Chapter XIV: Part V: , Birds. U.S. Army, Signal Corps, Washington, D.C. Pages (8)

McKnight, D. E., and B. L. Hiliker. 1970. The impact of oil
development on waterfowl populations in Alaska. Proc. Western
Assoc. State Game Fish Comm. 50:286-297.

Nelson-Smith, A. 1972. Oil pollution and marine ecology. Paul Elek,
Ltd., London. 260 pp.

Sanger, G. A. 1972. Preliminary standing stock and biomass
estimates of seabirds on the subarctic Pacific region. Pages
589-611 _in_ A. Y. Takenouti et al., eds. Biological oceanography
of the North Pacific. Idemitsu Shoten, Tokyo.

Savile, D. B. O. 1972. Evidence of tree nesting by the marbled
murrelet in the Queen Charlotte Islands. Can. Field-Nat.
86(4):389-390.

Sladen, W. J. L., and R. E. LeResche. 1970. New and developing
techniques in antarctic ornithology. Pages 585-616 _in_ M. W.
Holdgate, ed. Antarctic Ecology. Vol. I. Academic Press.

Sowl, L. W., and J. C. Bartonek. 1974. Seabirds--Alaska's most
neglected resource. Trans. N. Am. Wildl. Nat. Resour. Conf.
39:117-126.

Stander, G. H., and J. A. V. Venter. 1968. Oil pollution in South
Africa. Pages 251-259 _in_ Proceedings of the International
Conference on Oil Pollution of the Sea.

Swartz, L. G. 1966. Sea-cliff birds. Pages 611-678 _in_ N. J.
Wilimovsky and J. N. Wolfe, eds. Environment of the Cape Thompson
region, Alaska. U.S.A.E.C., Div. Tech. Inf., Oak Ridge, Tennessee.

Thompson, D. Q., and R. A. Person. 1963. The eider pass at Point
Barrow, Alaska. J. Wildl. Manage. 27(3):348-356.

Uspenskii, S. M. 1964. Present day problems of nature conservation
in the Arctic. Problems of the North 7:171-178.

Watson, G. E., and G. J. Divoky. 1972. Pelagic bird and mammal
observations in the eastern Chukchi Sea, early fall 1970. Pages
111-172 _in_ C. I. Merton et al., eds. An ecological survey in
the eastern Chukchi Sea. U.S. Coast Guard Oceanogr. Rep. 50.

Williamson, F. S. L., M. C. Thompson, and J. Q. Hines. 1966. Pages
437-480 _in_ N. J. Wilimovsky and J. N. Wolfe, eds. Environment
of the Cape Thompson region, Alaska. U.S.A.E.C., Div. Tech. Inf.,
Oak Ridge, Tennessee.

Mortality to Marine Birds Through Commercial Fishing

by

Warren B. King

International Council for Bird Preservation
Smithsonian Institution, Washington, D.C.

R. G. B. Brown

Canadian Wildlife Service
Dartmouth, Nova Scotia, Canada

and

Gerald A. Sanger[52]

U.S. National Marine Fisheries Service
Seattle, Washington

Abstract

Commercial fishing has been responsible for
incidental mortality of seabirds for centuries,
but with the advent of offshore salmon gill-net
fishing in the North Pacific in 1952 and in
the North Atlantic in 1965, the magnitude of
this kill has increased, and there is strong
indication that populations of some seabirds
are being adversely affected. Murres (_Uria_
spp.) are most frequently killed, although
several other species are caught in lesser
numbers. The seabird resources of several
nations are involved in this mortality.
Longline fishing and inshore gill-net fishing
for salmon and cod also are responsible for
mortality of seabirds, although usually not in
significant numbers.

That the activities of commercial fishermen have caused mortality of marine birds surprises no one nowadays. Traditions of exploitation of marine birds by fishermen date from previous centuries, and fishing has contributed to the extinction of some species. For example, great auks _(Pinguinus impennis)_ and other birds were used as food by fishermen fishing for Atlantic cod _(Gadus morhua)_ on the Grand Banks of Newfoundland since the beginning of that fishery in the early 16th century (Collins 1884; Lucas 1890). The last great auk died in 1844, but smaller species, such as storm-petrels (Hydrobatidae), greater shearwaters _(Puffinus gravis)_, and black-legged kittiwakes _(Rissa tridactyla)_, were used for food until rather recently (Templeman 1945). This practice has now lapsed, however.

Inshore Fisheries

Until the advent of the offshore salmon gill-net fisheries in the North Pacific in 1952 and the North Atlantic in 1965, most seabird mortality in these areas was the result of local fishing close to shore. Several records of such bird mortality have been published. For example, 8,000-10,000 seabirds--presumably mostly alcids--were reported caught annually off Hammerfest in northern Norway (Holgersen 1961). E. Brun (personal communication) reported that the longline fishery off the coast of Norway is having serious consequences on Norwegian populations of murres.

Numbers of alcids are caught in nets set for Atlantic salmon _(Salmo salar)_ around the coasts of Ireland and Scotland (Biddy 1971). A similar situation exists along the west Greenland coast, although it is overshadowed there by the direct exploitation of huge numbers of alcids by hunting. Nonetheless, in 1967 for example, 15,000 alcids were recovered from fish nets in southwestern Greenland, where they were sold as food (Evans and Waterston 1976). The annual salmon catch of the west Greenland inshore fishery has fluctuated between 60 and 1,500 metric tons and has averaged about 1,000 tons. There are no data comparing the relative catch of birds and fish in this fishery.

Atlantic cod follow the spawning capelin _(Mallotus villosus)_ inshore along the east coast of Newfoundland in late June and early July. They are traditionally fished with traps and handlines along this coast, but there has been a recent trend toward using drift nets set on the bottom. Since alcids feed extensively on capelin at this time, many are caught in the cod nets set in areas close to the large colonies off Witless Bay (D. N. Nettleship, personal communication). Additionally, gill nets are set at the surface for salmon in the same area. Common murres _(Uria aalge)_ are most affected, but Atlantic puffins _(Fratercula arctica)_ are also taken.

There are as yet no estimates of the total alcid mortality from this fishery, although the annual catch of birds is believed to be smaller during the present than during the last decade because the fishing effort is reduced, and fishermen in the area now avoid setting nets near alcid concentrations because of the annoyance of having to remove the birds from their nets. The Witless Bay colonies contain over 77,000 pairs of common murres, or 11% of the total eastern North American population, and over 235,000 pairs of Atlantic puffins, or 71% of the North American population outside of Greenland (Brown et al. 1975). The potential danger is obvious.

