The first workshop on the effects of ocean conditions on salmonid populations was held here in Newport, Oregon 13 years ago, in November 1983, just after the biggest El Niño of the century. The timing could not have been more appropriate because of the devastating effects of this ocean event on the survival and growth of coho (Oncorhynchus kisutch) and chinook (O. tshawytscha) salmon off Oregon and California. Warm seas farther north during this same period, however, were correlated with near record catches of Alaskan salmon. I believe that 1983 was the year of a paradigm shift—ocean factors were generally recognized as important factors to salmon survival and production.

The return of many of the participants of this first workshop (with precise homing back to Newport) is indeed gratifying, as is the attendance of marine scientists, freshwater fishery biologists, meteorologists, administrators, and environmentalists.

First to note is the general realization that ocean conditions are indeed important. This meeting is testimony to this recognition, and to the important progress toward understanding the relationships between the ocean environment and salmon survival. But with that progress, the complexity of responses, both physical and biological, in different scales of time and space, are clearly emerging. In my opinion, major changes in our thinking over the past 10 to 15 years can be grouped into four categories: El Niños, interdecadal climate shifts, hatcheries, and ocean carrying capacity.

The big 1982-83 El Niño event had pronounced effects on salmon growth and survival as well as the species composition of zooplankton and of nektonic competitors and predators of salmon in coastal waters (Pearcy 1992). Since the late 1970s, the Southern Oscillation Index has remained low for an unprecedented duration, culminating in the prolonged 1991-95 El Niño period (Trenberth and Hurrell 1995; Hargreaves, Beamish, and Neville this volume) which coincided with the continued decline of some southern stocks in the 1990s.

Although poor ocean conditions are often blamed on El Niños, we now realize the significance of interdecadal changes in climatic forcing. Progress on these regime shifts has been impressive since 1983. Papers by McLain (1984), Rogers (1984), and Chelton (1984) presented at the first workshop identified the abrupt change in ocean temperatures that occurred during the late 1970s, and Rogers (1984) pointed out the pronounced increase in production of Alaskan salmon after 1976. However, the more recent papers by Brodeur and Ware (1992), Francis and Hare (1994), Beamish and Bouillon (1995), and Hare and Francis (1995) cogently correlated major biological and physical changes associated with this large-scale regime shift, especially for the Alaskan fisheries and the Gulf of Alaska. Since the late 1970s, salmon catches in Alaska have exceeded historic highs. But for many stocks in the southern portions of their ranges, off California and Oregon, ocean conditions have remained generally unfavorable and populations are extremely low. The underlying mechanism for these low-frequency shifts in climate variability remains uncertain (see Ware 1995).

Hatcheries are no longer perceived as a panacea for increased production as they were in earlier years. Now they are considered as one of the causes for declining populations of wild stocks. In 1983, we had a working group on hatcheries and several papers were presented on saltwater releases of salmon by authors who were employed by private salmon ranching companies. It is revealing that there is no working group on hatcheries at this meeting, and private salmon ranching is defunct in Oregon. The emphasis today is clearly on restoration of declining wild stocks rather than on hatcheries.

Although the oceanic or high seas carrying capacity for salmon has been evaluated (see Pearcy 1992 for review), and both Rogers (1980) and Peterman (1984) reported negative correlations between numbers and size of sockeye salmon, this issue has recently returned to the fore—even more evidence is accumulating that the size of returning adult salmon has decreased and age at maturity has increased over much of their range (e.g., Kaeriyama 1989, Ishida et al. 1993, Helle and Hoffman 1995).

Based on the recommendations of the 1983 workshop, my own personal biases, and the results of this conference, here are some of my recommendations for future research on ocean-salmon interactions. Many of these recommendations will be amplified in the reports of the individual working groups.

Three basic questions need to be addressed to reveal how ocean variability affects growth and survival. These are the "when," "where," and "how" questions. When in the life history is the most variable and highest mortality? Where—in estuarine, coastal, or oceanic environments? How does mortality vary in time and space? And what are the mechanisms that result in changes in growth and mortality? We need more than correlations to predict changes in production and to assist managers. To answer these questions we need the following types of information.

We need precise and accurate time series for statistics both on salmon and on ocean conditions. These include data on catch and escapement of adults, and numbers of smolts entering the ocean, as well as data on size and age of both smolts and adults, in order to a) study variations in marine survival and growth, b) compare trends in survival among species, among stocks, and between hatchery and wild fish, and c) determine if similar trends are related to where smolts enter the ocean or where they inhabit the ocean later in life.

We need to improve our synoptic data sets and time series for physical, atmospheric, and biological factors that will be needed to relate salmon growth and survival to ocean conditions.

More studies are required to determine the survival success of different life history strategies within a stock (e.g., Reimers 1973). But such studies should be repeated for several years to determine interannual variability in the survival of life history groups and how they may relate to changing ocean conditions. Similarly, releases of smolts at different times, at different sizes, and at different locations should be compared with relevant ocean conditions.

We need better information on distributions of stocks in coastal and oceanic waters, both horizontally and vertically; how their ocean environments change over time; and how distributions of stocks overlap. When and where and under what conditions are growth rates suppressed? To answer these questions we need to identify fish with tags (including acoustical and archival tags), thermal marks, scale patterns, genetic markers, etc.

