Oceanan Abridged History

The Ocean Ecology of Salmon in the Northeast Pacific OceanAn Abridged History

WILLIAM G. PEARCY* CollegeofOceanic and Atmospheric Sciences, Oregon State University Corvallis, Oregon 97331, USA

STEWART M. MCKINNELL North Pacific Marine Science Organization Post Office Box 6000, Sidney, British Columbia V8L 4B2, Canada

Abstract.Research on the ecology of salmon in the northeast Pacific Ocean began in the early 20th century. Charles Gilbert and Willis Rich demonstrated the basis for the stock concept and were instrumental in changing common misconceptions of the times. Later in the 1900s, research endeavors, primarily under the auspices of the International North Pacific Fisheries Commission, led to important studies on the distribution and migration of maturing salmon on the high seas. Research on the early juvenile period was initiated later, especially after the 1982-1983 El Niño clearly showed the limits of the ocean's carrying capacity along the west coast of the United States. There is now good evidence for both intra- and interspecific competition among salmon in the open ocean and for correlations between variable physical environments, such as El Niflos and regime shifts, and survival of salmon during early ocean life. How mortality rates are affected by physical forcing, food availability, predation, and food web structure and how these effects will be modified by climate change and global warming are all major challenges for the future.

Introduction increased exponentially over the years (e.g., Figure 1), we sought to examine more closely Scientific research has led to major advances how the early pioneering research formed the in understanding the biology of Pacific salm- basis of our current knowledge. on (Groot and Margolis 1991; Quinn 2005). Although our focus is the history of re- But if we stop to consider how this progress search on the ocean ecology of salmon, we was achieved, it becomes apparent that the begin with the story of salmon in freshwater steps to enlightenment have been far from where they were first observed and harvested linear. In this chapter, we explore some of the after leaving the ocean. Thereafter, the bulk paradigms, the myths, and the science that of our discussion is organized around five have enhanced and sometimes misled our un- paradigms that focused our thoughts: derstanding of the biology of Pacific salmon Oncorhynchus spp. Although the quantity of Homogeneous salmonversus home- research on the life of salmon in the sea has stream theories, Migrationslocal or distant, * Corresponding author: [email protected] Critical periods,

7 8 PEARCY AND MCKINNELL

200

150

100

0 c c?\ \cø2\ Decade

Figure 1. Summary of the research on juvenile salmon in the ocean, cited by decade, from Canadian and U.S. publications (Beamish et al. 2003; Brodeur et al. 2003).

Ocean carrying capacity and density de-densities bring about interactions with vari- pendence, and ous terrestrial predators, including humans Ocean variability. (Homo sapiens). Because the salmon were such an important food source for the indig- We conclude with comments on future challenges to a better understanding of theenous peoples of western North America, the characteristics of their upstream migration importance of freshwater and ocean environ- were of great interest. The lack of a written ments for the survival of Pacific salmon. In history precludes an in-depth examination of the interests of brevity, we are selective rather the first -.8,000 years of the study of salmon than inclusive in our choice of references, es- biology, but some clues to the status of knowl- pecially for the period of copious research in edge were captured by the first European-ori- more recent times. We do not always follow gin fur trappers and explorers. With the onset a linear chronology of history and often hop- and growth of global commercial interests in scotch through time. salmon, the cannerymen added to our knowledge, and finally (after the stocks declined) Homogenous Salmon the scientists began their study. During these The anadromous life history of Pacific salm-early years, the ocean life of the salmon was on, combined with their size and high localunseen, unstudied, and unknown. THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 9 During the 19th and early 20th centuries,young coho salmon 0. kisutch returned 3 there were many misconceptions about theyears old, thus corroborating his age estimates freshwater phase of the life history of thesefrom scales. anadromous fishes. It was believed by many, Ultimately, Charles Henry Gilbert (Fig- including the renowned biologist from Indi-ure 2) found that his observations were not ana University, David Starr Jordan, that Pacific consistent with Jordan's beliefs, and Gilbert salmon did not have any special ability to find aevolved as a proponent of the stock concept home stream. Together with his former studentin Pacific salmon. Gilbert's comparative re- at Indiana, Charles Henry Gilbert, they wrote,search on Pacific salmon along the North "It is the prevailing impression that the salmonAmerican coast established his reputation have some special instinct which leads them to(Dunn 1996). After 5 years of informal study return to spawn in the same spawning-groundsof the salmon of British Columbia, Gilbert where they were originally hatched. We fail towas retained in 1914 by John Pease Babcock find any evidence of this in the case of the Pa- to establish a program of ongoing scientific cific coast salmon, and we do not believe it toinvestigation of the biology of sockeye salm- be true. It seems more probably that the youngon 0. nerka in British Columbia. Vestiges of salmon, hatched in any river, mostly remain inthat program continue today. the ocean within a radius of 20, 30, or 50 mi of Gilbert's research on sockeye salmon in its mouth" (Jordan and Gilbert 1887). Likewise,British Columbia was instrumental in the es- Hugh Smith, then director of the U.S. Bureau oftablishment of a full-time director and staff of Fisheries, believed that there were no significantbiologists at the Pacific Biological Station at differences between the races of salmon. TheNanaimo, British Columbia in 1924 (Foerster returning fish selected their spawning streams 1955). The first director of that station, Dr. W. randomly and stocks of a particular speciesA. Clemens, and his wife, Dr. Lucy S. Cle- were genetically homogeneous, with any dif-mens, continued the work of measuring and ferences due to the effects of different environ-aging sockeye salmon in the major fisheries ments (Jordan 1904; Ricker 1972). into the late 1940s. While Jordan was the first Pacific salm- Gilbert studied the patterns of growth, on biologist, Gilbert's eventual contributionrecorded like tree rings, on salmon scales to to salmon science was considerably greater.learn about the basic life history of salmon in Jordan came to the West Coast as presidentfreshwater and the ocean. In his seminal 1914 of the new Stanford University and shortlypaper on sockeye salmon from British Co- thereafter appointed Gilbert as its first profes-lumbia, he showed that salmon from different sor of zoology. Both were involved in somerivers had different numbers of circuli in the of the historic cruises of the research steamerfreshwater phase of growth. He concluded that Albatross. This ship made many cruises intofish from different rivers had different life his- Alaskan waters to study salmon in the latetories, returned to their natal streams as differ- 1800 s and early 1900s (Roppel 2004). ent age groups, and were essentially isolated Some of Gilbert's earliest successes con-from other populations (Gilbert 1914-1925). cerned the interpretation of the age of a salmonHe describes the significance of this discov- from its scales. In a letter to John Pease Bab-ery on the occasion of his departure from the cock, deputy commissioner of fisheries, Prov-employ of the Province of British Columbia ince of British Columbia, on January 20, 1913, in 1924: "The practical significance of this de- Gilbert discussed how he collaborated withtermination is of great economic importance, Willis Rich to demonstrate that fin-clipped since it demonstrates that, in order to maintain 10 PEARCY AND MCKINNELL