There are few data on mortality of seabirds from inshore commercial fisheries in the North Pacific. Some mortality of alcids has been shown to take place in Cook Inlet, Alaska, from beach-netting for Pacific salmon (_Oncorhynchus_ spp.) adjacent to seabird rookeries and from drift-netting in the inlet (D. A. Snarski, personal communication), but this mortality has not been quantified.

Bilateral agreements between the United States and Japan, the U.S.S.R. and the Republic of Korea, concerning the use of inshore waters adjacent to some of the Aleutian Islands, Kodiak, Nunivak, St. Matthew, St. George, Kayak, and Forrester Islands permit trawling, longlining, and loading fish and fuel in some of these areas and at certain periods. Although these activities may affect the seabirds of these areas, the extent of the effects are not known (U.S. Department of the Interior, Alaska Planning Group 1974). Murie (1959) indicated, however, that the disappearance of the ancient murrelet _(Synthliboramphus antiquus)_ from Sanak Island, Gulf of Alaska, was probably due as much to fisheries as to the blue fox industry. It has been suggested that the Japanese murrelet _(Synthliboramphus wumizusumi)_ may have declined as the result of fishing activities near breeding sites off the coast of Japan (Bourne 1971).

Atlantic Offshore Gill-net Fishery

In 1965, Denmark began an offshore gill-net fishery for Atlantic salmon in the Davis Strait off the coast of west Greenland. The offshore fishery catch increased from 36 metric tons in 1965 to more than 1,200 metric tons in 1969, and then gradually decreased.

The fact that large numbers of seabirds--almost entirely thick-billed murres _(Uria lomvia)_--were being drowned in the salmon gill nets was brought to the attention of the International Council for Bird Preservation at its 15th World Conference in 1970. The Council's recommendation was submitted to the Danish government and stated: "... having noted that during the 1969 fishing season about 250,000 individuals of Brunnich's guillemot or thick-billed murre _(Uria lomvia)_, a pelagic diving bird, were caught in these drift nets and drowned, which number represents no less than 25 percent of the Greenland population and exceeds its annual reproductive capacity; urges the Danish Government, and the national governments of all other countries involved in this fishing, to take all possible measures to eliminate this very serious problem."

The figures in the recommendation were not supported by research; they appeared instead to have been derived from the observed mortality on an offshore fishery vessel in 1965, which was then related to the salmon catch on that vessel and applied to the total catch of the inshore fishery in 1964 (Anonymous 1969). Studies in 1969 and 1970 by the Fisheries Research Board of Canada finally gave a firm basis for the earlier, though poorly substantiated concern. On the basis of the assumption that the ratio of salmon to murres caught in experimental fishing applied to the commercial fishery, an estimate of an annual mortality of 0.5 million murres (±50%) was made on the basis of a salmon catch of 1,200 metric tons (Tull et al. 1972).

The birds being killed were from colonies in west Greenland, the eastern Canadian Arctic, and possibly east Greenland and Spitzbergen. Coupled with other known causes of mortality (particularly hunting on the Greenland and Newfoundland coasts, an unknown but definitely substantial kill from oil pollution, a calculated mortality of pre-fledging young, and an unknown natural post-fledging mortality) there is no doubt that the estimated annual production of 1.5 million chicks from west Greenland and the Canadian Arctic was less than the estimated total annual mortality (Tull et al. 1972). Thus, it comes as no surprise that recent surveys of murre populations of west Greenland and the Canadian Arctic have revealed massive declines in numbers (Evans and Waterston 1976; D. N. Nettleship, personal communication). It is therefore encouraging news that, as a result of an agreement between the United States and Denmark, the offshore salmon gill-net fishery was terminated at the end of the 1975 season. The inshore fishery remained in operation, however, but was restricted to a total annual salmon catch of 1,100 metric tons.

Pacific Offshore Salmon Gill-net Fishery

In the north Pacific Ocean, the Japanese gill-net fisheries for salmon (_Oncorhynchus_ spp.), which have operated since 1952, might be expected to have an even more destructive effect on seabirds, since the annual salmon catch by the three Japanese salmon drift-net fisheries was about one hundred times that in west Greenland in recent years. The first, the mothership fishery, comprising about 369 catcher-boats[53] serviced by 11 mother-ships, operates west of 175°W and generally north of 46°N during the summer. The second, the land-based fishery of about 325 ocean-going vessels, operates west of 175°W and south of 46°N; and the third, the coastal fishery, made up of about 1,380 short-haul vessels, operates off Hokkaido. The relative salmon catches of these three fisheries is on the order of 1:1.34:0.65.

Data collected on U.S. National Marine Fisheries Service research vessels in 1974 (obtained through the cooperation of Francis M. Fukuhara and Richard Bakkala, Northwest Fisheries Center, Seattle, Washington) give, for the first time, an estimate of the magnitude of the incidental seabird kill of the Japanese salmon gill-net fishery. The kill data are available only from the mothership area and from an area east of it to 165°W. The Japanese salmon fishery is restricted to waters west of 175°W by agreement with the United States. Bird kills from the other two areas may be estimated by the relative salmon catch figures for the areas, assuming that seabird densities, species composition, and catch effort are similar.

An estimate of the total kill of seabirds in the mothership area may be made by calculating the bird mortality per length of gill-net set by research vessels, multiplied by the total length of gill nets set by the 369 catcher-boats of the Japanese mothership fishery. About 4,666 km of nets are set and retrieved daily during the approximately 65-day fishing season. The estimated annual mortality in the mothership area is about 75,000 to 250,000 birds. The lower number is based on data from 10 cruises (450 km of nets set) west of 175°W, within the area of the mothership fishery. The higher number is based on data from 20 cruises, including those in the first figure, west of 165°W, and covering the period 18 April to 3 September 1974 (956 km of nets set), whereas the mothership fishery usually operates between mid-May and late July. Assuming similar seabird densities and catch per unit of effort in the areas of the land-based and coastal fisheries, the estimated annual mortality is between 214,500 and 715,000 birds. Since 1952, as many as 4.7 million birds may have been killed by the Japanese salmon gill-net fishery. It must be stressed that seabird densities and catch per unit of effort are not known to be similar for the areas in question; consequently the projection of bird kill figures from one area to all three is speculative.

