a reprint from american scientist...daniel pauly, villy christensen, rainer froese and maria lourdes...

A reprint from

American Scientistthe magazine of Sigma Xi, The Scientific Research Society

This reprint is provided for personal and noncommercial use. For any other use, please send a request to Permissions,American Scientist, P.O. Box 13975, Research Triangle Park, NC, 27709, U.S.A., or by electronic mail to [email protected].©Sigma Xi, The Scientific Research Society and other rightsholders

Roll on, thou deep and dark blueOcean–roll!

Ten thousand fleets sweep over thee in vain;

Man marks the earth with ruin–his control

Stops with the shore

When Lord Byron wrote theselines, nearly 200 years ago, ex-

ploitation of the seas was already wellunder way, and some marine species,such as right whales, were on the pathto extinction. Yet the poet’s verse sup-poses that the sea is immune from en-vironmental degradation, a commonmisconception then and now. Peoplecan, in fact, ruin the sea as surely asthey can ruin the land. The only differ-ence is that ecological destruction inthe ocean is harder to see, particularlywhen the damage is inflicted on the

delicate and largely invisible web ofmarine life.

If one could don magic spectacles—with lenses that make the murkydepths of the ocean become transpar-ent—and look back several centuries toan age before widespread abuse of theoceans began, even the most casual ob-server would quickly discover that fishwere commonly much more abundant.And many now-depleted species ofmarine mammals (a list that includesnot just right whales but Hector’s dol-phins, Caribbean monk seals andSteller sea lions, among others) would,by comparison, appear plentiful. Butwithout such special glasses, the differ-ences between past and present oceanswould indeed be hard to discern.

The opacity of the sea thus accounts,in part, for the disregard that most peo-ple display toward the warnings thatwe and others have been trying to con-vey. But the remoteness and impene-trability of this habitat are not the onlyimpediments to our efforts. We havealso been frustrated by the very rich-ness of the ocean, in that it provides aseemingly never-ending supply ofseafood. Indeed, with only rare excep-tions, the total catch from global fish-eries justs keep going up.

The premier source for such statisti-cal information about world fisheries isthe Food and Agriculture Organization,one of the technical organizations of theUnited Nations. Surveys compiled bythe FAO are perhaps not as accurate oras detailed as many scientists wouldlike, but they are in most cases all that isavailable. Two years ago, we examineda large set of FAO statistics on fisheriesand showed that they do in fact por-tend disaster, despite the continuinggrowth in total amount being takenfrom the sea each year. The alarm ringswhen one realizes that the kinds of ani-

mals being caught have been changingin a fundamental way: In essence,rather than being satisfied with the bigfish, people are now routinely going af-ter many of the little ones too. Al-though the situation is not exactly anal-ogous to eating one’s seed corn, theresult may be equally catastrophic.

Little fish serve the marine ecosystemin various ways, notably as prey for big-ger fish. One straightforward way to de-scribe the feeding hierarchy uses the no-tion of trophic level. By convention,marine plants (such as the tiny, driftingphytoplankton) and various forms of or-ganic detritus make up the first trophiclevel; herbivores and detritivores are as-signed to the second trophic level; andthe first and higher-order carnivores aresaid to occupy trophic levels rangingfrom three to five.

More precisely, the trophic level ofsuch predators can take on non-inte-gral values, because the diet of theseanimals is commonly somewhatmixed. For example, an adult jackswimming around the Caribbeanmight eat equal amounts of herbivo-rous zooplankton (tiny plant-eating an-imals, which have a trophic level oftwo) and small fish that have a trophiclevel of three (say, because they con-sume only the local herbivorous zoo-plankton). This jack would then belongto trophic level 3.5.

Unfortunately, the trophic level of acaptured fish is not stamped on its side.So it is not an easy task to track the av-erage trophic level of all the fish caughteach year. But after modeling the foodwebs of many marine ecosystems, werealized that we had accumulated agreat deal of knowledge about the ap-proximate trophic levels of commonlyfished species. All that was needed wasto combine these estimates with FAOcatch statistics collected since the 1950s.

46 American Scientist, Volume 88© 2000 Sigma Xi, The Scientific Research Society. Reproduction

with permission only. Contact [email protected].