Figure 2. Photograph of Charles Henry Gilbert, obtained with the assistance of the Gilbert lchthyological Society and was provided for our use courtesy of Dorothy T. Gilbert. the runs to a given district, it will not sufficeRiver and concluded that Chinook salmon to install a hatchery on any convenient stream0. tshawytscha also consisted of many dis- into which the entire hatchery output may betinct populations that migrated to sea at dif- liberated, or to make joint and uniform regula-ferent sizes and at different times. Based on tions apply to all streams alike. Each streamscale characteristics and distinct cycles of must be given separate consideration in orderabundance, Rich (1939) reasoned that Pa- that each may receive its own quota of fry, ascific salmon returned to their natal streams to the run to each stream is self-dependent." spawn and that each species was composed Willis Rich, one of Gilbert's first stu-of many local, independent, and self-perpetu- dents, extended Gilbert's pioneering researchating populations. He wrote that "conserva- on life history diversity to the Columbiation of a species as a whole resolves into the THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 11 conservation of every one of the componenttion in oceanic waters, switching to local cues groups." Much later, in a small coastal river, such as olfaction in coastal waters. Reimers (1973) found that Chinook salmon The work of Gilbert and Rich dispelled had different life histories, residing in thethe myth that salmon of the same species were river and estuary for different periods and ex-one homogeneous population. This had enor- periencing different rates of return. mous implications for the direction of salmon Homing to natal streams, fundamentalresearch, now based upon the stock concept, to Gilbert and Rich's theories, was suspectedand the management of stocks as discrete units. many years before being clearly demonstrat-The evolution of understanding that maturing ed (Chamberlain 1907). Once it was clearsalmon home to their natal streams is now key that salmon did home to a natal stream, theto their population biology and management. mechanism by which this was accomplishedThis is the basis for the stock concept, the became the major scientific question. Whileprinciple that stocks are unique and should be tagging sockeye salmon off the west coast ofmanaged as discrete units (Ricker 1972). British Columbia in 1924, Clemens (BCLA Later researchers, such as Russell Foerster, 1926) severed the olfactory nerves of 259 ofFerris Neave, Andrew Pritchard, and William 515 tagged individuals. Of the 256 that were Ricker, made great contributions to our knowl- not treated, 59 tags (23%) were recovered inedge about the freshwater phases of salmon. or off the Fraser River while only 24 of 259Initially, they believed that factors in freshwa- (9.3%) with severed olfactory nerves wereter and fishing limited the abundance of return- taken in or off the Fraser River. No tags wereing spawners (see Foerster 1955; Beamish et recovered elsewhere, indicating the impor-al. 2003). These beliefs focused management tance of olfaction in homing. Later, Hasleron hatchery production, harvest, escapement and Wishy (1951) performed several experi- targets, and production in freshwater. ments that demonstrated the role of olfaction Lichatowich (1999) pointed out that the for homing of salmon in freshwater. belief in the homogeneous salmon often had Less is known about homing and naviga- self-serving implications for the commercial tion of salmon from the open ocean to theirfisheries, for fish hatchery operators and their spawning grounds. Larkin (1975) stated, "Cer-desire to transfer eggs among watersheds, and tainly none of the present theories of migrationfor a general lack of management of the salm- is adequate to account for the apparent phe- on fisheries of the late 1800s and early 1900s. nomenal ability of salmon to navigate whereIndividual stocks were simply not important. there are few apparent navigational cues" asAn example of the attitudes of the day was ex- each individual performs the migration oncepressed by representatives of the commercial with no possibility of learning from a parent.salmon industry: "The State of Washington Salmon often undertake remarkable long dis-Fish Commissioners are fully alive to the fact tance migrations at sea. Today, several hypoth-that only by a complete system of increasing eses are postulated to explain the phenomenalthe amount of fish by artificial propagation can orientation and navigation in the open oceanit be hoped to save the industry from the expe- (see Healey 2000 and Quinn 2005 for re-rience of the Eastern States." views). On the basis of apparently directed and well-timed migrations, Healey and Groot Letter to Premier and Council of the Province of Brit- (1987), Healey (2000), and others concludedish Columbia by the Salmon Canners' Association. Pages 617-619 in Sessional Papers, Second Session, that species use a combination of mechanisms, Ninth Parliament of the Province of British Columbia, including compass and bicoordinate naviga- Session 1901. Victoria, Canada. 12 PEARCY AND MCKINNELL The inclination to dabble in salmon biol-some scientists thought that salmon entered ogy has a long history. The belief that stocksthe ocean and stayed relatively close to shore could be transplanted among different riverand, when mature, swam up the nearest avail- systems, an underlying assumption of theable stream. By the early 1900s, there was a homogeneous concept, was advocated by thegrowing impression that sockeye salmon re- first explorer to cross North America to theturning to the Fraser River arrived from the Pacific Ocean by land. After leaving what isopen sea to the northwest of the Strait of Juan now Bella Coola, British Columbia on Augustde Fuca (BCLA 1903). The lack of food in 16, 1793, Alexander McKenzie recorded intheir stomachs upon arrival led to specula- his journal, "If I could have spared the time,tion that these sockeye salmon had traveled and had been able to exert myself, II...] ita considerable distance from their feeding was my intention to have taken some salmongrounds. The indigenous peoples of British alive, and colonized them in the Peace River;Columbia reported that the start of the run though it is very doubtful whether that fishoccurs in the outer islands a month or more would live in waters that have not a commu-before they make their appearance near the nication with the sea." (Sheppe 1962). mouths of streams (Gilbert 1913). Although science made significant con- Gilbert (1913) made some of the first ob- tributions to the early knowledge of Pacificservations on coastal migrations of juvenile salmon, it is important to remember that itsalmon. He ascertained certain differences was incremental. New science and ultimatelyamong the species, but the questions posed better ideas were often ignored because theywere rather simple: "After leaving the rivers, were inconvenient or contrary to local knowl-no young sockeyes are on record from salt edge and prevailing paradigms of the timeswater along the BC coast. The young of all viz, technology can improve nature. Exploi-other salmon species can be caught in traps in tation or development of salmon resourcesJuan de Fuca strait; the sockeye must pursue usually preceded a good understanding ofa different course. It is not improbable that science. Accepted science arrived late. Thethey strike directly for the outer coasts, pass- first canneries were established in the 1 860sing through the deep-water channels; but of so the industry had 40 years of experiencethis we have no direct evidence. During the with salmon, and the indigenous peoples hadyears of their sojourn in the sea their habits more than 8,000 years of experience, beforeare wholly a matter of inference." Gilbert and any formally trained scientists appeared toRich (1927) published their investigations assert their dominance in the knowledge ofof sockeye salmon runs to the Karluk River, salmon biology. Some topics have not pro- Alaska. gressed far in more than 100 years. The rush With the advent of motor-powered fish to develop a salmon farming industry beforeboats at the beginning of the 20th century, their impacts could be evaluated, includingsalmon fisheries expanded their range sea- those of escaped nonnative fish on native Pa-ward of the estuaries, and interest grew in the cific species, is a case in point. ocean distribution and interception of Pacif- ic salmon. Some of the first research on the Ocean Migrations: Local or Distant? ocean distribution of salmon was conducted The first reports of the ocean life of Pacificby O'Malley and Rich (1919) who studied salmon came from fishermen as they ex- migrations of sockeye salmon approaching panded their pursuits seaward from estuar-the Fraser River. In 1925, fisheries admin- ies to the coastal regions. As we have noted,istrators and scientists from Alaska, British THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 13

