very little fishing by any country on bluewhiting Micromesistius poutassou, however, in the late1970s Norwegian industrial fisheries targeted thisspecies, largely for use as fishmeal (Pawson 1979).

The increased importance of small pelagic species inCeltic Sea landings may, in part, be a response to scar-city of more traditional target species such as cod, butit might also be related to technical innovation and thedevelopment of new fishing gear designs. The incor-poration of synthetic fibres into fishing gears in the1950s and 1960s allowed large-scale mid-water trawlingand purse-seining by industrial fleets for the first time.This, in turn, greatly increased the vulnerability of smallpelagic stocks world-wide (Caddy & Garibaldi 2000).

Almost all the new Celtic Sea fisheries of the last30 years or so (blue whiting, mackerel, horse-mackerel)have targeted species that feed mainly on zooplankton(Sorbe 1980; Mehl & Westgård 1983; Ben-Salem1988). This is in contrast to the post-war fisheries,which targeted high trophic level species such as cod,hake, haddock and small sharks. Indeed, many of thesepredatory fishes feed primarily on those species whichare now being harvested by the fishery (Du Buit 1982;Olaso et al. 1996; Velasco & Olaso 1998), whichimplies that modern vessels are operating a full trophiclevel lower than their post-war counterparts. Even thenon-exploited species Capros aper, which has becomemuch more common in recent trawl surveys, is azooplanktivorous species (Macpherson 1979, 1981),

In the present analyses only fish catches and landingswere considered. It is clearly possible that if inverte-brates were included then the observed trends in meantrophic level might have been even more marked, assuggested by Pauly et al. (1998a). However, the relative



importance of invertebrate landings has changedremarkably little over recent decades, representing4·6% (c. 12 177 tonnes) in 1973 and 5·0% (c. 10 845tonnes) in 1998. This suggests that there has been verylittle diverting of effort from finfish capture to poten-tially profitable invertebrate species such as Nephropsnorvegicus (Linnaeus).

The observation that the FIB index significantlyincreased over the 53-year time-series (Fig. 3c),confirms that fisheries have expanded at a rate of morethan 150 tonnes year−1, the factor that would have beennecessary to result in no change in the FIB, assuming atrophic level decline of c. 0·03 year−1 and a 10-foldincrease in fishery productivity for each full trophiclevel decline (Pauly, Christensen & Walters 2000b).Christensen (2000) observed that the FIB index for thenorth-east Atlantic as a whole increased greatly upuntil 1976, but has subsequently declined, indicatingthat the decrease in mean trophic level is no longerbeing matched by a sufficient increase in landings.

There is very little evidence that low trophic level spe-cies (such as Capros aper) increase in abundance as aresult of an easing of predatory pressure (a top-downtrophic cascade effect) (Pinnegar et al. 2000). However,climate is thought to have a major impact on lowtrophic level pelagic fish abundances. Southward,Boalch & Maddock (1988) demonstrated that theabundance of herring Clupea harengus and pilchardSardina pilchardus (Walbaum) occurring off the south-west of England closely corresponded with fluctuationsin water temperature. Pilchard were generally moreabundant and extended further to the east whenclimate was warmer, whilst herring were generally moreabundant in cooler times. This pattern has apparentlybeen occurring for at least 400 years, and majorchanges were noted in the late 1960s as waters cooledand spawning of pilchard was inhibited. During thistime mackerel Scomber scombrus, another cold-waterarctic-boreal species, started to become very abundant,and these environmental changes coincided with thedevelopment of the mackerel and subsequently otherpelagic fisheries (Lockwood & Shepherd 1984; South-ward, Boalch & Maddock 1988). Changes in the dis-tribution of other Celtic Sea fish species have beenobserved to coincide with cold periods (Coombs 1975;Cushing 1982; Planque & Frédou 1999), includingdeclines in Merluccius merluccius, Trachurus trachurus,Lophius piscatorius, Mullus surmuletus, Conger conger(Linnaeus) and Pollachius pollachius (Linnaeus), whichare considered warm-water species (Cushing 1982), andincreases in Gadus morhua, Molva molva, Pleuronectesplatessa Linnaeus, Scyliorhinus canicula (Linnaeus)and Microstomus kitt, which are considered cold-waterspecies.