Fishing Down Aquatic Food Webs

Industrial fishing over the past half-century has noticeably depleted the topmost links in aquatic food chains

Daniel Pauly, Villy Christensen, Rainer Froese and Maria Lourdes Palomares

Daniel Pauly is a professor at the Fisheries Centreof the University of British Columbia. He receivedhis doctorate from the University of Kiel in 1979.His research focuses on the management of fish-eries and the modeling of aquatic ecosystems. VillyChristensen leads the Fisheries Resources Assess-ment and Management Program at the Interna-tional Center for Living Aquatic Resources Man-agement in Manila. He earned a Ph.D. in 1992from the Danish Royal School of Pharmacy. Rain-er Froese is leader of the Fishbase Project at the In-ternational Center for Living Aquatic ResourcesManagement. He received his doctorate from theUniversity of Hamburg in 1990. Froese’s currentwork includes the design of large biological data-bases and the analysis of the information con-tained therein. Maria Lourdes Palomares alsoworks on the Fishbase Project. She earned her doc-torate from the Ecole Nationale SupérieureAgronomique de Toulouse in 1991. Her researchinterests cover the comparative studies of the biol-ogy of fishes. Address for Pauly: Fisheries Centre,University of British Columbia, 2204 Main Mall,Vancouver, B.C., Canada V6T 1Z4. Internet:[email protected]

Doing so uncovered a systematicshift in the composition of global cap-ture fisheries, with the average trophiclevel showing a slow slide downwardover the past half-century. To judge thesignificance of our result requires anunderstanding of the procedures weemployed, of the statistics underlyingour analysis and of the various con-cerns that have been raised since ourstudy was first published.

On the LevelHistorically, the notion of assigningtrophic levels in ecology passedthrough two stages. During the firstphase, initiated by Charles Elton of theUniversity of Oxford and RaymondLindeman of Yale University, trophiclevels were seen as pre-defined cate-gories into which organisms could be

pigeonholed—perhaps shoehorned isthe better metaphor, given that com-plex feeding habits could not readilybe accounted for. Fish that normallyconsume zooplankton were assigned atrophic level of three, because zoo-plankton (with a trophic level of two)normally consume phytoplankton(trophic level of one). This procedureignores the fact that some zooplanktonare carnivorous and thus should be as-signed a trophic level somewhat higherthan two, which implies that theirpredators should in turn be placed attrophic levels above three.

Such complications forced this over-ly simplistic scheme to give way in1975, when the late Frank H. Rigler, azoologist at the University of Toronto,published an influential critique of theconcept. It was then that William E.

2000 January–February 47© 2000 Sigma Xi, The Scientific Research Society. Reproduction


Figure 1. Overview of the Tsukiji fish market in Tokyo hints at the variety of offerings available to seafood lovers. But the mix of speciesbeing caught and sold around the world will probably not remain as rich as it is today. The authors have determined that the output of globalfisheries—and presumably the compositions of the aquatic ecosystems from which these fisheries draw—are slowly shifting away from toppredators toward organisms positioned lower on the food chain. This trend signals that many of today’s fisheries will in time collapse.

120

100

80

60

40

20

01950 1960 1970 1980 1990

year

catc

h (m

illio

ns o

f ton

s) global total

invertebrates

pelagic fish

demersal fish

discards

Figure 2. Global totals for the amount of fishcaught during the past half-century providea misleadingly reassuring view of the stateof the world’s fisheries. The increasing catchof small, free-swimming (pelagic) specieshas masked stagnation in take of bottom-dwelling (demersal) fish. (Source: UnitedNations Food and Agriculture Organization.)

James L. Stanfield/National Geographic Image Collection

Odum (of the University of Virginia’sDepartment of Environmental Sci-ences) and Eric J. Heald (of the Univer-sity of Miami’s Rosenstiel School ofMarine and Atmospheric Sciences)pointed out that more precise estimatesof trophic level could be obtained fromactual observations of diet. Their ad-vance, as followed in current practice,treats trophic level as an empiricallydeterminable property of a species—like average size or metabolic rate.