Columbia, Washington, Oregon, and Cali- The International North Pacific Fisheries fornia met and established the International Commission Pacific Salmon Investigating Federation toThe most significant period of ocean research develop a cooperative plan for studying theon the distribution and migrations of salmon problems associated with migrating Pacificfollowed formation of the International North salmon (BCLA 1926). One of the researchPacific Fisheries Commission (INPFC). By priorities was to identify the origins andthe 1930s, the Japanese salmon fleet was fish- migrations of stocks under commercial ex- ing throughout the northwest Pacific and into ploitation by tagging and releasing salmonthe Bering Sea to gain access to an abundance at sea. Tagging experiments in 1927 andof salmon and other species that are relatively 1928 revealed that Chinook salmon fromrare along the coast of Japan. Following the California were feeding off the west coast ofwar in the Pacific in the early 1 940s, the victors Vancouver Island and that Chinook salmoninsisted on a clause in the 1951 San Francisco from the Columbia River, Skeena River, andPeace Treaty that obligated Japan to be part Fraser River intermingled off the west coastof an international fisheries regulatory body of the Queen Charlotte Islands (Clemensin the North Pacific. This led to the Interna- 1932). Sockeye salmon from the Nass andtional Convention for the High Seas Fisheries Skeena rivers were captured in fish traps asof the North Pacific Ocean with three signa- far north as Sumner Strait, Alaska (BCLAtories: Canada, Japan, and the United States. 1926). The terms of the convention established the Collectively(reviewedbyFoersterInternational North Pacific Fisheries Com- 1955), these studies showed limited migra-mission in 1953. As little was known of the tion of pink salmon 0. gorbuscha and chumbiology of Pacific salmon or their prey and salmon 0. keta, short travel distances forpredators on the high seas (Hartt 1962), a key coho salmon, with all tagged fish recov-role of the new commission was to conduct ered in the year they were tagged. Chinookresearch on the distribution and biology of salmon, on the other hand, traveled longsalmon and other marine fishes that swam in distances along the coast. Those taggedthe open North Pacific Ocean (Foerster 1955). west of Vancouver Island or in the QueenAfter 1992, the INPFC metamorphosed into Charlotte area were recovered to the souththe North Pacific Anadromous Fish Commis- in Puget Sound and off the Columbia River,sion, which continues research on salmonids supporting the theory of a northwest feedingin the open ocean today. migration of young fish and a southeast mi- Within a year of ratification of the con- gration of maturing adults (Foerster 1955).vention by the governments, scientific studies Later, Davidson (1937) tagged fish off theof the biology of Pacific salmon on the high west coast of British Columbia and record- seas were underway and tagging salmon on ed their recovery in rivers throughout thethe high seas was an integral part of these in- Pacific Northwest. He found that Chinookvestigations. Bulletins of the INPFC provid- salmon made extensive migrations from theed information on the distribution patterns of Columbia River to Vancouver Island. Sucheach species of Pacific salmon in the ocean- studies were a major impetus to treaties andcoho (Godfrey et al. 1975), sockeye (French agreements for the management of migrato-et al. 1976), chum (Neave et al. 1976), Chi- ry stocks after it was recognized that oceannook (Major et al. 1978 and Takagi et al. fisheries could affect the abundance of these1981), and steelhead 0. mykiss (Burgner et migratory populations. al. 1992). These studies revealed the broad 14 PEARCY AND MCKINNELL ranges and overlapping oceanic distribu- Vancouver Island into the Bering Sea. From tions of many species. Individuals were oftencatches, they concluded that most juvenile tagged thousands of kilometers from their na-salmon (but not juvenile steelhead) migrated tal streams. Because Canada and the Unitednorth and west along a narrow coastal belt or States were primarily interested in potentialcorridor. Later studies confirmed that juve- interceptions of various stocks, these studiesnile sockeye, pink, chum, coho, and Chinook tended to focus on oceanic migrations in rela-salmon were rarely found beyond the conti- tion to continent or country of origin. nental shelf (Jaenicke and Celewycz 1994; Few studies considered distributions ofWelch et al. 2003). Pacific salmon on the high seas at the stock! Research on juvenile salmonids in the population level because suitable techniquesoceans off California, Oregon, Washington, for identifying discrete salmon populationsCanada, and Alaska are reviewed by Brodeur had not yet been developed. Gilbert (1914)et al. (2003), Beamish et al. (2000, 2003), and was aware of the possibility of stock-specificHeardetal. (2001), respectively. Straty (1974) patterns of freshwater scale growth from hiswas one of the first to sample juvenile salmon studies of scales of maturing sockeye salmonwith purse seines to show the distribution and as they returned to the Fraser River. The pat-migratory routes of separate stocks of sock- terns of freshwater growth on the scales ofeye salmon from Bristol Bay. Healey (1980) returning adults varied by date and fishingrelated differences in estuarine residence location. Gilbert surmised that these differ-times and life histories of juvenile salmon ences reflected the different origins of theseto their feeding habits, offshore movements, sockeye salmon from among the many sock-and foraging success in the Strait of Georgia. eye nursery lakes within the Fraser basin. Ex- Groot et al. (1985) followed juvenile sockeye ploiting the knowledge of regional variationsfrom the mouth of the Fraser River to Queen in scale growth patterns was one of the tech- Charlotte Strait in a study that spanned four niques used to estimate the continent of originseasons. of salmon caught on the high seas, but it was With the advent of tiny coded-wire tags many years before studies of the distributionsin the 1970s that could be injected into large of individual populations could be consid- numbers of juvenile salmonids, more detailed ered. Most early studies of migrations wereinformation became available on migrations based on tagging (Foerster 1955). A notableduring and after the first year in the ocean, es- study by Straty (1975) based on exploratorypecially for coho and Chinook salmon (e.g., fishing and marking fish described the adultsee Myers et al. 1996 based on catches in the migration of Bristol Bay sockeye salmon andopen ocean and Weitkamp and Neely 2002 how stocks segregated by river of origin (asfor correlations between hatchery latitudes much as 200 km from the mouths of homeand the location of ocean catches). Walker rivers). and Myers (1992) reviewed stock identifica- The first extensive research on juveniletion techniques for salmon on the high seas. salmonids in the sea was the pioneering re- Genetic markers are now used to distin- search by Hartt (1980) and Hartt and Dellguish different stocks at sea (Seeb et al. 2004; (1986). They provided information on di-Beacham et al. 2005). Recently, it has been ets, growth, travel directions and rates, andpossible to test hypotheses about stock-spe- migration patterns based on purse seiningcific aggregations of salmon using coded- and tagging of fish during the summers ofwire tags recovered from salmon caught on 1956-1970 in the eastern North Pacific fromthe high seas (McKinnell et al. 1997). How- THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 15 ever, even with these new genetic and adultRiver discharge and sea surface temperatures tagging techniques, there is still much to learnthat may affect the distribution of sockeye in about ocean migrations and where stocks arethe ocean prior to migration (Wickett 1977; at specific times of the year when criticalGroot and Quinn 1987). Blackbourn (1987) events influence their survival. proposed a temperature model to explain how Besides the prolific research on oceanrun timing was related to ocean temperatures migrations, the INPFC fostered importantin the Gulf of Alaska, and Thomson et al. research by Canadian, Japanese, and U.S.(1992) modeled how ocean currents affect scientists on the physical oceanography ofthe latitude of landfall and migration speeds the environment occupied by salmon in theof Fraser River sockeye. North Pacific (e.g., Fleming 1955; Dodimead et al. 1963; Favorite et al. 1976), includingThe Critical Period Concept ocean circulation, temperature and salinityA critical period is an interval of time of high structure, and fronts and domains that relatedand variable mortality that is believed to de- to the distribution of salmon. Relationshipstermine the survival of a year-class. The hy- among salmon distributions and ocean con-pothesis for critical periods in the early ocean ditions revealed that various species and agelife of Pacific salmon is based on the mor- groups had affinities for certain environmen-tality schedules observed in the life tables tal conditions and water masses in the Northof fishes and other animals where mortality Pacific, such as the subarctic boundary, therates usually decrease rapidly and abruptly Alaskan Stream and subarctic gyre. Favoritewith the age of a cohort. Godfrey (1958) and Hanavan (1960) reported that the salinityconcluded that it was the early marine life front at about 42°N coincided approximate-of salmon that determined subsequent adult ly with the southern distribution of salmonabundances. Parker's (1962) postulate of ad- in the spring. Temperatures also influenceditional mortality during the first year associ- distributions and migrations. Manzer et al.ated with estuarine and early ocean life was (1965) listed preferred temperature rangeslater verified by his research (Parker 1968) for each species, how they changed season-where he showed the highest mortality rates ally, and the importance of upward domingfor pink salmon (2-4%/d) occurred during the of isotherms. first 40 dat sea. Later, Hartt (1980) called this Studies of the migration routes of re-period the first critical summer in the ocean. turning Fraser sockeye salmon have alsoBax (1983) also found highest mortality rates been related to ocean conditions. Babcockvery early in ocean life for chum salmon. was the first to report (BCLA 1903) that By the late 1900s, it was apparent that we maturing sockeye salmon appeared to mi-understood less about salmon during the first grate around Vancouver Island either fromfew months at sea than all the other phases the north through Johnstone Strait or fromof their life history (Healey 1980). This was the south through Juan de Fuca Strait (e.g.,due, in part, to the fact that the overwhelming Hamilton 1985; Groot and Quinn 1987). Themajority of salmon biologists were studying percentage of fish migrating to the north oreither freshwater or maturing ocean phases. south has varied greatly, with most arrivingDuring the late 1970s and early l980s, many via the southern route during 1953-1977,studies were initiated on the ecology of ju- and thereafter, more to the north (Groot andvenile salmonids at sea (Figure 1). Major Quinn 1987; McKinnell et al. 1999). Thesereasons for a spate of new ocean research variations have been correlated with Fraserwere the drastic declines of many stocks in 16 PEARCY AND MCKINNELL the U.S. Pacific Northwest following the bigbut this relationship broke down after 1981 1982-1983 El Niño, low-population abun-(see Pearcy 1992, 1996). Recent studies are dances in the Pacific Northwest and the en-consistent with a critical period in early ocean suing listings of many species/stocks underlife and indicate survival rates of several spe- the U.S. Endangered Species Act. The 1982-cies of salmon are related to marine condi- 1983 El Niflo was the nail in the coffin of thetions experienced by juvenile salmon just idea that the ocean had an unlimited capacityprior to or during out-migration (e.g., Mueter to support whatever salmon juveniles couldet al. 2005). A suite of environmental factors be produced in freshwater. have been correlated with the ocean survival Studies that found correlations betweenof Oregon coho salmon. These include con- environmental factors during early oceanditions during the winter before ocean entry, life and subsequent adult returns supportedthe period of ocean entry, and the winter af- the critical period hypothesis. Vernon (1958)ter ocean entry (e.g., Koslow et al. 2002; Lo- found that high AprilAugust temperaturesgerwell et al. 2003). Beamish and Mahnken and low salinities in Georgia Strait were in-(2001) posited that early marine growth and versely related to abundance of pink salmonattainment of a critical size by the end of the the following year (see also Holtby et al.first summer determines mortality during the 1990). Wickett (1958) found that low sea sur-first fall and winter in the ocean. face temperature in June was associated with We conclude that mortality of salmon low pink salmon survival along the centralin the ocean generally decreases with age coast of British Columbia, and Blackbournand that high rates of mortality often occur (1985) showed a significant relationship be-soon after ocean entrance but may also occur tween marine survival and salinity duringlater in life before, during, or after their first summer and fall of ocean entry. A positivewinter in the ocean. During unusually poor relationship between early ocean growth ofocean conditions, high rates of mortality may Fraser River sockeye and marine survivaloccur after the first summer in the ocean life suggested to Henry (1961) that conditions fa-as evidenced by several salmon populations voring good growth also favor good survival.during the 1982-1983 El Niflo (Wooster and Van Hyning (1973) found an inverse trendFluharty 1985). Although mortality rates are between the fall Chinook abundances in theknown to be correlated with ocean condi- Columbia River and sea surface temperaturestions, the exact mechanisms are still basically during the first few months of life at sea. unknown. Research on coho salmon from Oregon also provided evidence for the importanceCarrying Capacity of the Ocean of ocean conditions to survival. The number"Carrying capacity is a measure of the bio- of early-maturing male coho salmon (jacks)mass of a given population that can be sup- was usually a good predictor of the numberported by the ecosystem. It changes over time of adult coho returning the following year,with the abundance of predators and resourc- indicating that the adult run size was deter-es. Resources are a function of productivity mined largely during the first few months inof prey populations and competition" (U.S. the ocean. Moreover, a strong relationshipGLOBEC 1996). By this definition, pro- was found between the intensity of coastalcesses that affect abundances, either through upwelling during the first summer in thelower trophic levels and availability of food ocean and subsequent survival (Gunsolus(bottom-up) or predation (top-down), as well 1978; Scarnecchia 1981; Nickelson 1986),as density-dependent factors, and both intra- THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 17 and interspecific competition, all affect theproduction of fishes is "likely controlled by ocean carrying capacity for salmonids andbottom-up (food web) processes rather than their survival rates. self-regulated by effects of fish abundance Unlimited carrying capacity of the oceanon their zooplankton prey.. .at least when for salmon was an early paradigm and a stim-considered over the entire life history and ulus for increased hatchery production. Sev-entire North Pacific." This conclusion finds eral studies based on the standing stocks ofsome support in the large-scale and concur- zooplankton supported this theory. LeBras-rent increases in both zooplankton and fish seur (1972) estimated that salmon ate onlybiomass in the North Pacific (e.g., Beamish about 10% of the annual net production, di-and Bouillon 1993; Brodeur and Ware 1992) rectly or indirectly, of large copepods. S angerand recently developed estimates of the re- (1972) estimated that salmon consume onlylationship between primary production and about 3.4% of the zooplankton productionfish catches (Ware and Thomson 2005). Sugi- in the North Pacific. Similarly, Favorite andmoto and Tadokoro (1997) also thought that Laevastu (1979) postulated that salmon con-bottom-up processes were most important on sumed only a small fraction of the availabledecadal and longer time scales, but that top- zooplankton and that the ocean could sup-down control may affect zooplankton dynam- port more salmon at that time. Walters et al.ics on shorter time scales. (1978) also concluded that ocean food limita- Some studiesclearly document top- tion was unlikely for juvenile salmon unlessdown control of salmon survival in the ocean only a small fraction of the total zooplankton(Fresh 1996; Beamish et al. 2003; Brodeur is available. These estimates assumed thatet al. 2003). Predation by coho salmon on salmon were relatively unselective feederspink and chum fry and size-selective mortal- on zooplankton and did not consider sea-ity has been reported by Parker (1971) and sonality of the food supply. We now knowHargreaves and LeBrasseur (1985). Common that chum salmon often specialize on gelati- murres Uria aalge prey on coho juveniles nous zooplankton (Welch and Parsons 1993;(Bayer 1986) as do spiny dogfish Squalus Welch 1997), and maturing salmon in theacanthias (Beamish et al. 1992). Top-down open ocean often select micronekton as prey,effects have been documented in other eco- including gonatid squids (LeBrasseur 1966;systemsin Calilfornia kelp forests (Halp- Pearcy et al. 1988; Kaeriyama et al. 2004).em et al. 2006) and by the removal of a top In coastal waters off Oregon, juvenile salmo-predator, the Atlantic cod Gadus morhua that nids consume only a small portion of the totalproduced cascading effects through lower available prey and do not appear to be foodtrophic levels (Worm and Myers 2003; Frank limited during most years (Peterson et al.et al. 2005). 1982; Brodeur et al. 1992). Fisher and Pearcy Although coastal upwelling enhances (1988) also found little evidence that chronicprimary and secondary production, it also af- food shortages affected the growth of surviv-fects the distribution and abundance of other ing juvenile coho salmon, except during Elanimals and salmon predators. Fisher and Niño years of very low productivity. Pearcy (1988) and Pearcy (1992) hypoth- Perry et al. (1998) reviewed bottom-upesized that during periods of weak coastal and top-down processes regulating epipe-upwelling and low productivity, coho smolts lagic fish production in the subarctic Pacific. off Oregon and Washington are confined to They noted evidence for top-down control ofa narrow zone of upwelled waters close to zooplankton by salmon but concluded thatshore where both competitors and predators 18 PEARCY AND MCKINNELL are concentrated and where growth may bethe cause of the variable growth might be due slow and predation intense. Moreover, duringto "more rigorous competition among the warm years, there is an influx of predatorsschool" (Gilbert 1914). He also looked at the from the south that probably increases pre-number of cases of salmon packed per year dation mortality. On the other hand, duringversus the average weight of sockeye salmon years of strong upwelling, a broad region offrom the Fraser River but found no positive cool, turbid water exists where forage animalstrend, no doubt in part because of the 4-year such as northern anchovy Engraulis mordax,cycle of abundance in the Fraser River. smelts (Osmeridae), and herring Clupea pal- Larkin (1975) questioned whether there lasii are more common and provide a bufferwere density-dependent interactions among to predation on juvenile salmon, which aredifferent stocks of salmon at sea. He con- dispersed over a large area and less suscepti-cluded that good evidence was lacking since ble to predation. This interaction is supportedfluctuations in both American and Asian re- by Holtby et al.'s (1990) research that showed gions were usually in phase, indicating large- that marine survival and early ocean growthscale environmental influences rather than of juvenile coho salmon was positively re- out-ofphase competitive interactions. Later, lated to the intensity of coastal upwelling andBeamish and Bouillon (1993) summarized that the survival of juvenile coho salmon wasthe total landings of pink, chum, and sockeye positively related to the abundances of juve-salmon from America and Asia and found nile herring of about the same size, suggest- similar long-term trends that suggested that ing that herring provide a buffer to predation.climate and ocean conditions are important in The interaction between bottom-up and top-basin-wide salmon production, possibly over- down processes is again clearly illustratedriding any obvious density-related signal. for juvenile pink salmon in Prince William Contrary to the earlier paradigm of unlim- Sound where predation rates are linked toited ocean carrying capacity, many research- food availability and prey switching (Willetteers have found evidence for competition et al. 2001). In sum, both top-down and bot- among maturing fish in the high seas. Rogers tom-up processes are undoubtedly important.(1980) related the annual mean weight of re- They are not mutually exclusive and likelyturning sockeye salmon to run size for Bristol interact in complex ways in the ocean. HowBay sockeye salmon and found an inverse re- the ecosystems react to the changes in the rel- lationship (Figure 3). Peterman (1984) found ative strengths of these processes in responseinverse relationships between reconstructed to changes in external forcing and how thesetotal abundances of sockeye salmon in the processes affect salmon productivity are allGulf of Alaska and the mean length of Brit- relevant issues that will affect managementish Columbia sockeye salmon, also suggest- and conservation of salmon resources in theing density-dependence. McKinnell (1995) future. showed that the negative relationship between mean length of British Columbia sock- Density Dependence eye salmon and Bristol Bay sockeye salmon Gilbert appears to have been the first to con-abundance held only for age-classes that had sider the possibility of density effects. Asspent 3 years at sea, presumably sharing a early as 1914, he observed that the length andcommon feeding area. Peterman (1987) also weight of sockeye salmon returning in thenoted that the mean adult size of returning dominant cycle year of 1913 were smallerFrazer River pink salmon was inversely relat- than those returning in 1912 or 1914 and thated to the ratio of abundance of pink salmon to THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 19