Since 1988, mean temperatures in the Celtic Sea,and north-east Atlantic generally, have been higher

than during the three previous decades, and this hascoincided with declines in the abundance of cold-waterspecies including cod and haddock. O’Brien et al.(2000) showed that a combination of reduced spawningstock biomass due to overfishing and warm conditionsresulting in reduced rates of recruitment were probablyresponsible for the recent near-collapse of cod stocks inthe North Sea. Many of the species that increase inabundance during cool periods (e.g. cod, dogfish andling) feed at high trophic levels, whilst many thatincrease during warm periods feed at lower trophiclevels (e.g. horse-mackerel, pilchard and Argentinaspp.). Thus warming can exacerbate apparent declinesin trophic level caused by fishing.

Capros aper is a species with a southerly distribution,and therefore we might expect increases in abundanceduring periods of warming, as has been observed here(Fig. 2c). Farina, Freire & Gonzalez Gurriaran (1997)also observed a recent increase in Capros aper at sites inthe Bay of Biscay. These authors suggested thatincreased productivity associated with upwelling mighthave been responsible for such an influx, and severalother non-commercial and low trophic level specieswere similarly observed to have increased over recentyears (e.g. Argentina sphyraena Linnaeus). MonkfishesLophius piscatorius and Lophius budegassa Spinola, bycontrast, declined in abundance, as was observed forthe Celtic Sea (Fig. 2c). Bakun (1990) has demon-strated that the warming of the north-east Atlantic,by intensifying wind stress over the ocean surface, hasled to an acceleration of coastal upwelling, which mayhave brought about the observed changes in demersal andpelagic fish distributions (Farina, Freire & GonzalezGurriaran 1997).

Caddy (2000a,b) and De Leiva Moreno et al.(2000) have proposed that pelagic resources of thenorth-east Atlantic have greatly benefited fromanthropogenic nutrient inputs in recent years, andthat Pauly et al. (1998a) may have underestimated theimportance of bottom-up effects, particularly withreference to semi-enclosed seas. Eutrophication mayrender benthic habitats less suitable for demersal pred-atory and piscivorous species (e.g. cod), due toreduced oxygen levels in bottom waters. This cancontribute to the establishment of an ecosystemdominated by small pelagic species (De LeivaMoreno et al. 2000). The Irish Sea (ICES area VII a),which adjoins the Celtic Sea (Fig. 1), was specificallyhighlighted by De Leiva Moreno et al. (2000)because plankton productivity has increased in thisregion markedly. Although the Celtic Sea is lessenclosed, and is open to the north-east Atlantic, wecannot discount the possibility that eutrophicationmight also have contributed to the long-term trendswe observed in this region. Capros aper has recentlyappeared in much greater numbers in the western Medi-terranean (Abad & Giráldez 1990), a semi-enclosedsea, highly impacted by nutrient inputs (Caddy &Garibaldi 2000).



Fish prices are greatly influenced by changes in con-sumer income, current tastes or preferences and theprice of alternative products on the market. They arealso influenced by the ability of suppliers (fishermen) toland desirable fish, which in turn is reliant upon theavailability of target species in the environment,capture restrictions and/or quotas, technologicalinnovation as well as the weather (Lawson 1984;Ludicello, Weber & Wieland 1999). Analysis of our 22-year time-series of market prices for Celtic Sea fishdemonstrated that high trophic level species havebecame relatively more expensive in comparison withlow trophic level species, i.e. the steepness of the slopein the relationship between trophic level and price (theRPI) has increased. This is evidently because theexpanding pelagic fisheries flooded the market withlow trophic level fish and forced their prices down (e.g.horse-mackerel prices fell by 35% between 1984 and1985). Conversely, many high trophic level fish, e.g.hake, cod, haddock, monkfish and small sharks, havebecome scarce (supply has declined but demand remainshigh) and their prices have risen. Similar trends havebeen observed on a world-wide basis and, as a group,the price of cod, hake and haddock (ISSCAAP group32) rose from $714 per tonne in 1989 to $1080 per tonnein 1994 (accounting for inflation), reflecting world-widedeclines in their abundance (OECD 1997; Ludicello,Weber & Wieland 1999). Sumaila (1998) suggested that,globally, low trophic level species have become morevaluable in relation to high trophic level species over thepast 50 years, which would result in a declining globalRPI. Our analysis does not support this theory for theCeltic Sea; however, it is possible that in regions with alonger history of intensive exploitation the RPI mightbegin to decline, as surmised. The decline in the meantrophic level of landings we observed suggests somesubstitution has occurred, but this has not yet greatlyaffected fish prices. Increases in the price of previouslyless desirable species (but still high trophic level), suchas saithe Pollachius virens, relative to other gadoids,does, however, point towards changes in the targetingof species within some ISSCAAP groups.