In the mothership area and adjacent seas to the east, in addition to murres (48% of birds killed), significant numbers of shearwaters, _Puffinus_ spp. (27%); puffins (9%); and fulmars, _Fulmarus glacialis_ (5%) are killed, as are lesser numbers of small alcids, albatrosses (_Diomedea_ spp.), and storm-petrels. The murres and puffins taken in the mothership area are of U.S. and U.S.S.R. origin, and the shearwaters come from New Zealand, Australia, and Chile. In the coastal fishery area, Japanese and U.S.S.R. alcids are taken. Available knowledge of the populations of the species making up the bulk of the kill, which has been taking place for 20 years, is insufficient to suggest whether their annual reproduction can tolerate such losses. Prohibition of fishing within 160 km of North Pacific seabird breeding islands would help to decrease losses of alcids of U.S. origin, but would not help the shearwaters from the southern hemisphere.

Comparison of statistics of the salmon fisheries and associated bird kills from the North Atlantic and the North Pacific shows that the North Atlantic salmon fishery is concentrated in a relatively small area which is also along a major migration pathway of murres. Virtually all seabird mortality is confined to one species. Enough information is at hand to indicate that this cause of mortality, in conjunction with others known to be significant, is causing a drastic decline in the thick-billed murre population.

In the North Pacific, on the other hand, the fishery is more widely dispersed and the ratio of seabirds to salmon caught is much lower. Furthermore, several species are subject to mortality. No information is available to indicate whether alcid populations (which make up two-thirds of the kill) are stable or decreasing. The shearwaters, primarily sooty _(Puffinus griseus)_ and slender-billed _(P. tenuirostris)_, appear to be able to sustain not only these losses but also a sizable harvest of birds of the year (the so-called muttonbirds) on their New Zealand and Australian breeding grounds. Thus, although the latest estimates of the total standing stock of seabirds in the North Pacific in summer may be as high as 100 million (Sanger and King, this volume), and thus only about 1 of every 200 birds in the North Pacific region may be caught, the fact that a few species, particularly murres, are selectively caught raises questions about the impact of this fishery on populations of these species.

The U.S.-Japan Migratory Bird Convention of 1973 specifically protects all of the species thought to be subject to gill-net mortality in the Pacific. Thus, the Japanese salmon fleet apparently operates in constant violation of this convention.

Mortality of Albatrosses

A recent analysis of recoveries of Laysan albatrosses _(Diomedea immutabilis)_ and black-footed albatrosses _(D. nigripes)_ banded on the northwest Hawaiian chain from 1937 to 1969 showed that of a sample of 532 recovered birds, 57.4% of the Laysan species and 49.5% of the black-footed species were caught on fishhooks or in nets, and the means of recovery of many additional birds was thought to have been the same (Robbins and Rice 1974). It is likely that the large majority are taken on Japanese and U.S.S.R. longline tuna fishing gear. Although this cause of mortality is insignificant in terms of the total population of either species (only 0.2% of banded Laysan and 0.8% of banded black-footed albatrosses have been recovered by any means away from their breeding grounds), these species are protected by the U.S.-Japan Migratory Bird Convention. Furthermore, the possibility exists that individuals of the endangered short-tailed albatross _(Diomedea albatrus)_ might be killed in this manner.

Long-term Effects of Developing Capelin Fishery in Northwest Atlantic

Capelin are important food fish for many seabirds in the northwest Atlantic, and the development and expansion of this fishery off eastern Canada must be carefully monitored. In theory, the capelin fishery ought not to seriously affect the birds because it is designed to exploit a surplus of capelin artificially created by the overfishing of Atlantic cod, the capelin's most important predator. It is hoped that there is no prospect of the overfishing that may have contributed to the recent drastic decline of the Peruvian anchovy _(Engraulis ringens)_ and the seabird species dependent on it (Paulik 1971). However, the relative influence of overfishing and "El Niño" oceanographic conditions on the decline remains unclear. North Atlantic seabirds are, in any case, more versatile in their feeding habits (Belopol'skii 1961). But, the threat may be a subtle one. The important point to the seabirds may well be not merely the survival of a reasonably large capelin stock, but the presence of capelin schools in high densities in certain areas or at certain seasons. Lower densities might, for example, reduce the foraging efficiency of breeding birds, and hence their nesting success. The very large common murre colony on Funk Island, Newfoundland (500,000 pairs: Tuck 1960), might be particularly vulnerable. It lies close to an area where capelin are especially abundant and one which is already being exploited by the developing fishery.

References

Anonymous. 1969. Seabird slaughter. Sports Fish. Inst. Bull. 203:5.

Belopol'skii, L. O. 1961. Ecology of sea colony birds of the
Barents Sea. (Transl. from Russian.) Israel Program for
Scientific Translations, Jerusalem. 346 pp.

Biddy, C. J. 1971. Auks drowned by fishnets. Seabird Rep. No. 2.

Bourne, W. R. P. 1971. General threats to seabirds. ICBP [Int.
Counc. Bird Preservation] Bull. 11:200-219.

Brown, R. G. B., D. N. Nettleship, P. Germain, C. E. Tull, and
T. Davis. 1975. Atlas of eastern Canadian seabirds. Canadian
Wildlife Service, Ottawa. 220 pp.

Collins, J. W. 1884. Notes on the habits and methods of capture of
various species of seabirds that occur on the fishing banks off
the east coast of North America and which are used as bait for
catching codfish by New England fishermen. U.S. Comm. Fish Fish.
Rep. 1882:311-335.

Evans, P., and G. Waterston. 1976. The decline of the thick-billed
murre in Greenland. Polar Rec. 18:283-286.

Holgersen, H. 1961. On the movements of Norwegian _Uria aalge_. (In
Norwegian, English summary.) Sterna 4:229-240.

Lucas, F. A. 1890. Expedition to the Funk Island, with observations
on the history and anatomy of the Great Auk. Rep. U.S. Natl.
Mus., 1887-1888:493-529.

Murie, O. J. 1959. Fauna of the Aleutian Islands and Alaska
Peninsula. U.S. Fish Wildl. Serv., N. Am. Fauna 61. 406 pp.

Paulik, A. J. 1971. Anchovies, birds, and fishermen in the Peru
Current. Pages 156-185 _in_ W. W. Murdoch, ed. Environmental
Resources and Society. Sinauer Associates, Inc., Stamford, Conn.