Ages 1.2 and1.3 Ages 2.2 and 2.3 4.0

3.8

3.6

3.4

3.2

3.0

2.8

2.6 C

1.8 0

1.6 I I I I I I I 0 10 20 30 40 50 0 10 20 Run inMillions

Figure 3. Mean weights of sockeye salmon in the Bristol Bay runs as a function of the number of fish in the run (from Rogers 1980). Females are indicated by open circles, m ales by solid dots. zooplankton at Ocean Station "P," again sug-Peterman 1984, 1987), a likely result of intra- gesting competition. Pink salmon are usuallyspecific competition for food. smaller during years of high abundance than Others have reported evidence for inter- in years of low abundance (Davidson andspecific interactions among salmon in the Vaughan 1941; Blackboum and Tasaka 1984;ocean. Tadokoro et al. (1996) observed that 20 PEARCY AND MCKINNELL the diet of chum salmon depended on theneither Emlen et al. (1990) nor Nickelson abundance of pink salmon. When pink salm-(1986) found evidence for interactions be- on abundance was high, macrozooplanktontween wild and hatchery coho from Oregon biomass can be reduced by intense feedingthat affected survival. Cooney and Brodeur by pink salmon (Shiomoto et al. 1997). Both(1998) and Beamish et al. (1997) questioned sockeye salmon growth at sea and abundancewhether massive production of hatchery salm- declined significantly following years of highon could increase returns, especially during Asian pink salmon abundance (Ruggerone etperiods of low ocean productivity. Peterman al. 2003), with pink salmon influencing the(1991) felt that because of density-dependent abundance, species, and energetic content ofgrowth and survival, in addition to potential the prey of the other salmon species (Aydinresponses by predators, increased production et al. 2000; Ruggerone and Nielsen 2004).of more smolts by hatcheries may have lim- These studies clearly indicate how interspe-ited benefits. In the Columbia River, Levin cific competition among species of salmo-et al. (2001) showed that during years of nids for food can influence the populationpoor upwelling, greater releases of hatchery dynamics among species and suggest thatChinook corresponded with years of higher pink salmon can exert top-down control onmortality of wild Chinook salmon. The bil- their prey. lions of hatchery-reared salmon released Declines in the average size or size-at-into the North Pacific Ocean along with the age of maturing salmon at sea is more recentdepressed runs of endangered or threatened evidence for density-dependence (Ishida etstocks raise concerns about hatchery-wild al. 1993; Bigler et al. 1996; Helle and Hoff-stock interactions in the ocean (Beamish et man 1998; Pyper and Peterman 1999). Theseal. 1997; Heard 1998). Replacement of wild declines were associated with increased totalsalmon with hatchery-reared salmon has abundance of salmon during the 1980s andbeen observed in some regions of the Pacific 1990s in the subarctic Pacific, including largeOcean in recent years (e.g., Sweeting et al. increases in hatchery production on both2003). This practice is a major management sides of the Pacific (Beamish et al. 1997).issue that needs broad ecological assessment, This again suggests that density-dependenceespecially in view of climate shifts in ocean in the open ocean could be an important con-productivity. sideration in the enhancement and management of salmon stocks. Abundance during theOcean Variability 1990s may have approached the carrying ca-As in most other environments, ocean ecosys- pacity of the ocean to produce salmon. S ever-tems fluctuate on various temporal and spatial al alternative hypotheses have been proposedscales that ultimately lead to variations in the to explain these changes, however, includingproductivity of Pacific salmon. El Niños have changes in ocean productivity and selectionaffected coastal fisheries off South America of larger fish by fishermen resulting in genet-for centuries (Quinn et al. 1987), but only ic changes (Ricker 1981; Quinn 2005). recently have their full effects in the Pacific Issues concerning competition betweenNorthwest been described in detail (Sette wild and hatchery salmon arose during theand Isaacs 1960; Wooster and Fluharty 1985; late 20th century. Gunsolus (1978) noted thatChavez et al. 2002). The warming in the during periods of weak upwelling along theCalifornia Current system during the 1958 Oregon coast, releasing more hatchery fishEl Niño was associated with a significantly did not result in greater production. However,higher proportion of Fraser River sockeye THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 21 migrating north around Vancouver Island andal. 2000). Similar patterns of low-frequency later arrival at the Fraser River (Royal and Tul- variability are also apparent in the stand- ly 1961) and the record delay of the run of Co-ing stocks of zooplankton in the California lumbia River fall Chinook (Van Hyning 1973).Current (Roemmich and McGowan 1995) The 1982-1983 El Niflo and the climate re-and in the subarctic Pacific (Brodeur and gime shift of 1976-1977 were historic eventsWare 1992). They also appear in variations that woke up many salmon biologists about thein freshwater conditions that affect salmonid critical importance of the ocean conditions toproduction (Lawson et al. 2004) and in the survival and run size of Pacific salmon (Pearcydramatic changes in the species abundances et al. 1985; Wooster and Fluharty 1985; John-of pelagic nekton off Oregon and Washington son 1988). (Emmett and Brodeur 2000). Regime chang- One of the first scientists to recognizees in ocean and freshwater productivity are what we might now call production regimesimportant considerations in management of was Tanaka (1962). He fit a series of Rickersa!monid resources. These changes also need curves to different sequences of years to ii- to be considered when evaluating our efforts lustrate how the years of variable productiv- to improve habitat conditions for rearing and ity appeared to occur in stanzas. Canadianspawning fish. scientists argued against this interpretation (Anonymous 1962). Given the absence ofChallenges for the Future evidence to the contrary, Ricker (1958) as- Although we now know that the carrying ca- sumed that the variability of points aroundpacity of the ocean for salmon varies over a his stock-recruit curves was the result ofrange of time and space scales, from interan- stochastic interannual variability rather thannual variability in seasonal upwelling to El persistent production regimes, an idea thatNifloSouthern Oscillation frequencies to in- persisted at least into the l990s. terdecadal and millennial events, we are still Low-frequency fluctuations in averageuncertain about how these external forcings levels of salmon abundance are known fromaffect food web structure, community dy- long-term records (e.g., Chatters et al. 1995;namics, and salmon ecology. How do vary- Finney et al. 2000). Variations in abundanceing conditions affect the relative importance during the past century have been linked toof interactions among trophic levels, bottom- large-scale atmosphereocean interactionsup and top-down processes, and the rates of over the North Pacific. The Pacific Decadalpredation, growth, and survival of salmon? Oscillation is an index of the dominant pat-What stages in the life history of salmon have tern of sea surface temperature variability inrapid and variable changes in survival that af- the North Pacific Ocean (Mantua et al. 1997).fect year-class success? Where do these in- It shows that regime shifts (i.e., periods ofteractions occur in the ocean? Are all species abrupt change) occurred in 1925, 1947,1977,or stocks of salmon affected by similar pro- 1989, and 1998 (Hare and Mantua 2000).cesses? After more than 100 years of study, These climate regimes tend to be correlatedthere is no life table for any species of Pacific with periods of change in salmon catchessalmon and rarely are we able to distinguish in the North Pacific and often show inversefreshwater and marine mortality. Interactions production trends in Alaska and the Pacificbetween freshwater, estuarine, and marine Northwest (Pearcy1992,1996; Beamishphases need to be evaluated and measured. and Bouillon 1993; Francis and Hare 1994;Furthermore, what are the effects and roles of Mantua et al. 1997; Hare et al. 1999; He!!e ethatchery salmon in this variable ocean? 22 PEARCY AND MCKINNELL