Had there been no substitution and the relativedistribution of prices determined only by the scarcityof different species in the environment, we would nothave expected the observed levelling off of RPI after1987 (Fig. 6). In Fig. 6, we hypothesize that at somepoint in time or level of fishing effort (P), the relation-ship between fish prices and their abundance in theenvironment becomes uncoupled. This phenomenonwould occur when the price of high trophic level speciesincreases so much that consumers increasingly opt foralternative protein sources, either low trophic levelfishes or non-fish products. In the Celtic Sea, it is notclear whether the expansion of pelagic fisheries, and thusthe levelling out in RPI, was primarily a response toscarcity of traditional high trophic level target species,

or was due to environmental change which, in turn, led toincreases in pelagic stocks, thus making the fisheries viablefor the first time, or indeed whether the fishery only becameviable because of improvements in gear technology.

In the past fish prices were largely governed by localpatterns of demand and supply (e.g. cod and haddockin the UK; Taylor 1960), thus the RPI would be a usefulindicator of trends in the underlying ecosystem.With increasing globalization, however, prices, par-ticularly of high value species, tend to reflect relativeabundance or scarcity on global markets. Asche,Bremnes & Wessells (1999) found that salmon priceswere globally integrated, and that different salmon spe-cies (from the US and Norway) exhibited similar trends,irrespective of national boundaries. Hannesson (1999)reported similar integration among whitefish prices (e.g.cod and haddock) across US and European markets.

Our analysis indicates that fisheries typically willtarget species at high trophic levels. Although decreasedabundance of high trophic level species will ultimatelyhave negative economic consequences, the reduction inmean trophic level of the fish community may allow thesystem to sustain higher fishery yields overall. Manage-ment objectives for a particular fishery will depend on therelative value that society attaches to either economicprofit or protein production. In developed western societiessuch as those surrounding the Celtic Sea, fish areprimarily viewed as a luxury food and managers are

Fig. 6. Conceptual model illustrating how the relative priceindex (RPI) might develop over time, assuming increasingfishing effort. (1) Prices of high trophic level species increase asthey become scarce, and there may be some substitutionwithin high trophic level species. (2) Increases in the price ofhigh trophic level species are partly offset by substitution withlow trophic level species, i.e. the increase in both cancels out,and thus the slope of the relationship between trophic leveland price (the RPI) stays the same. (3) High trophic level speciesbecome so scarce that they become commercially extinct. Thefishery switches to targeting low trophic level species, forcingthe price of these species up and thereby reducing the slope(RPI) until eventually negative. (4) Prices of low trophic levelspecies become higher than prices of higher trophic levelspecies, e.g. Sumaila (1998). Therefore RPI < 0.



more likely to restrict fishing mortality on high trophiclevel species in an attempt to maximize long-term eco-nomic yields (e.g. by catch or effort control). This is lesslikely to be the case in developing nations where fishremain an essential source of protein (Kent 1998).