Robbins, C. S., and D. W. Rice. 1974. Recoveries of banded Laysan
albatrosses _(Diomedea immutabilis)_ and black-footed albatrosses
_(D. nigripes)_. Pages 232-271 _in_ W. B. King, ed. Pelagic
studies of seabirds in the Central and Eastern Pacific Ocean.
Smithson. Contrib. Zool. 158.

Templeman, W. 1945. Observations on some Newfoundland seabirds.
Can. Field-Nat. 59:136-147.

Tuck, L. M. 1960. The murres. Canadian Wildlife Service, Ottawa.
260 pp.

Tull, C. E., P. Germain, and A. W. May. 1972. Mortality of
thick-billed murres in the west Greenland salmon fishery. Nature
(Lond.) 237 (5349):42-44.

U.S. Department of the Interior, Alaska Planning Group. 1974. Final
environmental impact statement, proposed Alaska Coastal National
Wildlife Refuges. 678 pp.

FOOTNOTES:

[52] Present address: U.S. Fish and Wildlife Service, Office of Biological Services--Coastal Ecosystems. 1011 E. Tudor Road, Anchorage, Alaska 99503.

[53] This figure is based on data through 1971. Since then, the number of catcher-boats has decreased to 332 in 1974 (F. M. Fukuhara, personal communication).

Interactions Among Marine Birds and Commercial Fish in the Eastern Bering Sea

by

Richard R. Straty and Richard E. Haight

National Marine Fisheries Service
Auke Bay Fisheries Laboratory
Auke Bay, Alaska 99821

Abstract

The high primary and secondary productivity
of the eastern Bering Sea makes it one of
the greatest producers of commercial fish
and largest congregating areas of marine
birds in the world. The fish and birds are so
interrelated that fluctuations in the abundance
of one may well be responsible for changes in
the abundance of the other. The seasonal and
annual variation in the impact of birds on fish
is a function of the life history, food habits,
growth rate, and final size of the fish species
of concern and of the distribution, abundance,
and feeding habits of bird populations--plus
the effects of the environment on these
factors. Stages in the life history of some of
the important commercial fish and shellfish
of the Bering Sea directly or indirectly
influenced by marine birds are identified.

The eastern Bering Sea is one of the world's richest fish-producing areas and is also one of the world's major congregating areas for marine birds. The large extent of the continental shelf and the climatic and oceanographic characteristics of the eastern Bering Sea combine to make this region extremely productive biologically. The distribution and abundance of plankton, benthos, and fish determine the distribution, time, and character of the migration of marine birds in the eastern Bering Sea (Shuntov 1961). Several studies have illustrated the close relation between marine birds and the biological properties of surface waters (Tuck 1960; Bourne 1963; Solomensen 1965). Spatial and temporal variations in the abundance of the fish families Clupeidae (herring), Gadidae (codfish), Osmeridae (capelin), and Ammodytidae (sand lance) are thought to be major determinants of the breeding seasons, breeding places, and movements of boreal seabirds (Ashmole 1971). The timing of breeding among larids and alcids is related to the seasonal changes in the surface waters inhabited by Ammodytidae and Clupeidae in the North Sea (Pearson 1968).

The eastern Bering Sea contains members of these and other fish families that are extensively exploited by man; the fish are also important as forage for other species of commercial fish, marine mammals, and marine birds. During some part of their life cycles, all fish species feed on plankton, nekton, benthos, or other fishes.

The incidental use or dependence of marine birds on commercial fish and the items on which the fish feed account for the major interaction between man and these two groups of animals.

In this paper, we consider how marine birds and fish interact. Although some of what we present is only speculative, we identify certain areas that have received little or no scientific study, areas in which further research is needed for a better understanding of the role of commercial fish in the ecology and dynamics of marine birds in the eastern Bering Sea.

Commercial Fish Resources of the Eastern Bering Sea

Most of the fishing in the eastern Bering Sea is done by Japan and the Soviet Union. Japan resumed fishing in the Bering Sea in 1953 (7 years after World War II), the Soviet Union started fishing in the region in 1959, and since the early 1960's both nations have accelerated their exploitation of Bering Sea fish stocks (Chitwood 1969).

Species of major concern to Japan and the Soviet Union include fish--walleye pollock _(Theragra chalcogramma)_, yellowfin sole _(Limanda aspera)_, Pacific cod _(Gadus macrocephalus)_, Pacific ocean perch _(Sebastes alutus)_, Pacific herring _(Clupea harengus pallasi)_, and sablefish _(Anoplopoma fimbria)_--and snow crabs (_Chionoecetes_ spp.). The distribution of the principal species being harvested in Bristol Bay and the eastern Bering Sea are shown in Figs. 1, 2, and 3. The weight of each of the major species in the total catches made by foreign and domestic fishermen in 1973 is shown in Table 1. In 1972, the catch of commercial finfish in the eastern Bering Sea alone amounted to 5% of the total world catch of marine fishes (H. Larkins, personal communication).

Most species of commercial fish in the Bering Sea are in a state of decline or in a depressed condition from overexploitation (Table 1). This is indicated by a reduction in the catch per unit of effort and in the mean size of fish in the commercial catch (H. Larkins, personal communication). The notable exception is the king crab (_Paralithodes_ sp.), which has increased in abundance in recent years as a result of reduced foreign fishing.

Table 1. _Foreign and domestic catch of fish and shellfish in the
eastern Bering Sea, including Bristol Bay, 1973._

----------------------------------------------
Catch
Species (metric tons)
----------------------------------------------
Fish
Pollock 1,500,000
Flatfish 125,000
Pacific cod 45,000
Herring 35,033
Salmon 11,785
Sablefish 7,000
Pacific halibut 222
Other 40,000
Shellfish
King crabs 26,798
Snow crabs 17,694
Shrimp Minor
----------------------------------------------

Routes of Interaction Between Marine Birds and Commercial Fish

The obvious ways in which marine birds and fish of commercial importance interact in the eastern Bering Sea are illustrated by the simplified food web diagram in Fig. 4. The major animal groups and species included in two of the categories in this figure--secondary producers (invertebrate forage) and intermediate carnivores (commercial and forage marine fish and shellfish)--are as follows:

_Secondary producers_

Zooplankton and micronekton
Copepods
_Calanus_ spp.
_Eucalanus_ spp.

Euphausiids
_Thysanoessa_ spp.

Amphipods
_Parathemisto_ spp.
_Gammarus_ spp.