So how exactly do marine biologistsestimate trophic level for a givenspecies? One way is first to determinethe average trophic level of the variousthings that the organism eats and thenadd one to the value obtained. Follow-ing this formula, marine biologistsstudying Pacific bluefin tuna (Thunnusthynnus orientalis), a carnivore that con-sumes smaller fish and large inverte-brates such as squid, typically assignthe adults trophic levels ranging from4.2 to 4.5, despite wide variation fromplace to place in the exact compositionof their diets. Still, there can be sub-stantial shifts in the value of trophiclevel that apply to different points indevelopment of this and other species,because dramatic changes in size andbehavior take place during the life ofmost fishes.

Another complication is that it is of-ten quite hard to determine what allgoes into the stomach of a fish. This dif-ficulty can be overcome to a large de-gree by doing what we did: consideringthe organism as part of an integratedecosystem and modeling the web ofconnections between the animals andplants in enough detail to gain a reason-able understanding of what each crea-ture is consuming, at least on average.

A second method for estimatingtrophic level relies on a curious phe-nomenon of nitrogen biochemistry. Ni-trogen, like the other common build-ing blocks of life, comes in distinctforms. The most common isotope, ni-trogen-14, has seven protons and sevenneutrons in the nucleus. But a secondstable isotope with eight neutrons, ni-trogen-15, is also found in nature. In-terestingly, biochemical reactions dis-criminate to a small degree betweenthese two different kinds of atoms. Sothe nitrogen incorporated into the tis-sues of a sea creature does not have ex-actly the same isotopic ratio as is pre-sent in its food. Rather, the flesh of theanimal tends to collect the heavier iso-tope. (The same enrichment takes place



tro

ph

ic le

vel 1

tro

ph

ic le

vel 2

tro

ph

ic le

vel 3

tro

ph

ic le

vel 4

tro

ph

ic le

vel 5

1.0

2.0

0.0

2.2

3.2

3.0

phytoplankton

copepods

people

anchovies

people

squid

swordfish

marine trophic levels

Figure 3. Trophic levels were initially defined to include only discrete steps (left). Organicdetritus and microscopic plants (phytoplankton) occupy the first trophic level. Tiny zoo-plankton, which feed on phytoplankton, reside at the second level. Creatures that eat zoo-plankton sit at the third level, and so forth. But many marine creatures feed from multipletrophic levels and so could not be fit into this classic scheme. Thus the modern approachallows the assignment of trophic level to span a continuum rather than forcing it to take onintegral values. Marine biologists would, for example, assign the anchovy, which supple-ments its main diet of phytoplankton with some zooplankton, to a trophic level of about 2.2;people fishing for anchovies (and eating a diet of only these small fish) would then beassigned a trophic level of 3.2 (right).

among terrestrial organisms. In this re-spect, you are not what you eat.)

Measurements of the nitrogen fromvarious aquatic creatures have shownthat the isotopic ratio shifts by a rough-ly constant amount from one trophiclevel to the next—no matter whatspecies are involved. So the analysis ofnitrogen isotopes from an organismprovides a convenient way to ascertainits position in the food web withouthaving to know exactly what it con-sumes. All that is necessary is to com-pare the measured isotopic ratio withthat obtained from whatever organismsits at the bottom of the local foodchain. The difference in these two val-ues (divided by the constant differencebetween adjacent levels) immediatelyshows how far above base level thecreature should be placed.

Recently, we compared values oftrophic level determined in this way tothose we obtained from the more tradi-tional approach, using Prince WilliamSound in Alaska as a sample ecosys-tem. Happily, the two sets comparedfavorably (Figure 4). This result adds toour certainty that the 220 estimates oftrophic level that we used in our analy-sis of global fisheries, for which we em-ployed just one method (modeling thefood web), were largely correct.

We had, perhaps, less confidence inthe accuracy of the FAO statistics forthe catch from various parts of theworld—a perennial problem for scien-tists trying to understand the function-ing of global fisheries. Some of the con-cern arises from the breadth of thebrush that most countries use in report-ing their catches to the FAO. Althoughthe surveys sometimes track individualspecies (cod, for example), the statisticsoften lump multiple species togetherand report values only at the level ofgenera (such as for hake), families (her-ring and sardines) or even higher taxo-nomic groupings (all bony fishes).