Acknowledgements

We are particularly grateful to Steve Warnes for help-ing with groundfish survey data, Gillian Taylor of theBiomedical Mass-Spectrometry Unit, NewcastleUniversity, for providing the stable isotope analysis,and Trevor Hutton for advice and discussion regardingfishery economics. This work was funded by theEuropean Community under Framework V, projectcontract QLRT-1999-01609 (development of structur-ally detailed statistically testable models of marinepopulations, dst2), and the UK Ministry of Agriculture,Fisheries and Food contract MF0316.

References

Abad, R. & Giráldez, y A. (1990) Concentrations de Capros apersans la Mer d’Alboran (Mediterranée Espagnole). RapportCommission Internationale Pour la Mer Mediteranée, 32, 256.

Asche, F., Bremnes, H. & Wessells, C.R. (1999) Productaggregation, market integration, and relationships betweenprices: an application to world salmon markets. AmericanJournal of Agricultural Economics, 81, 568–581.

Azam, F., Fenchel, T., Field, J.G., Gray, J.S., Meyer-Reil, L.A.& Thingstad, F. (1983) The ecological role of water-columnmicrobes in the sea. Marine Ecology-Progress Series, 10,257–263.

Badalamenti, F., Pinnegar, J.K., Polunin, N.V.C. &D’Anna, G. (2000) Estimates of trophic level in the red mul-let Mullus barbatus: comparison between gut-contents andstable-isotope data. CIESM Workshop Series, 12, 19–21.

Bakun, A. (1990) Global climate change and intensificationof coastal ocean upwelling. Science, 247, 198–201.

Ben-Salem, M. (1988) Régime alimentaire de Trachurustrachurus (Linnaeus, 1758) et de T. mediterraneus (Stein-dachner, 1868), (poissons, téléosteens, carangidae) de laprovince Atlantic-Méditerranéenne. Cybium, 12, 247–253.

Bowman, R.E. (1986) Effect of regurgitation on stomachcontent data of marine fisheries. Environmental Biology ofFishes, 16, 171–181.

Cabana, G. & Rasmussen, J.B. (1996) Comparison ofaquatic food chains using nitrogen isotopes. Proceedingsof the National Academy of Sciences of the United Statesof America, 93, 10844–10847.

Caddy, J.F. (2000a) Ecosystem structure before fishing?Fisheries Research, 47, 101.

Caddy, J.F. (2000b) Marine catchment basin effects versusimpacts of fisheries on semi-enclosed seas. ICES Journal ofMarine Science, 57, 628–640.

Caddy, J.F. & Garibaldi, L. (2000) Apparent changes in thetrophic composition of world marine harvests: the perspectivefrom the FAO capture database. Ocean and Coastal Man-agement, 43, 615–655.

Caddy, J.F., Csirke, J., Garcia, S.M. & Grainger, R.J.L.(1998) How pervasive is ‘fishing down marine food webs’?Science, 282, 1383–1384.

Christensen, V. (1995) A model of trophic interactions in theNorth Sea in 1981, the year of the stomach. Dana, 11, 1–28.

Christensen, V. (1998) Fishery-induced changes in a marineecosystem: insight from models of the Gulf of Thailand.Journal of Fish Biology, 53 (Supplement A), 128–142.

Christensen, V. (2000) Indicators for marine ecosystems affectedby fisheries. Marine and Freshwater Research, 51, 447–450.

Coombs, S.H. (1975) Continuous plankton records showfluctuations in larval fish abundance during 1948–72.Nature, 258, 134–136.

Cushing, D.H. (1982) Climate and Fisheries. Academic Press,London.

Deb, D. (1997) Trophic uncertainty vs parsimony in food webresearch. Oikos, 78, 191–194.

De Leiva Moreno, J.I., Agostini, V.N., Caddy, J.F. &Carocci, F. (2000) Is the pelagic-demersal ratio from fisherylandings a useful proxy for nutrient availability? A prelim-inary data exploration for the semi-enclosed seas aroundEurope. ICES Journal of Marine Science, 57, 1091–1102.

Du Buit, M.H. (1982) Essai d’évaluation de la prédation dequelques téléostéens en Mer Celtique. Journal of Du ConseilInternational Pour l’Exploration de la Mer, 40, 37–46.