Pteropods
_Spiratella_ spp.
_Clione_ spp.

Chaetognaths
_Sagitta_ spp.

Benthos
Polychaetes
_Nereis_ spp.
_Euroe_ spp.

Molluscs
_Mytilus edulis_
_Tonicella_ spp.
_Fusitriton oregonensis_

Echinodermata
_Strongylocentrotus_ spp.

Crustacea
Gammaridae
Mysidae
_Idothea_ spp.
_Pagurus_ spp.
_Hapalogaster_ spp.
_Sclerocrangon_ spp.

_Intermediate carnivores_

Eggs (littoral, adhesive)
Clupeidae

Pelagic larvae
Gadidae
Pleuronectidae
Osmeridae
Ammodytidae
Salmonidae
Gadidae
Pandalidae

Juvenile and small adults
Clupeidae
Osmeridae
Ammodytidae
Salmonidae
Gadidae
Pandalidae

Large adults
Clupeidae
Gadidae
Pleuronectidae
Salmonidae
Scorpaenidae
Lithodidae
Majidae
Pandalidae

Marine birds
Alcidae
Procellariidae
Laridae
Phalacrocoracidae

In our discussion, we mainly consider predation by birds on commercial fish and competition between birds and commercial fish for food. The extent of these interactions determines the potential for birds and fish to influence each other's abundance. The extent of the interactions also determines the impact of man's commercial harvest of fish on the abundance of birds or of the bird's harvest on the abundance of fish.

The extent of the interaction between marine birds and commercial fish depends on the abundance, distribution, feeding habits, and life history of the fish species of concern. We have limited our discussion to examples of the major commercial pelagic and demersal fish and shellfish of the eastern Bering Sea. We also use as examples those species of marine birds whose abundance in the eastern Bering Sea and feeding habits give them the greatest potential for influence on, or being influenced by, fish abundance.

Abundance and Feeding Habits of Marine Birds in the Eastern Bering Sea

Information on the general abundance and distribution of the most important marine birds in the eastern Bering Sea in the summer and winter is scattered among many published and unpublished reports: Shuntov (1961, 1966), Sanger (1972), Bartonek and Gibson (1972), and Ogi and Tsujita (1973); and surveys by D. T. Montgomery and W. E. Oien ("Bristol Bay waterbird survey, 1972," unpublished report of the U.S. Bureau of Sport Fisheries and Wildlife, Alaska area) and by J. G. King and D. E. McKnight (1969, "A waterbird survey in Bristol Bay and proposals for future studies," unpublished report of the U.S. Bureau of Sport Fisheries and Wildlife and the Alaska Department of Fish and Game, Juneau, Alaska).

In summer, the most abundant birds appear to be the procellariids, mainly the slender-billed shearwater _(Puffinus tenuirostris)_ and Pacific fulmar _(Fulmarus glacialis)_; the alcids, mainly the common murre _(Uria aalge)_, thick-billed murre _(U. lomvia)_, tufted puffin _(Lunda cirrhata)_, horned puffin _(Fratercula corniculata)_, and the ancient murrelet _(Synthliboramphus antiquus)_; and the larids, mainly the glaucous-winged gull _(Larus glaucescens)_ and the black-legged kittiwake _(Rissa tridactyla)_.

In winter, the alcids and larids appear to be the most abundant groups, the procellariids having been reduced by the departure of the slender-billed shearwaters for breeding grounds in the southern hemisphere. The selection of the types of food to be consumed by these marine birds is a function of their morphological and physiological adaptations and of the resultant feeding behavior. Ashmole (1971) classified the feeding behavior of various genera of marine birds and the relative importance of the kinds of food eaten by each group; this information for some of the Bering Sea bird species occurring in the genera listed by Ashmole (1971) is summarized in Fig. 5.

Fish and invertebrates are evidently of moderate to major importance in the diet of these marine birds (Fig. 5). The extent to which a given fish species is fed upon by or is in competition with marine birds for food is determined by the life history of the fish. Most pelagic and some demersal fish and shellfish are more subject to predation by pursuit diving birds than by birds restricted to the near-surface waters. Invertebrates appear to be equal to or more important than fish in the diets of birds feeding in near-surface waters (Fig. 5).

Predation by Marine Birds

The literature contains numerous accounts of marine birds feeding on marine fish and shellfish of commercial importance. Some studies quantify the impact of some bird species on certain species of commercial fish (Outram 1958; Shaefer 1970; Wiens and Scott 1976) and shellfish (Glude 1967). Other studies have shown that in some regions the value of guano produced by birds may exceed the value of the commercial fish they consume (Jarvis 1970). Some fish of worldwide commercial importance that are important in the diets of marine birds are listed in Table 2.

Table 2. _Fish of worldwide commercial importance in the diets of
some marine birds._

--------------------------------------------------------
Fish Shearwaters Murres Puffins Fulmars Gulls

Anchovy X -- -- -- --
Sardines X -- -- -- --
Herring X X X X X
Sprat X -- -- -- --
Pilchard X -- -- -- --
Capelin -- X X -- X
Salmon -- X -- -- --
Mackerel -- X -- -- --
Pollock -- X -- X --
Haddock -- X -- -- --
Cod -- X -- -- --
--------------------------------------------------------

The significance of bird predation on pelagic or demersal fish and shellfish (Fig. 5) depends on the feeding behavior of the birds and on the life history of the fish (e.g., distribution, abundance, growth, and adult size). Pursuit diving birds, such as murres and puffins, can consume fish at greater depths than can birds that feed near the surface, such as shearwaters, kittiwakes, fulmars, and gulls.

Aspects of the Life Histories of Fish Related to Predation by Marine Birds

Fish that are pelagic during part of their lives, such as salmon and herring, and forage fish like smelt, capelin, and sand lance, are vulnerable to greater predation by a wider variety of marine birds than are bottom-dwelling demersal fish, such as pollock, cod, sole, ocean perch, and halibut, as well as king and snow crabs. Some species that live on the bottom as adults have pelagic stages during which they are vulnerable to predation by marine birds. Juveniles of some demersal species (pollock, cod, halibut, some species of sole, and king crabs) are sometimes found in shallow water where they might be subject to predation by birds.