Still, with no similarly comprehen-sive alternatives, we had no choice butto employ the FAO statistics, despitethis and other shortcomings. The re-sults were nevertheless clear and rea-sonably consistent (Figure 5). In mostplaces we looked, the average trophiclevel of the catches has declined overthe years. In the northwest Atlantic, forexample, trophic level plunged from apeak of nearly 3.7 in 1965 to 2.8 in 1997(the last year for which statistics areavailable), while in the northeast At-lantic it fell from about 3.6 to 3.4. In the

Mediterranean, the figure also dimin-ished, but it did so more gradually(slipping from about 3.1 to 3.0 duringthe same period).

Hold the AnchoviesExceptions to this general pattern havecertainly happened, but they are forthe most part easy to explain. For ex-ample, average trophic levels of thecatches from the South Atlantic and theWestern Pacific have climbed some-what during the past few decades—aneffect we ascribe to the development ofnew fisheries in these regions. And theaverage trophic level of the fish takenfrom the South Pacific rose sharplyduring the 1970s, from about 2.3 to 2.9.This increase corresponds to the col-lapse of a huge industrial fishery forPeruvian anchoveta, a species with amarkedly low trophic level (2.2), andthe growth of a more modest fisheryfor horse mackerel, which occupies aconsiderably higher position in the ma-rine food web (trophic level 3.3).

Indeed, the rise and fall of the Peru-vian anchoveta fishery creates a promi-nent dip in the curve for the global ma-rine catch, which otherwise shows alargely steady decrease from a value ofabout 3.4 in the early 1950s to less than3.1 today.

Might this worrisome trend bemerely some sort of statistical artifact?One could imagine, for example, thatthe decline we noted merely reflectschanges in the way the FAO collectedits information, say, by using onlycoarse groupings at first and refiningthe categories later. To test how our re-sults change when many species arelumped together, we regrouped the rel-atively detailed totals available for theNorth Atlantic, so that the accountingdistinguished different kinds of fishonly at high taxonomic levels. The re-sult of this exercise was to lessen thedecline seen. Hence the shift we com-puted with the full statistics must be, ifanything, underestimating the true ex-tent of the drop.



20151050deviation of N / N ratio (per mil)15 14

tertiary consumersLarus crassirostris

secondary consumersOctopusSebastes tririttatusSebastes taczanoskiiActiniaria

primary consumersHemigrapsus sanguineusHenriciaSeptifer virgatusMytilus edulisPolychaeta

primary producers and detritusdetritusZosteraSarugussum thunbergiiHizikia fusifoliaAlaria crassifoliaGloiopeltis furucata

Figure 4. Nitrogen isotopes provide an indi-rect means for determining trophic level.The ratio of the two stable isotopes of nitro-gen (14N and 15N) changes with position onthe food chain because organisms concen-trate the heavier isotope in their tissues. Thedifference in nitrogen isotope compositionfrom level to level is approximately con-stant, as Minagawa and Wada (1984) showedfor the waters off the coast of Hokkaido,Japan (above). The authors have used thismethod to check their more conventionalstrategy for assigning trophic level (by mod-eling the relevant food webs on a computer)for certain test cases, such as for PrinceWilliam Sound (right).

1

2

3

4

5

1 2 3 4 5nitrogen isotope estimate

conv

entio

nal e

stim

ate

Admittedly, the procedure of assign-ing a single trophic level to a speciesrests on a rather shaky scientific foun-dation, because the feeding habits ofmany fishes change as they grow.Some begin life eating only planktonand hop up a full trophic level whenthey begin devouring other fish. Otherspecies maintain roughly the same diet(and trophic level) throughout life. Stillother species slide backward a step onthe ladder, as larvae feeding on zoo-plankton and later turning to plants ordetritus for sustenance when theyreach maturity.

Fortunately, ignoring these compli-cations only makes our estimates oftrophic-level decline more conserva-tive. Why? Because some 86 percent ofthe worldwide marine catch is madeup of species that either maintain a con-stant diet throughout life or increase introphic level as they grow. In the firstcase, assigning a single trophic level

does not skew the results. The secondcase creates the possibility of bias, be-cause there is a tendency in many fish-eries to catch smaller and smaller spec-imens over time. But the actual trophiclevel of the catch from such a fisherywould typically be decreasing, whereasour analysis would show it to be con-stant. That is, our assigning of a singletrophic level to a species probably miss-es some of the real change that is takingplace. Here again, we conclude that thetrophic level of the global catch might,in actuality, be declining faster than wehad documented.