Technology will be important in answer-Acknowledgments ing some of these questions. Tagging withWe are very grateful to M. C. Healey, R. D coded-wire tags, external macrotags, andBrodeur, A. Schoener, and anonymous re- archival tags has already provided importantviewers for their helpful comments on the information, but the information typically re-manuscript. lies on a fishery to provide recaptures. New types of tags, smart tags, are providing excit-References ing new information, and their contributions Anonymous. 1962. Further considerations on the to our knowledge will increase in the future. status of the sockeye stocks of the Fraser and Archival or data storage tags that record tem- Skeena river systems with respect to Articles perature, depth, swim speed, and geoloca- 111(1 )(a) and IV of the international convention are available (Boehlert 1997). Some of tion for the high seas fisheries of the North these have revealed surprising information Pacific Ocean. International North Pacific on diel changes in the thermal environment Fish Commission Bulletin 9. and vertical distribution of salmon (WalkerAydin, K. Y., K. W. Myers, and R. V. Walker. et al. 2000). Acoustic tagging studies hope to 2000. Variation in summer distribution of provide new data on early ocean survival and prey of Pacific salmon (Oncorhynchus spp.) in the offshore Gulf of Alaska in relation to migrations (Welch et al. 2004). And genetic oceanographic conditions, 1994-98. North data will be important to identify the stock Pacific Anadromous Fish Commission Bul- compositions, distributions, and migrations letin 2:43-54. (e.g., Seeb et al. 2004; Beacham et al. 2005). Bax, N.J. 1983. Early marine mortality of marked Satellites and remote sensors will continue juvenile chum salmon (Oncorhynchus keta) to provide valuable information on the struc- released in Hood Canal, Puget Sound, Wash- ture and dynamics of the ocean that can be ington, in 1980. Canadian Journal of Fisher- related via geographical information systems ies and Aquatic Sciences 40:426-435. to distributions and migrations of salmon. Bayer, R. D. 1986. Seabirds near an Oregon salm- New and improved modeling and statistical on hatchery in 1982 and during the 1983 El Niflo. Fisheries Bulletin 84:279-286. approaches will also become important with BCLA (British Columbia Legislative Assembly). our increasing ability to collect these new 1903. Sessional papers: annual report of the and more extensive data on salmonid behav- Commissioner of Fisheries of British Colum- ior and their dynamic environments. All of bia. Queen's Printer, Victoria. these will enhance our knowledge of the en-BCLA (British Columbia Legislative Assembly). vironments and ecology of salmon at sea and 1926. Sessional papers: annual report of the improve our ability to manage and conserve Commissioner of Fisheries of British Colum- stocks in the future. bia. Queen's Printer, Victoria. Climate change and its effects on PacificBeacham, T. D., J. R. Candy, B. McIntosh, C. salmon is a big challenge for the future (e.g., MacConnachie, A. Tabata, K. Kaukinen, Beaugrand and Reid 2003; Mote et al. 2003; L. Deng, K. M. Miller, R. E. Withler, and Payne et al. 2004). Understanding the im- N. Varnavskaya. 2005. Estimation of stock composition and individual identification of pacts of climate change on the physical and sockeye salmon on a Pacific Rim basis using biological environments, in both freshwater microsatellite and major histocompatability and the ocean, how they interact, and how we complex variation. Transactions of the Amer- can ameliorate these effects, will be vital to ican Fisheries Society 134:112-1146. the productivity of many populations in theBeamish, R. J., and D. R. Bouillon. 1993. Pacific future. salmon production trends in relation to cli- THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 23

mate. Canadian Journal Fisheries and Aquat-Blackbourn, D. J., and M. B. Tasaka. 1984. Adult ic Sciences 50:2270-2291. scale growth and population dynamics of Beamish, R. J., and C. Mahnken. 2001. A critical pink salmon from the Fraser River, Brit- size and period hypothesis to explain natural ish Columbia. Pages 97-105 in K. L. Fresh regulation of salmon abundance and the link- and S. L. Schroder, editors. Proceedings of age to climate and climate change. Progress the 1983 Northeast Pacific Pink and Chum in Oceanography 49:423-437. Salmon Workshop, Rosario Resort, Orcas Is- Beamish, R., Y. Ishida, V. Karpenko, P. Livings- land. January 1983. Washington Department ton, and K. Myers, editors. 2000. Workshop of Fisheries, Olympia. on factors affecting production of juvenile Boehlert, G. W. 1997. Application of acoustic and salmon. North Pacific Anadromous Fish archival tags to assess estuarine, nearshore Commission Technical Report 2. and offshore habitat utilization and move- Beamish, R. D., C. Mahnken, and C. M. Neville. ment of salmonids. NOAA Technical Memo 1997. Hatchery and wild production of Pa- NMFS. NOAA-TM-NMFS-SWFS-236. cific salmon in relation to large-scale, natu- Brodeur, R. D., and D. M. Ware. 1992. Long-term ral shifts in productivity of the marine envi- variability in zooplankton biomass in the sub- ronment. ICES Journal of Marine Science arctic Pacific Ocean. Fisheries Oceanography 54: 1200-1215. 1:32-38. Beamish, R. J., I. A. Pearsall, and M. C. Healey. Brodeur, R. D., R. C. Francis, and W. G. Pearcy. 2003. A history of the research on the early 1992. Food consumption ofjuvenile coho and marine life of Pacific salmon off Canada's Chinook salmon on the continental shelf off Pacific coast. North Pacific Anadromous Fish Washington and Oregon. Canadian Journal Commission Bulletin 3. of Fisheries and Aquatic Sciences 49:1670- Beamish, R. J., B. L. Thomson, and G. A. Mc- 1685. Farlane. 1992. Spiny dogfish predation onBrodeur, R. D., K. W. Myers, and J. H. Helle. Chinook and coho salmon and the potential 2003. Research conducted by the United effects on hatchery-produced salmon. Trans- States on the early ocean life history of Pa- actions of the American Fisheries Society cific salmon. North Pacific Anadromous Fish 121:444-455. Commission Bulletin 3:89-13 1. Beaugrand, G., and P. C. Reid. 2003. Long-termBurgner, R. L., J. T. Light, L. Margolis, T. Oka- changes in phytoplankton, zooplankton and zaki A Tautz, and S. Ito. 1992. Distribution salmon related to climate. Global Change Bi- and origins of steelhead trout (Oncorhynchus ology 9:801-8 17. mykiss) in offshore waters of the North Pa- Bigler, R. S., D. W. Welch, and J. H Helle. 1996. cific Ocean. International North Pacific Fish- A review of size trends among North Pa- eries Commission Bulletin 51. cific salmon (Oncorhynchus spp.). CanadianChamberlain, F. M. 1907. Some observations on Journal of Fisheries and Aquatic Sciences salmon and trout in Alaska. United States Bu- 53:455-465. reau of Fisheries, Document No. 627, Wash- Blackbourn, D. J. 1985. The "salinity" factor and ington, D.C. the marine survival of Fraser River pinksChatters, J. C., V. L. Butler, M. J. Scott, D. M. and other stocks of salmon and trout. Pages Anderson, and D. A. Neitzel. 1995. A paleo- 67-75 in B. G. Shephard, editor. Proceed- science approach to estimating the effects of ings 1985 Northeast Pacific Pink and Chum climatic warming on salmonid fisheries of Salmon Workshop. Department of Fisheries the Columbia River basin. Pages 489-496 and Oceans, Vancouver. in R. J. Beamish, editor. Climate change and Blackbourn, D. J. 1987. Sea surface temperature northern fish populations. Canadian Special and pre-season predictions of return timing Publication of Fisheries and Aquatic Sci- in Fraser River sockeye salmon (Oncorhyn- ences 121. chus nerka). Canadian Special Publication ofChavez, F. P., C. A. Collins, A. Huyer, and D. L. Fisheries and Aquatic Sciences 96:269-306. Mackas, editors. 2002. Observations of the 24 PEARCY AND MCKINNELL