_Demersal Fish and Shellfish_

The early life histories of the commercially important demersal fish of the eastern Bering Sea are quite different (Table 3). For example, the eggs and larvae of Pacific halibut _(Hippoglossus stenolepis)_ generally occur at depths greater than 100 m (Hart 1973), whereas those of pollock and yellowfin sole are found at or near the surface (Musienko 1963, 1970). The eggs of Pacific cod are demersal, but the larvae are oceanic (pelagic) and occur from 25-150 m (Mukhacheva and Zviagina 1960).

In their juvenile stages, many demersal fish frequent the near-surface waters (Table 3), where they become vulnerable to predation by piscivorous marine birds. Juvenile pollock, for example, form into small schools that usually move about close to the bottom but sometimes move into areas as shallow as 3 m. Juvenile Pacific cod prefer the warmer water close to shore and may be found within 10 m of the surface (Moiseev 1953). The young of many species of flatfish, such as yellowfin sole, rock sole _(Lepidopsetta bilineata)_, and flathead sole _(Hippoglosoides elassodon)_, remain for a time in shallow warm water after assuming a demersal existence. Yellowfin sole 2-2.5 cm in total length may be found in abundance in areas as shallow as 5 m (Fadeev 1965; Moiseev 1953).

Table 3. _Informal listing of life history information on
selected species of commercial and forage fish and shellfish to
show vulnerability to predation by marine birds._ (? indicates no
information available.)

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Yusa 1954
(cm)[54] Tanino et al. 1959
=Walleye pollock= 31-35 95,700 Feb.- April Kobayashi 1963
(_Theragra_ 46-50 324,400 June -May Musienko 1963, 1970
_chalcogramma_ Serobaba 1968
Pallas) Hart 1973[55]

Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 0.1-0.2 0-10 Feb.-June { 12 at 6-7°C
{ 20.5 at 3.4°C[57]
Larval 0.4-0.9 10-25 March-? > 25 at 6-7°C
Larval 0.9-? 25-? ?-Sept. ?
Juvenile 2.2-4.1 0-?[58] Summer --
Juvenile 6.0-30.0 4-37 Summer --
Adult 30.0-70.0 0-386 -- --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Moiseev 1953
(cm)[54] Mukhacheva and
=Pacific cod= 60 1,200,000 Jan.- ? Zviagina 1960
(_Gadus_ 78 3,300,000 March Musienko 1970;
_macrocephalus_ Hart 1973[55]
Tilesius)

Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 0.1-0.11 100-250 Demersal { 8-9 at 11°C
{ 17 at 5°C
{ 28 at 2°C
Larval 0.5-3.2 25-150 Feb.-Aug. ?
Juvenile ? 10-? Summer --
Adult 40.0-99.0 0-900 -- --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Stevenson 1962
(cm)[54] Musienko 1970
=Pacific herring= 20.5-22.0 26,600 May- Varies Rumyantsev and
(_Clupea_ 28.0-31.0 77,800 June Darda 1970
_harengus pallasi_ Reid 1972
Valenciennes) Hart 1973[55]

Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 0.1-0.2 0-12 Demersal 10-20[57]
Larval 0.9 0.5-8 May-June }
Larval 1.3 0.5-8 June-July } 42-56
Larval 2.5 1-6 July-Aug. }
Juvenile 2.5-20.5 0-? March-Nov. --
Adult 20.5-31.0 0-140 March-Nov. --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period
(cm)[54] Clemens and
=Capelin= ? 3,000 June- ? Wilby 1961
(_Mallotus_ ? 6,000 July Musienko 1970
_villosus_ 10.3 6,670 Hart 1973
(Muller)) ? 60,000
Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 0.1 <20 Demersal 14-?
Larval 0.5-? ? June-? ?
Juvenile ? ? March-Nov.(est.) --
Adult ? 0-? March-Nov. --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Musienko 1963, 1970
(cm)[54] Kashkina 1970
=Pacific sand lance= -- ? June- [59] Hart 1973
(_Ammodytes_ Aug.
_hexapterus_ Pallas)
Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg ? ? Demersal ?
Larval 0.7-3.4 0-? June-Sept. ?
Juvenile 3.6-9.6 0-? ? --
Adult 26 0-? ? --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Paraketsov 1963
(cm)[54] Lisovenko 1965
=Pacific ocean perch= 26 10,000 March- ? Lyubimova 1965
(_Sebastes_ 44 180,000 May Kashkina 1970[60]
_alutus_ (Gilbert))

Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg[61] -- -- -- --
Larval[62] 0.6-? [62] March-Aug. ?
Juvenile 6.2 37-128 -- --
Juvenile 10.4 37-154 -- --
Juvenile 14.7-21.3 37-230 -- --
Adult 21.3-51.0 37-420 -- --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period
(cm)[54] Novikov 1964
=Pacific halibut= 75 101,723 Oct.- ? Hart 1973
(_Hippoglossus_ 135 2,800,837 March
_stenolepis_
Schmidt) Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 0.3-0.4 40-935 Oct.-March 48 at ?
Larval 0.8-1.5 >200 Nov.-May }
Larval 1.5-2.9 <100 May-Sept.} 70-98
Juvenile 3.4-4.2 7-43 -- --
Juvenile 19-25 7-45 -- --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Moiseev 1953
(cm)[54] Pertseva-Ostraumova
=Yellowfin sole= 1954; Musienko 1963;
(_Limanda_ 26.1-28.0 1,295,000 June- July Fadeev 1965;
_aspera_ 40.1-42.0 3,319,500 Aug. Kashkina 1965_a_,
(Pallas)) 1965_b_[55]
Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 0.07-0.09 >0 June-Aug. 9.4 at 13.1°C[57]
Larval 0.2-1.2 >0 July-Oct. ?
Juvenile 2.1-2.5 5-15 -- --

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period
(cm)[54] Kurata 1960, 1964
=King crabs= 9.4 55,408 April- ? Korolev 1964
(_Paralithodes_ 17.1 444,651 June Rodin 1970
_camtschatica_ (Tilesius))

Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)
Egg -- 100-200[63] -- ?
Zoeal} 0.55-0.65 ? April-July {33 at 7-10°C
Zoeal} {23 at 12.3-12.5°C
Glaucothoeal 0.38x0.18 ? May-? ?
Juvenile ? 1-? -- ?