But what if this effect applies only tothe fish being netted, not to those re-maining in the wild? Perhaps marineecosystems are not changing at all,only people’s choice of what they ex-tract. Although such a selection bias isconceivable for lightly exploited fish-eries, we do not consider it likely to bea widespread phenomenon. Indeed, in

two cases one can test the hypothesisdirectly: in the Gulf of Thailand andover the Cantabrian Shelf, offshorefrom northern Spain. For the Gulf ofThailand, we determined the change inaverage trophic level (during periodsof intense industrial fishing) using sci-entific trawl surveys of fish popula-tions, rather than relying on the catchstatistics. Francisco Sanchez of the In-stituto Español de Oceanografia inSantander and several of his Spanishcolleagues did the same for the Can-tabrian Shelf. And both groups foundthat fishing pressure indeed altered themakeup of the ecosystem enough to de-press the average trophic level. So wefirmly believe that the average trophiclevel elsewhere is also truly declining.

It takes very little to convince one-self that this situation is alarming—forseafood lovers as well as for environ-mentalists. After all, the average troph-ic level of the global catch has alreadyslipped from 3.4 to 3.1 in just a fewdecades, and there are not many moreappetizing species to be found belowthis level. (Recall that 2.0 correspondsto copepods and other tiny zooplank-ton, creatures that are unlikely ever tobe filling one’s dinner plate.) So if thetrend continues, more and more re-gions are likely to experience completecollapse of their fisheries.

Faultfinders can argued that our in-terpretation may be overly gloomy.Maybe the shift toward lower values ofaverage trophic level simply reflects anincrease in number of plankton feedersin many places, perhaps species thatare now thriving from the coastalplankton blooms that take place whenfertilizer leached from farmers’ fields iscarried to the ocean by rivers andstreams. We have examined this possi-bility by comparing the decreases in av-erage trophic level with the increases inamount of seafood being caught. Insome places and times (for example,the Mediterranean Sea before 1986, orthe South Pacific during the 1960s), de-clining trophic levels are matched byappropriately large increases in theamount hauled in each year. But for themost part, diminishing trophic levelsdo not coincide with burgeoning popu-lations of plankton feeders. The overallpicture is indeed quite bleak.

Medium and MessageOur study has prompted many fish-eries scientists to review the statisticsthey have collected over the years for



19502.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

1960 1970 1980 1990year

mea

n tr

ophi

c le

vel

Mediterranean

19502.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

1960 1970 1980 1990year

northwest Atlanticm

ean

trop

hic

leve

l

19502.5

2.6

2.7

2.8

2.9

3.0

3.1

3.2

3.3

3.4

3.5

1960 1970 1980 1990year

global freshwater

19502.5

2.6

2.7

2.8

2.9

3.0

3.1

3.2

3.3

3.4

3.5

1960 1970 1980 1990year

global marine

Figure 5. Mean trophic level of the catch from many regions, including the Mediterranean Sea(upper left) and the northwest Atlantic Ocean (upper right) has declined over the past 50 years.The global mean for freshwater fisheries shows almost continuous decline (lower left), but thegeneral trend for marine fisheries (lower right) is complicated by the rise of the fishery for Peru-vian anchoveta in the late 1950s and its subsequent collapse in the early 1970s. Because an-chovies are assigned to a comparatively low trophic level, output from this productive fisherycreates a prominent dip in the record of global marine catch (blue line). Omitting statistics fromthe area containing this fishery produces a smooth decline (black line).

places of interest to them so that theycan judge whether these ecosystems aresuffering from a decline in trophic level.Our approach offers them a new toolfor assessing whether fishing in a par-ticular region is sustainable. But thepublication of our results also engen-dered considerable critique from somemembers of the fisheries community.We welcomed this scrutiny and havetried to provide others with the meansto examine the foundation of our analy-sis and to test the procedures we used.