Eaton, D.R. (1983) Scad in the North-East Atlantic. Labor-atory Leaflet No. 56. Ministry of Agriculture, Fisheries andFood, Directorate of Fisheries Research, Lowestoft, UK.

Farina, A.C., Freire, J. & Gonzalez Gurriaran, E. (1997) Demersalfish assemblages in the Galician continental shelf andupper slope (NW Spain): spatial structure and long-termchanges. Estuarine, Coastal and Shelf Science, 44, 435–454.

Gilbert, R.O. (1987) Statistical Methods for Environmental Pol-lution Monitoring. Van Nostrand Reinhold, New York, NY.

Greenstreet, S.P.R. & Hall, S.J. (1996) Fishing and ground-fishassemblage structure in the north-western North Sea: ananalysis of long-term and spatial trends. Journal of AnimalEcology, 65, 577–598.

Gulland, J.A. (1987) The effect of fishing on community struc-ture. South African Journal of Marine Science, 5, 839–849.

Hannesson, R. (1999) Globalisation of seafood prices: someexamples. ACP-EU Fisheries Research Report, 5, 15–16.

Hesslein, R.H., Hallard, K.A. & Ramal, P. (1993) Replace-ment of sulphur, carbon and nitrogen of growing broadwhitefish (Coregonus nasus) in response to a change in diettraced by δ13C, δ15N and δ34S. Canadian Journal of Fisheriesand Aquatic Sciences, 50, 2071–2076.

Horwood, J. (1993) The Bristol channel sole Solea solea (L.):a fisheries case study. Advances in Marine Biology, 29, 215.

ICES (1906–90) Bulletin of the Statistique Des PêchesMaritimes 1903–1987, Vols 1–72. International Council forthe Exploration of the Sea. Copenhagen, Denmark.

Jennings, S., Greenstreet, S.P.R. & Reynolds, J.D. (1999)Structural change in an exploited fish community: a conse-quence of differential fishing effects on species with contrastinglife histories. Journal of Animal Ecology, 68, 617–627.

Jennings, S., Pinnegar, J.K., Polunin, N.V.C. & Boon, T. (2001a)Weak cross-species relationships between body size andtrophic level belie powerful size-based trophic structuring infish communities. Journal of Animal Ecology, 70, 934–944.

Jennings, S., Pinnegar, J.K., Polunin, N.V.C. & Warr, K.J.(2001b) Impacts of trawling disturbance on the trophicstructure of benthic invertebrate communities. MarineEcology-Progress Series, 213, 127–142.

Kent, G. (1998) Fisheries, food security and the poor. FoodPolicy, 22, 393–404.

Kline, T.C. & Pauly, D. (1998) Cross-validation of trophiclevel estimates from a mass-balance model of PrinceWilliam Sound using 15N/14N data. Fishery Stock AssessmentModels Alaska Sea Grant College Program (eds F. Funk,T. J. Quinn II, J. Heifetz, J. E. Powers, J. F. Schweigert,P. J. Sullivan & C.-I. Zhang), pp. 693–702. Alaska Sea GrantCollege Program, Fairbanks, AK.

Laurec, A., Biseau, A. & Charuau, A. (1991) Modelling technicalinteractions. ICES Marine Science Symposia, 193, 225–236.

Lawson, R. (1984) Economics of Fisheries Development.Frances Pinter, London, UK.

Lockwood, S.J. & Johnson, P.O. (1977) Horse MackerelTrachurus trachurus. Laboratory Leaflet No. 38. Ministry



of Agriculture, Fisheries and Food, Directory of FisheriesResearch, Lowestoft, UK.

Lockwood, S.J. & Shepherd, J.G. (1984) An assessment ofthe Western mackerel stock. Journal of Du Conseil Inter-national Pour l’Exploration de la Mer, 41, 167–180.

Ludicello, S., Weber, M. & Wieland, R. (1999) Fish, Marketsand Fishermen: The Economics of Overfishing. Earthscan,London, UK.

Macpherson, E. (1979) Estudio sobre el régimen alimentariode algunos peces en el Mediterráneo occidental. MiscelaneaZoologica, 5, 93–107.