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Haynes 1973[55]
(cm)[54] Jewett and
=Snow crabs= Haight[64]
(_Chionoecetes_ ? ? ?[65] ?
_bairdi_ Rathbun)
Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg 100[63] -- ?
Prezoeal 0.22-0.28 ? May-? 1-2 at 2.5°C
1st zoeal 0.50-0.56 ? Summer ?
2d zoeal ? 0-10 Summer ?
Megalopal 0.30-0.35x ? Summer --
0.18-0.21
Juvenile 0.44-0.48x ? -- --
0.32-0.35

Fecundity Spawning season Source of data
Length Mean no. Total Peak
of female of eggs period period Ito 1968; Kon 1970
(cm)[54] Haynes 1973;
=Snow crabs= Motoh 1973
(_Chionoecetes_ ? ? ?[65] ? Jewett and
_opilio_ Haight[64]
(O. Fabricius))
Total Depth from Seasonal Duration of
Life length surface period of life stages
stage (cm)[56] (m) pelagic life (days)

Egg ? 93[60] -- ?
Prezoeal -- ? May-? }
1st zoeal 0.48-0.54 ? Summer }
2d zoeal 0.62-0.71 ? Summer } 63-66 at 11-13°C
Megalopal 0.29-0.33 ? Summer }
0.19
Juvenile 4.4-4.8x ? -- --
3.2-3.5

The commercially important king and snow crabs of the eastern Bering Sea also have larval stages that are pelagic (Table 3). Zoeae and megalopa of snow crabs are found near the surface where they are vulnerable to plankton-feeding marine birds. The eggs of king crabs are attached to the abdomen of the female, but after hatching, the larvae become pelagic and occur near the surface. They are planktonic through five larval stages before settling to the bottom to take up demersal residence (Kurata 1960, 1964). These larvae attain a length of 5.5-6.5 mm and spend 33 days or more in the plankton (Kurata 1960). Even after the young king crabs have settled to the bottom, they may still frequent water shallow enough to make them vulnerable to predation by some marine birds. Juvenile king crabs 1 and 2 years of age appear to prefer shallower water than do older crabs. In southeastern Alaska, during the spring, small juvenile crabs have been observed in pods at depths as little as 1 m below the low tide level.

The available life stages of king and snow crabs and commercially important demersal fish (Table 3) represent an enormous food supply for other fishes and marine birds. Predation by marine birds on pelagic eggs and on the larval and juvenile stages of demersal fish is not well documented, probably because the rapid digestion rate of birds makes species identification of these stages difficult. Investigators must often depend on the presence of the hard parts of fish (such as scales and otoliths) in the stomachs of birds to identify the species eaten. Because these hard parts have not yet formed in the larvae and most juveniles, predation by marine birds on older fish is more apparent on examination of stomach contents. Full understanding of predation by marine birds on demersal fish and shellfish requires additional data on when and where the egg, larval, and juvenile stages are present.

_Pelagic Fish_

Many fish, such as herring, capelin, smelt, and salmon, are pelagic for part of their lives, particularly during the spring and summer feeding periods. The extent of predation by marine birds on these species depends primarily on the location of their spawning grounds, their growth rates, and the size of the adults. The spawning location determines the extent of predation on eggs, whereas growth rate and adult size determine during how much of its lifetime a given fish species is vulnerable to the wide variety of marine birds.

Herring spawn in intertidal and subtidal zones and spend most of their post-larval lives in bays or estuaries near the coast. They deposit their adhesive eggs primarily on vegetation, and the eggs are particularly vulnerable to predation by a wide variety of marine and terrestrial birds. Outram (1958) estimated that gulls alone accounted for 39% of the egg loss on the spawning grounds at Vancouver Island, British Columbia. When herring larvae hatch, they are between 0.7 and 0.8 cm long; when they metamorphose about 6-8 weeks later, they are between 2.6 and 3.5 cm long. Thereafter, juvenile herring grow rapidly and reach a length of about 7-10 cm before winter. Although herring as old as 13 years and up to 38 cm long have been reported in Alaska, they seldom exceed 30 cm and 11 years of age (Rounsefell 1929). During spring and summer, herring are commonly within 10 m of the surface, but in winter, they are in water 100-140 m deep. Although herring are particularly vulnerable to predation in spring and summer, they are available to marine birds during most of their life.

The life history of capelin is somewhat different than that of herring--they live in the open sea near the surface and throughout the water column most of their lives. Sometime in June or early July, they migrate in large schools toward shore to spawn (Musienko 1970). In British Columbia, capelin bury their eggs in coarse sand and gravel in the intertidal and subtidal zones. The larvae are 0.5-0.7 cm long at hatching and are carried by currents to the open sea where they develop in the plankton. Capelin attain an age of 5 years and a maximum length of about 22 cm; their small size makes them vulnerable to predation by marine birds most of their lives, and they are an important pelagic food fish for other commercial fish in the Bering Sea.

The sand lance reaches a maximum size of 20-26 cm and is vulnerable to bird predation during most of its life. Little information is available on the maximum age attained by this species in the Bering Sea, but because of its size, it is an important forage fish for many commercial fish species.

The five species of Pacific salmon of the eastern Bering Sea spawn in fresh water, unlike herring, capelin, and sand lance. Their eggs are not vulnerable to extensive predation by marine birds; gulls take mainly salmon eggs which have been dislodged from the gravel and are drifting or being rolled along the stream bottom by the current (Moyle 1966). After a few months to several years in fresh water, the juvenile salmon (5-14 cm long) enter the Bering Sea during late spring or early summer and migrate through these waters to feeding grounds, primarily in the north Pacific Ocean. At maturity, the survivors return to their home streams and rivers to spawn. It is during the seaward migratory phase of their life cycle that salmon are most vulnerable to predation by marine birds.

The sockeye salmon _(Oncorhynchus nerka)_ is the most abundant and valuable species harvested by American fishermen in the waters adjacent to the Bering Sea and, as a result, the one that has been most extensively studied during early marine life. Juvenile sockeye salmon are between 8 and 14 cm long when they enter the Bering Sea between late May and early July. They are most abundant in the upper 1 m of water at night and the upper 2 m during the day (Straty 1974)--well within the regime that can be exploited by many species of marine birds.

The numbers of juvenile sockeye salmon migrating seaward from the Bristol Bay region of the Bering Sea in a single year has ranged between 46.3 and 370.4 million (H. Jaenicke, personal communication). This is equivalent to between 409 and 3,267 metric tons (on the basis of the mean weight of the juveniles when they enter the Bering Sea). These large numbers of juvenile sockeye salmon, plus juvenile chinook salmon _(O. tshawytscha)_, coho salmon _(O. kisutch)_, chum salmon _(O. keta)_, and pink salmon _(O. gorbuscha)_ from all other rivers entering the Bering Sea, represent a considerable input of energy from fresh water in the form of prime forage fish for other fishes, marine birds, and mammals. Young salmon enter the Bering Sea each year over a period of only 6 to 8 weeks and may follow rather discrete coastal migration routes through the Bering Sea (Fig. 6), with the result that predators have access to an abundant but transient food supply.