1997-98 El Niflo along the west coast of Northwest and Alaska Fisheries Center Pro- North America. Progress in Oceanography cessed Report 79-116, Seattle. 53. Favorite, F., A. J. Dodimead, and K. Nasu. 1976. Clemens, W. A. 1932. Pacific salmon migration: Oceanography of the subarctic Pacific Ocean The tagging of the spring salmon on the east region, 1960-71. International North Pacific coast of Vancouver Island in 1927 and 1928. Fisheries Commission Bulletin 33. Biological Board of Canada Bulletin 27. Finney, B. P., I. Gregory-Eaves, M. S. V. Douglas, Cooney, R. T., and R. D. Brodeur. 1998. Carrying and J. P. Smol. 2000. Fisheries productivity in capacity and North Pacific salmon produc- the northeastern Pacific Ocean over the past tion: stock-enhancement implications. Bul- 2,200 years. Nature (London) 416:729-733. letin of Marine Science 62:443-464. Fisher, J. P., and W. G. Pearcy. 1988. Growth of Davidson, F. A. 1937. Migration and homing of juvenile coho salmon (Oncorhynchus kisutch) Pacific salmon. Science 86(2220):55-56. in the ocean off Oregon and Washington, Davidson, F. A., and E. Vaughan. 1941. Rela- USA, in years of differing coastal upwelling. tion of population size to marine growth and Canadian Journal of Fisheries and Aquatic time of spawning migration in pink salmon Sciences 45:1036-1044. (Oncorhynchus gorbuscha) of southeasternFleming, R. H. 1955. Review of oceanography of Alaska. Journal of Marine Research 4:23 1- the northern Pacific. International North Pa- 246. cific Fisheries Commission 2:1-43. Dodimead, A. J., F Favorite, and T. Hirano. 1963.Foerster, R. E. 1955. The Pacific salmon (Ge- Review of oceanography of the subarctic Pa- nus Oncorhynchus) of the Canadian Pacific cific region. International North Pacific Fish- Coast, with particular reference to their oc- eries Commission Bulletin 13. currence in or near fresh water. International Dunn, J. R. 1996. Charles Henry Gilbert (1859- North Pacific Fisheries Commission Bulletin 1928): an early fishery biologist and his con- 1:5-56. tributions to knowledge of Pacific salmonFrancis, R. C., and S. R. Hare. 1994. Decadal-scale (Oncorhynchus spp.). Reviews in Fisheries regime shifts in large marine ecosystems of Science 4(2): 133-184. the northeast Pacific: a case for historical sci- Emlen, J. M., R. R. Reisenbichler, A. M. McGie, ence. Fisheries Oceanography 3:279-291. and T. E. Nickelson. 1990. Density-depen-Frank, K. T., B. Petrie, J. S. Choi, and W. C. dence at sea for coho salmon (Oncorhynchus Leggett. 2005. Trophic cascades in a for- kisutch). Canadian Journal of Fisheries and merly cod-dominated ecosystem. Science Aquatic Sciences 47:1765-1772. 308:1621-1623. Emmett, R. L., and R. D. Brodeur. 2000. RecentFrench, R. R., H. T. Bilton, M. Osaka, and A. changes in the pelagic nekton community off C. Hartt. 1976. Distribution and origin of Oregon and Washington in relation to some sockeye salmon (Oncorhynchus nerka) in physical oceanographic conditions. North offshore waters of the North Pacific Ocean. Pacific Anadromous Fish Commission Bul- International North Pacific Fisheries Com- letin 2:11-20. mission Bulletin 34. Favorite, F, and M. G. Hanavan. 1960. Oceano-Fresh, K. L. 1996. The role of competition and graphic conditions and salmon distribution predation in the decline of Pacific salmon south of the Alaska Peninsula and Aleutian and steelhead. Pages 245-275 in D. J. Stoud- Islands,1956. International North Pacific er, P. A. Bisson and R. J. Naiman, editors. Fisheries Commission Document 415. Pacific Salmon and their ecosystems, status Favorite, F., and T. Laevastu. 1979. A study of the and future options. Chapman and Hall, New ocean migrations of sockeye salmon and es- York. timation of the carrying capacity of the NorthGilbert, C. H. 1913. Contributions to the life-his- Pacific Ocean using a dynamical numerical tory of the sockeye salmon. Annual Report of salmon ecosystem model (NOPASA). NOAA the Commissioner of Fisheries, Number 1, Fisheries, National Marine Fisheries Service, Victoria. THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 25

Gilbert, C. H. 1914. Contributions to the life-his- evidence for North Pacific regime shifts in tory of the sockeye salmon. Paper 2. Annual 1977 and 1989. Progress in Oceanography Report of British Columbia Department of 42:103-146. Fisheries, Victoria. Hare, S. R., N. J. Mantua, and R. C. Francis. 1999. Gilbert, C. H. 19 14-1925. Contributions to the Inverse production regimes: Alaska and west life-history of sockeye salmon. Papers 1-10. coast Pacific salmon. Fisheries 24:6-14. Annual Reports British Columbia FisheriesHargreaves, N. B., and R. J. LeBrasseur. 1985. Department, 19 13-1924, Victoria. Species selective predation on juvenile pink Gilbert, C. H., and W. Rich. 1927. Investigations (Oncorhynchus gorbuscha) and chum salm- concerning the red-salmon runs to the Kar- on (0. keta) by coho salmon (0. kisutch). luk, River, Alaska. U.S. Bureau of Fisheries Canadian Journal of Fisheries and Aquatic Bulletin 43, Washington, D.C. Sciences 42:659-668. Godfrey, H. 1958. Comparisons of the index ofHartt, A. C. 1962. Movement of salmon in the return for several stocks of British Columbia North Pacific Ocean and Bering Sea as de- salmon to study variations in survival. Jour- termined by tagging, 1956-1958. Bulletin nal of the Fisheries Research Board of Cana- of the International North Pacific Fisheries da 15:891-900. Commission 6. Godfrey, H., K. A. Henry, and S. Machidori. 1975. Hartt, A. C. 1980. Juvenile salmonids in the ocean Distribution and abundance of coho salmon ecosystem-the critical first summer. Pages in offshore waters of the North Pacific Ocean. 25-67 in W. J. McNeil and D. C. Himsworth, International North Pacific Fisheries Com- editors. Salmonid ecoystems of the North Pa- mission Bulletin 31. cific. Oregon State University Press, Corval- Groot, C., and L. Margolis, editors. 1991. Pacific lis. Salmon Life Histories. UBC Press, Vancou-Hartt, A. C., and M. B. Dell. 1986. Early oceanic ver. migrations and growth of juvenile Pacific Groot, C., and T. P. Quinn. 1987. Homing mi- salmon and steelhead trout.International gration of sockeye salmon, Oncorhynchus North Pacific Fish Commission Bulletin 46. nerka, to the Fraser River. Fishery BulletinHasler, A. D., and W. J. Wisby. 1951. Discrimina- 85:455-469. tion of stream odors by fishes and its relation Groot, C., K. Cooke, G. Ellis, and R. Bailey. 1985. to parent stream behavior. American Natural- Data record of juvenile sockeye salmon and ist 85:223-238. other fish species captured by purse seineHealey, M. C. 1980. The ecology of juvenile and trawl in the Strait of Georgia, Johnstone salmon in Georgia Strait, British Columbia. Strait and Queen Charlotte Strait in 1982, Pages 203-239 in W. J. McNeil and D. C. 1983, 1984, and 1985. Canadian Data Report Himsworth, editors. Salmonid ecosystems of Fisheries and Aquatic Sciences 561. of the North Pacific, Oregon State University Gunsolus, R. T. 1978. The status of Oregon coho Press, Corvallis. and recommendations for managing the pro-Healey, M. C. 2000. Pacific salmon migrations in duction, harvest and escapement of wild and a dynamic ocean. Pages 29-54 in P. Harrison hatchery-reared stocks. Oregon Department and T. R. Parsons, editors. Fisheries ocean- of Fish and Wildlife Processed Report. ography: an integrative approach to fisheries Halpern, B. S., K. Cottenie, and B. R. Broitman. ecology and management. Volume 4. Black- 2006. Strong top-down control in southern well Scientific Publications Limited, Oxford, California kelp forest ecosystem. Science UK. 312:1230-1232. Healey, M. C., and C. Groot. 1987. Marine mi- Hamilton, K. 1985. A study of the variability of gration and orientation of ocean-type Chi- the return migration route of Fraser River nook and sockeye salmon. Pages 298-313 sockeye salmon (Oncorhynchus nerka). Ca- in M. J. Dadswell, R. J. Klauda, C. M. Mof- nadian Journal of Zoology 63:1930-1943. fitt, R. L. Saunders, R. A. Rulifson, and J. E. Hare, S. R., and N. J. Mantua. 2000. Empirical Cooper, editors. Common strategies of anad- 26 PEARCY AND MCKINNELL