As part of this same effort, we main-tain a site on the World Wide Web—Fishbase.org—to aid communicationamong interested fisheries scientistsand managers. We have also taken fulladvantage of the internet to dissemi-nate “Ecopath,” software we developedfor modeling the food webs of aquaticecosystems and determining such im-portant quantities as the trophic levelof various components.

We thus find ourselves manipulat-ing computer spreadsheets more oftenthan we would sometimes like. With-

out such work, we could never havemounted our study of global trophiclevels. But with today’s inexpensivepersonal computers and powerful soft-ware tools, the exercise has becomerather straightforward. Indeed, thetime was ripe to combine decades ofstatistical information about yearlycatches–the fodder of fisheries re-search—with the modeling of foodwebs, a branch of ecology that hadnow grown mature. We thus believethat the results of our investigationprovide a robust assessment of theshifts taking place in many regions ofthe globe—and for the world as awhole. Documenting the systematicdecline in trophic level exposes the im-mense influence that industrial-scalefishing has had on marine and fresh-waters ecosystems.

Clearly, we have skipped lightly hereover many of the details of individualfisheries in an effort to give a compre-hensive overview. Yet we see much val-ue in presenting such a broad summary,one that should raise awareness of the

scope of the changes that have takenplace in lakes, rivers, estuaries, coastalwaters and the open ocean. Perhaps ifmore people accept this message, fewerwill be lulled into thinking, with LordByron, that humankind’s ruinous wayshave marked only the earth.

BibliographyCaddy, J. F., J. Csirke, S. M. Garcia and R. J. R.

Grainger. 1998. How pervasive is “Fishingdown marine food webs”? Science 282:183.

Christensen, V. 1998. Fishery-induced changesin a marine ecosystem: Insights from mod-els of the Gulf of Thailand. Journal of Fish Bi-ology (supplement A) 53:128–142.

Cushing, D. H. 1998. The Provident Sea. Cam-bridge: Cambridge University Press.

Kline, T. C., Jr., and D. Pauly. 1998. Cross-vali-dation of trophic level estimates from a massbalance model of Prince William Sound us-ing 15N/14N data. In Proceedings of the Inter-national Symposium on Fishery Stock Assess-ment Models, ed. T. J. Quin II et al. Alaska SeaGrant College Program Report 98-01.

Lindeman, R. L. 1942. The trophic-dynamic aspect of ecology. Ecology 23:399–418.

Minagawa, M., and E. Wada. 1984. Stepwiseenrichment of 15N along food chains: Fur-ther evidence and the relation between δ15Nand animal age. Geochimica et Cosmochimica Acta 48:1135–1140.

Odum, E. P. 1969. The strategy of ecosystemdevelopment. Science 164:262–270.

Odum, W. E., and E. J. Heald. 1975. The detri-tus-based food web of an estuarine man-grove community. In Estuarine Research,(volume 1), ed. L. E. Cronin. New York:Academic Press.

Pauly, D. 1995. Anecdotes and the shiftingbaseline syndrome of fisheries. Trends inEcology and Evolution 10:430.

Pauly, D., and V. Christensen. 1995. Primaryproduction required to sustain global fisheries. Nature 374:255–257.

Pauly, D. V. Christensen, J. Dalsgaard, R. Froeseand F. Torres, Jr. 1998. Fishing down marinefood webs. Science 279:860–863.

Rigler, F. H. 1975. The concept of energy flowsand nutrients flows between trophic levels.In Unifying Concepts in Ecology, ed. U. W. H.van Dobben and R. H. Lowe-McConnell.The Hague: Dr. W. Junk B. V. Publishers.



Figure 6. Fishing off the coast of Peru regularly provides a plentiful catch of anchovies (photo-graph) because nutrient-rich water rises from depth in this region. This coastal upwelling is sopronounced that satellites carrying the proper sensors can chart this comparatively cold waterreaching the surface to the west of Peru and northern Chile (green on map). (Photograph courtesyof the Instituto del Mar del Peru, Proyecto Bitacoras de Pesca. Image of sea-surface temperaturefor May 1999 courtesy of the National Oceanic and Atmospheric Administration.)

Africa

SouthAmerica

a reprint from american scientist...daniel pauly, villy christensen, rainer froese and maria lourdes...

Documents