Macpherson, E. (1981) Resource partitioning in a Mediter-ranean demersal fish community. Marine Ecology-ProgressSeries, 4, 183–193.

MAFF (1999) United Kingdom Sea Fisheries Statistics 1998.HM Stationery Office, Norwich, UK.

Marchal, P. & Horwood, J. (1996) Long-term targets for theCeltic Sea mixed-species multi-metiers fisheries. AquaticLiving Resources, 9, 81–94.

Mehl, S. & Westgård, T. (1983) The Diet and Consumption ofMackerel in the North Sea. ICES CM 1983/H: 34. Interna-tional Council for the Exploration of the Sea. Copenhagen,Denmark.

Minagawa, M. & Wada, E. (1984) Stepwise enrichment of 15Nalong food chains: further evidence and the relationbetween δ15N and animal age. Geochimica CosmochimicaActa, 48, 1135–1140.

Morizur, Y., Berrow, S.D., Tregenza, N.J.C., Couperus, A.S.& Pouvreau, S. (1999) Incidental catches of marine-mam-mals in pelagic trawl fisheries of the northeast Atlantic.Fisheries Research, 41, 297–307.

Murawski, S.A. & Serchuk, F.M. (1989) Mechanized shellfishharvesting and its management: the offshore clam fishery ofthe eastern United States. Marine Invertebrate Fisheries:Their Assessment and Management (ed. J. F. Caddy),pp. 479–506. John Wiley & Sons, New York, NY.

O’Brien, C.M., Fox, C.J., Planque, B. & Casey, J. (2000)Climate variability and North Sea cod. Nature, 404, 142.

OECD (1997) Towards Sustainable Fisheries: Economic Aspectsof The Management of Living Marine Resources. Organisationfor Economic Co-operation and Development, Paris, France.

Olaso, I., Velasco, F., Pereda, P. & Perez, N. (1996) Importanceof Blue Whiting (Micromesistius poutassou) Discarded inthe Diet of Lesser-Spotted Dogfish (Scyliohinus canicula) inthe Cantabrian Sea. ICES CM 1996/Mini:2. InternationalCouncil for the Exploration of the Sea. Copenhagen, Denmark.

Pauly, D. & Christensen, V. (1995) Primary productionrequired to sustain global fisheries. Nature, 374, 255–257.

Pauly, D. & Palomares, M.L. (2000) Approaches for dealing withthree sources of bias when studying the fishing down marinefood web phenomenon. CIESM Workshop Series, 12, 61–66.

Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Torres, F.(1998a) Fishing down marine food webs. Science, 279,860–863.

Pauly, D., Christensen, V., Froese, R. & Palomares, M.-L.(2000a) Fishing down aquatic food webs. American Scien-tist, 88, 46–51.

Pauly, D., Christensen, V. & Walters, C. (2000b) Ecopath,Ecosim and Ecospace as tools for evaluating ecosystem impactof fisheries. ICES Journal of Marine Science, 57, 697–706.

Pauly, D., Froese, R. & Christensen, V. (1998b) Response toCaddy et al.: how pervasive is ‘fishing down marine foodwebs’. Science, 282, 1384–1386.

Pauly, D., Palomares, M.A., Froese, R., Sa-a, P., Vakily, M.,Preikshot, D. & Wallace, S. (2001) Fishing down Canadianaquatic food webs. Canadian Journal of Fisheries andAquatic Science, 58, 51–62.

Pawson, M.G. (1979) Blue Whiting. Laboratory Leaflet No.45. Ministry of Agriculture, Fisheries and Food, Directorateof Fisheries Research, Lowestoft, UK.

Perez, N., Trujillo, V. & Pereda, P. (1996) Discards of the

Trawl and Long Line Spanish Fleets in ICES Subarea VII in1994. ICES CM 1996/Mini:8. International Council for theExploration of the Sea.Copenhagen, Denmark.

Pinnegar, J.K. (2000) Planktivorous fishes: links between theMediterranean littoral and pelagic. PhD Thesis. Universityof Newcastle, Newcastle upon Tyne, UK.