romous and catadromous fishes. AmericanJordan, D. S. 1904. The salmon of the Pacific. Pa- Fisheries Society, Symposium 1, Bethesda, cific Fisherman 2:1. Maryland. Jordan, D. S., and C. H. Gilbert. 1887. The salm- Heard, W. R. 1998. Do hatchery salmon affect the on fishing and canning interests of the Pacific North Pacific Ocean ecosystem? North Pa- coast. Section V, volume 1. Pages 729-75 3 in cific Anadromous Fish Commission Bulletin G. B. Goode, editor. The fisheries and fishery Number 1:405-411. industries of the United States. Government Heard, W. R., J. A. Orsi, A. C. Wertheimer, M. Printing Office, Washington, D.C. V. Studevant, M. J. Murphy, D. G. Morten- Kaeriyama, M., M. Nakamura, R. Edpalina, J. son, B. L. Wing, and A. G. Celewycz. 2001. R. Bower, H. Yamaguchi, R. V. Walker, and A synthesis of research on early marine ecol- K. W. Myers. 2004. Change in feeding ecology of juvenile Pacific salmon in southeast ogy and trophic dynamics of Pacific salmon Alaska. North Pacific Anadromous Fisheries (Oncorhynchus spp.) in the central Gulf of Commission Technical Report 2:3-6. Alaska in relation to climate events. Fisheries Helle, J. H., and M. S. Hoffman. 1998. Changes in Oceanography 13:197-207. size and age at maturity of two North Ameri- Koslow, J. A., A. J. Hobday, and G. Boehlert. can stocks of chum salmon (Oncorhynchus 2002. Climate variability and marine survival keta) before and after a major regime shift in of coho salmon (Oncorhynchus kisutch) in the North Pacific Ocean. North Pacific Anad- the Oregon production area. Fisheries Ocean- romous Fish Commission Bulletin 1:81-89. ography 11:65-77. Helle, J. H., Y. Ishida, D. Noakes, and V. Rad- Larkin, P. A. 1975. Some major problems for fur- chenko, editors. 2000. Recent changes in ther study of Pacific salmon. International ocean production of Pacific salmon. North North Pacific Fisheries Commission 32:3-9. Pacific Anadromous Fish Commission Bul-Lawson, P. W., E. A. Logerwell, N. J. Mantua, R. letin Number 2. C. Francis, and V. N. Agostini. 2004. Envi- Henry, K. A. 1961. Racial identification of Fra- ronmental factors influencing freshwater sur- ser River sockeye by means of scales and its vival and smolt production in Pacific North- application to salmon management. Interna- west coho salmon (Oncorhynchus kisutch). tional Pacific Salmon Fisheries Commission Canadian Journal Fisheries and Aquatic Sci- Bulletin 12. ences 61:360-373. Hoitby, L. B., B. C. Anderson, and R. K. Kadowa-LeBrasseur, R. J.1966. Stomach contents of ki. 1990. Importance of smolt size and early salmon and steelhead trout in the northeast- ocean growth to interannual variability in ern Pacific Ocean. Journal of the Fisheries marine survival of coho salmon (Oncorhyn- Research Board of Canada 23:85-100. chus kisutch). Canadian Journal of FisheriesLeBrasseur, R. J. 1972. Utilization of herbivo- and Aquatic Sciences 47:2181-2194. rous zooplankton by maturing salmon. Pages Ishida, Y., S. Ito, M. Kaeriyama, S. McKinnell, 58 1-588 in A. Y. Takenouti, editor. Biologi- and K. Nagasawa. 1993. Recent changes in cal oceanography of the northern North Pa- age and size of chum salmon (Oncorhynchus cific Ocean. Idemitsu Shoten, Tokyo. keta) in the North Pacific Ocean and pos-Levin, P. S., R. W. Zabel, andJ. G. Williams. 2001. sible causes. Canadian Journal Fisheries and The road to extinction is paved with good in- Aquatic Sciences 50:290-295. tentions: negative association of fish hatcher- Jaenicke, H. W., and A. G. Celewycz. 1994. Ma- ies and threatened salmon. Proceedings of the rine distribution and size of juvenile Pacific Royal Society of London 286B: 1153-1158. salmon in southeast Alaska and northern Brit- Lichatowich, J. 1999. Salmon without rivers: a ish Columbia. Fishery Bulletin 92:79-90. history of the Pacific salmon crisis. Island Johnson, S. L. 1988. The effects of the 1983 El Press, Washington D.C. Niño on Oregon's coho (OncorhynchusLogerwell, E. A., N. Mantua, P. W. Lawson, R. kisutch) and Chinook (0. tshawytscha) salm- C. Francis, and V. N. Agostini. 2003. Track- on. Fisheries Research 6:105-123. ing environmental processes in the coastal THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 27

zone for understanding the predicting Oregon versity of Washington, Fisheries Research coho (Oncorhynchus kisutch) marine surviv- Institute, FRI-UW-9614, Seattle. al. Fisheries Oceanography 12:554-568. Neave, F., T. Yonemori, and R. G. Bakkkala. 1976. Major, R. L., J. Ito, S. Ito, and H. Godfrey. 1978. Distribution and origin of chum salmon in Distribution and origin of Chinook salmon offshore waters of the North Pacific Ocean. (Oncorhynchus tshawytscha) in offshore wa- International North Pacific Fisheries Com- ters of the North Pacific Ocean. International mission Bulletin 35. North Pacific Fish Commission Bulletin 38. Nickelson, T. E. 1986. Influences of upwelling, Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wal- ocean temperature, and smolt abundance on lace, and R. C. Francis. 1997. A Pacific inter- marine survival of coho salmon (Oncorhyn- decadal climate oscillation with impacts on chus kisutch) in the Oregon production area. salmon production. Bulletin of the American Canadian Journal of Fisheries and Aquatic Meteorological Society 78:1069-1079. Sciences 43:527-535. Manzer, J. I., T. Ishida, A. E. Peterson, and M. G. O'Malley, H., and W. H. Rich. 1919. Migration Hanavan. 1965. Salmon of the North Pacific of adult sockeye salmon in Puget Sound and Ocean. Part. 5: offshore distribution of salm- Fraser River. U.S. Bureau of Fisheries Docu- on. International North Pacific Fish Commis- ment 873. sion Bulletin 15. Parker, R. R. 1962. A concept of the dynamics of McKinnell, S. 1995. Age-specific effects of sock- pink salmon populations. Pages 203-211 in eye abundance on the adult body size of ma- N. J. Wilimovsky, editor. Symposium on pink jor British Columbia sockeye stocks. Canadi- salmon. H.R. MacMillan Lectures in Fisher- an Journal of Fisheries and Aquatic Sciences ies. University of British Columbia, Vancou- 52:1050-1063. ver. McKinnell, S., H. J. Freeland, and S. Groulx.Parker, R. R. 1968. Marine mortality schedules of 1999. Assessing the northern diversion of pink salmon of the Bella Coola River, central sockeye salmon returning to the Fraser River. British Columbia. Journal of the Fisheries Fisheries Oceanography 8:104-114. Research Board of Canada 25:757-794. McKinnell, S., J. J. Pella, and M. L. Dahlberg. Parker, R. R. 1971. Size selective predation among 1997. Population-specific aggregations of juvenile salmonid fishes in a British Colum- steelhead trout (Oncorhynchus mykiss) in the bia inlet. Journal of the Fisheries Research North Pacific. Canadian Journal of Fisheries Board of Canada 28: 1503-1510. and Aquatic Sciences 54:2368-2376. Payne, J. T., A. W. Wood, A. F. Hamlet, R. N. Mote, P. W., E. A. Parson, A. F. Hamlet, W. S. Palmer, and D. P. Lettenmaier. 2004. Mitigat- Keeton, D. Lettenmaier, N. Mantua, E. L. ing the effects of climate change on water re- Miles, D. W. Peterson, D. L. Peterson, R. sources of the Columbia River basin. Climate Slaughter, and A. K. Snover. 2003. Prepar- Change 62:233-256. ing for climate change: the water, salmon,Pearcy, W. G. 1992. Ocean ecology of North Pa- and forests of the Pacific Northwest. Climate cific salmonids. University of Washington, Change 61:45-88. Washington Sea Grant Program, Seattle. Mueter, F. J., B. J. Pyper, and R. M. Peterman.Pearcy, W. G. 1996. Salmon production in chang- 2005. Relationships between coastal ocean ing ocean domains. Pages 33 1-352 in D. J. conditions and survival rates of northeast Pa- Stouder, P. A. Bisson and R. J. Naiman, edi- cific Salmon at multiple lags. Transactions tors. Pacific salmon and their ecosystems: of the American Fisheries Society 134:105- status and future options. Chapman and Hall, 119. New York. Myers, K. W., K. Y Aydin, and R. V. Walker.Pearcy, W. G., R. D. Brodeur, J. M. Shenker, W. 1996. Known ocean ranges of stocks of Pa- W. Smoker, andY. Endo. 1988. Food habits of cific salmon and steelhead as shown by tag- Pacific salmon and steelhead trout, midwater ging experiments, 1956-1995. North Pacific trawl catches and oceanographic conditions Anadromous Fisheries Document 192. Uni- in the Gulf of Alaska, 1980-1985. Bulletin of 28 PEARCY AND MCKINNELL