The only published account of predation by marine birds on juvenile salmon in the Bering Sea is that of Ogi and Tsujita (1973). They found juvenile sockeye salmon in the stomachs of murres captured in gill nets in the eastern Bering Sea. The predation did not appear extensive, but most of the birds were captured outside or on the fringes of the main seaward migration route of the salmon. The foods of marine birds should be studied in conjunction with studies of the migrations of juvenile salmon.

Influence of Growth Rate and Adult Size of Fish on the Extent of Predation

Incubation time for fish eggs, the length of the pelagic larval period (Table 3), and the growth rate of juvenile fish are species-specific and temperature-dependent. The extent to which a fish species is subjected to predation by marine birds is directly related to the rate at which development and growth occur. For example, the less time it takes the pelagic eggs of demersal fish and shellfish to hatch and complete pelagic larval life, the less is the time they will be preyed on by marine birds. For fish species that are pelagic during their entire life, the rate of growth will determine how long they remain small enough for birds to eat. Some of the smaller pelagic fish, such as herring, capelin, and smelt, are vulnerable to bird predation most of their lives; larger pelagic species like salmon may be preyed on for only a very short time. The maximum size fish that can be eaten by marine birds is, therefore, important in evaluating predation on a given species of fish.

The literature on the food habits of marine birds contains little on the sizes of fish consumed. Tuck (1960) stated that murres probably will take fish up to 18 cm long. Ogi and Tsujita (1973) estimated the lengths of Pacific pollock in the stomachs of murres taken in the eastern Bering Sea at 24 cm.

Herring in the eastern Bering Sea reach an age of 11 years and grow to about 33 cm. Herring could, therefore, be taken during most of their lives by murres but during only the first few years by smaller birds such as fulmars and shearwaters. Capelin and some species of smelt would be vulnerable to birds during all their lives. Although the size of adult Pacific salmon varies with the species, they are all so large that they are not preyed upon by marine birds. Once in the ocean, juvenile salmon grow at such a rapid rate that they are probably not very vulnerable to marine birds after their first 4 to 6 months at sea. Limited studies on the growth of juvenile sockeye salmon in the eastern Bering Sea (Straty 1974) indicate they may double their size in their first 8 weeks at sea. A sockeye salmon that entered the Bering Sea at 12 cm in mid-June would be 24 cm long in August--the maximum size that a murre could eat; the fish could be eaten by smaller marine birds for much less time. Pink and chum salmon enter the sea at a smaller size than sockeye salmon and would be vulnerable to predation both by a greater variety of marine birds and for a longer period of time.

Competition Between Commercial Fish and Marine Birds

We do not know the importance of competition between marine birds and commercial fish in the eastern Bering Sea. Only a few investigators have even alluded to competition between marine birds and fish for food. Ogi and Tsujita (1973) mentioned that competition seemed to exist between murres and juvenile sockeye salmon for euphausiids in the eastern Bering Sea. We have listed some of the types of forage fish and invertebrates eaten by commercial fish (Table 4) and marine birds (Table 5) in the eastern Bering Sea; comparison of these two tables clearly indicates that competition could occur.

The principal factors determining the extent of competition between marine birds and fish are the numbers of birds and fish, the length of time that various life history stages of the fish are in association with the birds, and the abundance of the preferred foods at these times. The impact of competition depends on the adaptability of the birds and fish to alternative types of food.

The types and sizes of food eaten by fish vary with the life history stage--especially with size at each stage. For instance, very young herring eat the eggs and nauplii of copepods or small copepodite stages and barnacles. As herring grow, their diet includes small fish and larger zooplankton, such as mature copepods, amphipods, euphausiids, and pteropods. Pacific cod shorter than 9 cm feed on small crustaceans (Moiseev 1953), whereas larger cod eat young crabs, shrimp, and fish. Small juvenile sockeye salmon feed mainly on larval stages of euphausiids (Straty 1974), but larger juveniles also eat the more adult forms, which eventually make up a significant part of their diet (Nishiyama 1974).

The change in the diet of fishes with growth results in competition with a changing variety of marine birds. For example, deep-diving birds may replace surface feeders as the major bird competitors of the Pacific cod and pollock as these fish increase in size and seek deeper waters. The diet of cod changes from small crustaceans in shallow water to progressively larger food that eventually includes herring, sand lance, shrimp, and crabs. The change to herring and sand lance, and quite possibly small crabs, places the adult cod in competition with both the surface feeders and pursuit diving birds, but adult cod do not compete with birds for zooplankton.

Table 4. _Food items eaten by the adult stage of seven
commercially important species of fish in the eastern Bering Sea._

Pacific
Food Walleye Pacific ocean Yellowfin Pacific
item Herring Salmon pollock cod perch sole halibut

Invertebrates
Pteropods X X -- -- X -- --
Squid -- X -- X X -- X
Polychaetes X X X X -- X X
Copepods X X X -- -- -- --
Amphipods X X X X X X --
Euphausiids X X X -- X X --
Decapods X X X X X X X

Fish
Capelin X X X X -- X --
Sand lance -- X X X -- -- X

Table 5. _Forage fish and invertebrate foods eaten by seven
species of marine birds in the eastern Bering Sea._

Shear-
Food item waters Murres Puffins Murrelets Fulmars Kittiwakes Gulls

Forage fish
Sand lance X X X -- -- X X
Capelin -- -- X -- -- -- --

Invertebrates
Copepods -- -- -- -- -- X --
Euphausiids X X -- -- -- X --
Amphipods X X -- -- -- X --
Decapods X X -- -- -- X --
Pteropods -- X -- -- -- -- --
Chaetognaths -- -- -- -- -- -- --
Polychaetes -- X X -- -- X --
Squid X X -- -- X -- --

As pollock increase in size, they continue to feed mainly on zooplankton, but they change from copepods near the surface to euphausiids at mid-depths and near the bottom. Euphausiids are large and abundant zooplankters which, for the most part, are available only to deep-diving birds. Adult pollock also consume herring, sand lance, capelin, and other small fish.