Pinnegar, J.K. & Polunin, N.V.C. (1999) Differential fraction-ation of δ13C and δ15N among fish tissues: implications for thestudy of trophic interactions. Functional Ecology, 13, 225–231.

Pinnegar, J.K., Polunin, N.V.C., Francour, P., Badalamenti,F., Chemello, R., Harmelin-Vivien, M.-L., Hereu, B.,Milazzo, M., Zabala, M., D’Anna, G. & Pipitone, C. (2000)Trophic cascades in benthic marine ecosystems: lessons forfisheries and protected-area management. EnvironmentalConservation, 27, 179–200.

Planque, B. & Frédou, T. (1999) Temperature and the recruit-ment of Atlantic cod (Gadus morhua). Canadian Journal ofFisheries and Aquatic Science, 56, 2069–2077.

Polunin, N.V.C. & Pinnegar, J.K. (2002) Trophic ecology andthe structure of food webs. Handbook of Fish and Fisheries,Vol. 1 (eds P. J. B. Hart & J. D. Reynolds). Blackwell,Oxford, UK.

Post, D.M., Pace, M.L. & Hairston, N.G. (2000) Ecosystemsize determines food-chain length in lakes. Nature, 405,1047–1049.

Rochet, M.-J., Trenkel, V. & Péronnet, I. (2001) EstimationDes Rejets de Pêche Des Chalutiers Français En Mer Celtique.IFREMER Report DRV/RST/RH-22001–07. InstitutFrancais de recherche pour l’exploitation de la mer,Laboratoire Mathématique Appliquées à l’exploitation desResources Halieutiques et Aquacoles, Nantes, France.

Sen, P.K. (1968) Estimates of the regression coefficient basedon Kendall’s tau. Journal of the American Statistical Asso-ciation, 63, 1379–1389.

Sorbe, J.-C. (1980) Regime alimentaire de Micromesistius pou-tassou (Risso, 1826) dans le sud du Golfe de Gascogne. RevueDes Travaux Institut Des Pêches Maritimes, 44, 245–255.

Southward, A.J., Boalch, G.T. & Maddock, L. (1988) Fluc-tuations in the herring and pilchard fisheries of Devon andCornwall linked to change in climate since the 16th century.Journal of the Marine Biological Association of the UnitedKingdom, 68, 423–445.

Sumaila, U.R. (1998) Markets and the fishing down marinefood webs phenomenon. EC Fisheries Cooperation Bulletin,11, 25–28.

Taylor, R.A. (1960) The Economics of Whitefish Distributionin Great Britain. Gerald Duckworth and Co., London, UK.

Treganza, N.J.C., Berrow, S.D., Hammond, P.S. & Leaper,R. (1997) Harbour porpoise (Phocoena phocoena) by-catchin set gill nets in the Celtic Sea. ICES Journal of Marine Sci-ence, 54, 897–904.

Vander Zanden, M., Cabana, G. & Rasmussen, J.B. (1997)Comparing trophic position of freshwater fish calculatedusing stable nitrogen isotope ratios (δ15N) and literaturedietary data. Canadian Journal of Fisheries and AquaticScience, 54, 1142–1158.

Velasco, F. & Olaso, I. (1998) European hake Merluccius mer-luccius (L., 1758) feeding in the Cantabrian Sea: seasonal,bathymetric and length variations. Fisheries Research, 38,33–44.

Wainright, S.C., Fogarty, M.J., Greenfield, R.C. & Fry, B.(1993) Long-term changes in the Georges Bank food web –trends in stable isotopic compositions of fish scales. MarineBiology, 115, 481–493.

Warnes, S. & Jones, B.W. (1995) Species Distributions fromEnglish Celtic Sea Groundfish Surveys, 1984–1991. FisheriesResearch Technical Report No. 58. Ministry of Agriculture,Fisheries and Food, Directorate of Fisheries Research,Lowestoft, UK.

Received 15 May 2001; final copy received 22 February 2002

Documents

J. K. PINNEGAR, S. JENNINGS, C. M. O’BRIEN and N. V. C