the Ocean Research Institute, University ofReimers, P. E. 1973. The length of residence of Tokyo 26(2):2-78. juvenile fall Chinook salmon in Sixes River, Pearcy, W. G., J. Fisher, R. Brodeur, and S. John- Oregon. Research Report of the Fish Com- son. 1985. Effects of the 1983 El Niño on mission of Oregon 4(2). coastal nekton off Oregon and Washington.Rich, W. H. 1939. Local populations and migra- Pages 188-204 in W. S. Wooster and D. L. tion in relation to the conservation of Pacific Fluharty, editors. El Niño north. Niño effects salmon in the western United States and Alas- in the eastern subarctic Pacific. Washington ka. Migration and conservation of salmon. Sea Grant Program, Seattle. American Association for the Advancement Perry, R. I., D. W. Welch, P. J. Harrison, D. L. of Science Publication 8:45-47. Mackas, and K. L. Denman. 1998. Epipelagic Ricker, W. H. 1958. Maximum sustained yields fish production in the open subarctic Pacific: from fluctuating environments and mixed bottom up or self-regulating control. PICES stocks. Journal of the Fisheries Research Press 6:26-32. Board of Canada 15:991-1006. Peterman, R. M. 1984. Density-dependent growthRicker, W. E. 1972. Hereditary and environmen- in early ocean life of sockeye salmon (On- tal factors affecting certain salmonid popula- corhynchus nerka). Canadian Journal of Fish- tions. Pages 19-160 in R. C. Simon and P.A. eries and Aquatic Sciences 41:1825-1 829. Larkin, editors. The stock concept of Pacific Peterman, R. M. 1987. Review of the components salmon. H.R. McMillan Lectures in Fisheries. of recruitment of Pacific salmon. Pages 417- University of British Columbia, Vancouver. 429 in M. J. Dodswell, R. J. Klauda, C. M. Ricker, W. E. 1981. Changes in the average size Moffitt, R. L. Saunders, R. A. Rulifson and and average age of Pacific salmon. Canadian J. E. Cooper, editors. Common strategies of Journal of Fisheries and Aquatic Sciences anadromous and catadromous fishes. Ameri- 38:1636-1656. can Fisheries Society, Symposium 1, Bethes-Roemmich, D., and J. McGowan. 1995. Climate da, Maryland. warming and the decline of zooplankton in Peterman. R. M. 1991. Density-dependent marine the California Current. Science 267:1324- processes in North Pacific salmonids: lessons 1326. for experimental design of large-scale manip-Rogers, D. E. 1980. Density-dependent growth of ulations of fish stocks. ICES Marine Science Bristol Bay sockeye salmon. Pages 267-283 Symposia 192:69-77. in W. J. McNeil and D. C. Himsworth, edi- Peterson, W. T., R. D. Brodeur, and W. G. Pearcy. tors. Salmonid ecosystems of the North Pa- 1982. Food habits of juvenile salmon in the cific. Oregon State University Press, Corval- Oregon coastal zone, June 1979. Fishery Bul- lis. letin 80:841-85 1. Roppel, P. 2004. The steamer Albatross and early Pyper, B. J., and R. M. Petennan. 1999. Relation- Pacific salmon, Oncorhynchus spp. research ship among adult body length, abundance, in Alaska Marine Fisheries Review 66(3):21- and ocean temperature for British Columbia 31. and Alaska sockeye salmon (OncorhynchusRoyal, L.A., and J. P. Tully. 1961. Relationship of nerka), 1967-1997. Canadian Journal Fisher- variable oceanographic factors to migrations ies and Aquatic Sciences 56:1716-1720. and survival of Fraser River salmon. Califor- Quinn, W. H., D. 0. Zopf, K. S. Short, and R. T. nia Cooperative Oceanic Fisheries Investiga- W. K. Yang 1987. Historical trends and sta- tions 8:65-68. tistics of the Southern Oscillation, El Niño,Ruggerone, 0. T., and J. L. Nielsen. 2004. Evi- and Indonesian droughts. Fisheries Bulletin dence for competitive dominance of pink 76:663-678. salmon (Oncorhynchus gorbuscha) over oth- Quinn, T. P. 2005. The behavior and ecology of er salmonids in the North Pacific Ocean. Re- Pacific salmon and trout. American Fisheries views in Fish Biology and Fisheries 14:371- Society, Bethesda, Maryland and University 390. of Washington Press, Seattle. Ruggerone, 0. T., M. Zimmermann, K. W. Myers, THE OCEAN ECOLOGY OF SALMON IN THE NORTHEAST PACIFIC OCEAN 29

J. L. Nielsen, and D. E. Rogers. 2003. Com-Sweeting, R. M., R. J. Beamish, D. J. Noakes, and petition between Asian pink salmon (On- C. M. Neville. 2003. Replacement of wild corhynchus gorbuscha) and Alaskan sockeye coho salmon by hatchery-reared coho salmon salmon (0. nerka) in the North Pacific Ocean. in the Strait of Georgia over the past three de- Fisheries Oceanography 12:209-219. cades. North American Journal of Fisheries Sanger, G. A. 1972. Fishery potentials and esti- Management 23:492-502. mated biological productivity of the subarctic Tadokoro, K., Y. Ishida, N. D. Davis, S. Uey- Pacific region. Pages 56 1-574 in A. Y. Tak- anagi, and T. Sugimoto. 1996. Change in enouti, editor. Biological oceanography of chum salmon (Oncorhynchus keta) stomach the northern North Pacific Ocean. Idemitsu contents associated with fluctuation of pink Shoten, Tokyo. salmon (0. gorbuscha) abundance in the cen- Scarnecchia, D. L. 1981. Effects of streamfiow tral subarctic Pacific Ocean and Bering Sea. and upwelling on yield of wild coho salmon Fisheries Oceanography 5:89-99. (Oncorhynchus kisutch) in Oregon. Canadian Takagi, K., K. V. Aro, A. C. Hartt, and M. B. Dell. Journal of Fisheries and Aquatic Sciences 1981. Distribution and origin of pink salmon 38:471-475. (Oncorhynchus gorbuscha) in offshore wa- Seeb, L. W., P. A. Crane, C. M. Kondzela, R. L. ters of the North Pacific Ocean. International Wilmot, S. Urawa, N. V. Varnavskaya, and J. North Pacific Fisheries Commission Bulletin E. Seeb. 2004. Migration of Pacific rim chum 40. salmon on the high seas: insights from ge-Tanaka, S. 1962. Studies of the question of re- netic data. Environmental Biology of Fishes production of salmon stocks. International 69:21-36. North Pacific Fisheries Commission Bulletin Sette, 0. E., and J. D. Isaacs, editors. 1960: Sym- 9:85-90. posium on the changing Pacific Ocean inThomson, K. A., W. J. Ingraham, Jr., M. C. Heal- 1957 and 1958. California Cooperative Fish- ey, P. F. LeBlond, C. Groot, and C. G. Heal- eries Investigations Reports 7:13-217. ey. 1992. The influence of ocean currents on Sheppe, W. 1962. First man west; Alexander latitude of landfall and migration speed of McKenzie's journal of his voyage to the Pa- sockeye salmon returning to the Fraser River. cific coast of Canada in 1793. McGill Univer- Fisheries Oceanography 1:163-179. sity Press, Montreal. U.S. GLOBEC. 1996. Report on climate change Shiomoto, A., K. Tadokoro, K. Nagasawa, and Y. and carrying capacity of the North Pacific Ishida. 1997. Trophic relations in the subarc- ecosystem. U.S. GLOBEC Report 15. U.S. tic North Pacific ecosystem: possible feeding GLOBEC Scientific Steering Committee Co- effect from pink salmon. Marine Ecology ordinating Office, University of California, Progress Series 150:75-85. Berkeley. Straty, R. R. 1974. Ecology and behavior of juve- Van Hyning, J. M. 1973. Factors affecting the nile sockeye salmon (Oncorhynchus nerka) abundance of fall Chinook salmon in the Co- in Bristol Bay and the eastern Bering Sea. lumbia River. Research Reports of the Fish Pages 285-320 in D. W. Hood and E. J. Kel- Commission of Oregon 4(l):2-87. ley, editors. Oceanography of the Bering Sea.Vernon, E. H. 1958. An examination of factors af- University of Alaska, Fairbanks. fecting the abundance of pink salmon in the Straty, R. R. 1975. Migratory routes of adult Fraser River. International Pacific Salmon sockeye salmon, Oncorhynchus nerka, in the Fisheries Commission, Progress Report 5, eastern Bering Sea and Bristol Bay. NOAA New Westminster, British Columbia. Technical Report NMFS-SSRF-690. Walker, R. V., and K. W. Myers. 1992. Stock iden- Sugimoto, T., and K. Tadokoro. 1997. Interannu- tification techniques for salmon on the high al-interdecadal variations in zooplankton bio- seas. Northwest Science 66(2). mass, chlorophyll concentration and physical Walker, R. V., K. W. Myers, N. D. Davis, K. Y. environment in the subarctic Pacific and Ber- Aydin, K. D. Friedland, H. R. Carlson, G. W. ing Sea. Fisheries Oceanography 6:74-93. Boehlert, S. Urawa, Y. Ueno, and G. Anma. 30 PEARCY AND MCKINNELL

2000. Diurnal variation in thermal environ- competitive overlap for Pacific salmon(On- ment experienced by salmonids in the North corhynchusspp.). Fisheries Oceanography Pacific as indicated by data storage tags. 2:11-23. Fishery Oceanography 9:171-1 86. Welch, D. W., B. R. Ward, and S. D. Batten. 2004. Walters, C. J., R. Hilborn, R. M. Peterman, and Early ocean survival and marine movements M. J. Staley. 1978. Model for examining ear- of hatchery and wild steelhead trout(On- ly ocean limitation of Pacific salmon produc- corhynchus mykiss)determined by an acous- tion. Journal of the Fisheries Research Board tic array: Queen Charlotte Strait, British Co- of Canada 35:1303-13 15. lumbia. Deep-Sea Research II 5 1:897-909. Ware, D. M., and R. E. Thomson. 2005. Bottom-Wickett, W. P. 1958. Review of certain environ- up ecosystem trophic dynamics determine mental factors affecting the production of pink fish production in the Northeast Pacific. Sci- and chum salmon. Journal of the Fisheries Re- ence 308:1280-1284. search Board of Canada 15:1103-1 126. Weitkamp,L. A., and K. Neely. 2002. Coho salm- Wickett, W. P.1977. Relationship of coastal on(Oncorhynchus kisutch)ocean migration oceanographic factors to the migration of patterns: insight from marine coded-wire tag Fraser River salmon. International Commis- recoveries. Canadian Journal of Fisheries and sion Exploration of the Sea CM 19771M:26. Aquatic Sciences 59:1100-1115. Willette, T. M., R. T. Cooney, V. Patrick, D. M. Welch, D., J. Morris, M. Trudel, M. Thiess, T. Mason, G. L. Thomas, and D. Scheel. 2001. Zubkowski, M. Jacobs, P. Winchell, and H. Ecological processes influencing mortality of MacLean; 2003. A summary of Canadian juvenile pink salmon(Oncorhynchus gorbus- high seas salmon surveys in the Gulf of Alas- cha)in Prince William Sound, Alaska. Fisher- ka, 1995-2003. North Pacific Anadromous ies Oceanography 10(Supplement 1): 14-41. Fish Commission Document 712. Wooster, W. S., and D. L Fluharty, editors. 1985. Welch, D. W. 1997. Anatomical differences in the El Niflo north. Niño effects in the eastern sub- gut structure of Pacific salmon(Oncorhyn- arctic Pacific Ocean. University of Washing- chus):evidence for oceanic limits to salmon ton, Washington Sea Grant Program, Seattle. production? Canadian Journal of ZoologyWorm, B., and R. A. Myers. 2003. Meta-analy- 75:936-942. sis of cod-shrimp interactions reveals top- Welch, D. W., and T. R. Parsons. 1993.13C-615N down control in oceanic food web. Ecology values as indicators of trophic position and 84:162-173.