In my opinion, partitioning survival between freshwater and marine phases is an extremely urgent problem. As we have seen at this workshop, climatic shifts result in wet/cool vs. dry/warm periods that affect both the ocean and terrestrial habitats (Anderson this volume). Although we can assume that many hatchery smolts migrate short distances to sea, hence their post-smolt mortality is mainly in the ocean, we cannot assume this for wild stocks. Here, freshwater conditions are much more important (Bradford this volume). Therefore, we need to monitor numbers of wild smolts entering the ocean and the catch and escapement of surviving adults of "indicator" stocks so that both freshwater and marine survival rates can be estimated (similar to the long-term research at Carnation Creek). This will provide basic information on survival rates in these two environments and will allow evaluation of restoration efforts and recovery of depleted stocks. Estuarine mortality is undoubtedly important for some species and stocks and also needs to be studied by serial marking, mark and recapture, and monitoring smolts entering and leaving the estuary.

We need experiments to test hypotheses of density-dependent smolt mortality in the ocean and effects of hatchery releases on wild fish by releasing a large range in numbers of hatchery smolts, either in different locations in the same year or large and small releases in the same region during the same year.

Modeling can be useful in identifying critical processes that affect mortality, growth, or distributions. Examples from this workshop are the interactions between ocean circulation and distributions of salmon (Rand et al. this volume) and bioenergetic modeling (Brodeur this volume).

To understand mechanisms, we must go to sea. Integrated research on salmonid ecosystems is needed in both coastal and oceanic waters. In 1983, we recommended process-oriented research on a few selected stocks (Oregon Production Index (OPI) coho salmon and Bristol Bay sockeye salmon) to elucidate mechanisms linking ocean growth and survival to ocean conditions. Such research should include study of ocean distribution and migration, growth, feeding ecology, potential competitors and predators, and alternative prey of predators. Today, the only comprehensive study of the early ocean life of salmon is the SEA program in Prince William Sound (Cooney this volume).

Predation, in my opinion, is the most likely cause of high mortality rates of juvenile salmon at sea. Because smolts are only available to marine predators for short periods of the year, most predators must sustain themselves on more abundant forage animals. Therefore, the availability of alternate prey may be important in regulating predation rates on juvenile salmon. When usually abundant forage is not available due to unfavorable ocean conditions, predation on juvenile salmon is probably intensified. Moreover, as we have heard at this meeting, migratory piscivorous fishes such as mackerel and Pacific hake (Merluccius productus) may migrate into coastal waters in large numbers and prey on salmon smolts during warm ocean conditions (Hargreaves this volume).

One of the strong recommendations from 1983 was for long-term monitoring of the coastal ocean so that ocean dynamics can be related to trends of salmon survival and growth. The currently available time series from shore stations, on sea level, sea surface temperatures, and occasionally sea surface salinity, are inadequate for understanding mechanisms influencing survival and growth. For example, the relationship between upwelling intensity during the summer and OPI hatchery coho salmon survival was significant and positive from 1960 to 1975 (Nickelson 1986) but was insignificant and negative after 1976 (Pearcy in press). This difference is probably due to changes in the effectiveness of upwelling in recharging the euphotic zone with nutrients. After 1976, a deep stratified layer of warm water often persisted in coastal waters off southern California (Roemmich and McGowan 1995), and probably farther north in the California Current system. We need routine cruises by research vessels that monitor ocean conditions along predesignated transects from the coast. Such cruises, along with data on currents, sea temperatures, and phytoplankton from satellite measurements, drifters, and moored buoys, would provide relevant data on subsurface oceanography, circulation dynamics, and water types, and on factors affecting plankton productivity.

Also needed are basin scale studies of circulation to evaluate variability in large-scale physical forcing. The Chelton and Davis (1982) model showing that the California and Alaska current systems are linked and out-of-phase should be tested. This is an important concept in understanding regime shifts and inputs of subarctic waters into coastal waters (see Parrish this volume for alternative ideas). Deployment of moored and drifting buoys, satellite measurements, and ships of opportunity could all be used to further our knowledge of how circulation is related to general trends of salmonid survival and abundance.

Beamish, R. J., and D. R. Bouillon. 1995. Marine fish production trends off the Pacific coast of Canada and the United States. In R. J.Beamish (editor), Climate change and northern fish populations, p. 585-591. Can. Spec. Publ. Fish. Aquat. Sci. 121.

Brodeur, R. D., and D. M. Ware. 1992. Long-term variability in zooplankton biomass in the subarctic Pacific Ocean. Fish. Oceanogr. 1:32-38.

Chelton, D. B. 1984. Short-term climatic variability in the Northeast Pacific Ocean. In W. G. Pearcy (editor), The influence of ocean conditions on the production of salmonids in the North Pacific, p. 87-99. Oregon Sea Grant Program, Oregon State Univ., Corvallis. Publ. ORESU-W-83-001.

Chelton, D. B., and R. E. Davis. 1982. Monthly mean sea level variability along the west coast of North America. J. Phys. Oceanogr. 12:757-789.

Francis, R. C., and S. R. Hare. 1994. Decadal-scale regime shifts in the large marine ecosystems of the North-east Pacific: A case study for historical science. Fish. Oceangr. 3:279-291.

Hare, S. R., and R. C. Francis. 1995. Climate change and salmon production in the Northeast Pacific Ocean. In R. J. Beamish (editor), Climate change and northern fish populations, p. 357-372. Can. Spec. Publ. Fish. Aquat. Sci. 121.

Helle, J. H., and M. S. Hoffman. 1995. Size decline and older age at maturity of two chum salmon (Oncorhynchus keta) stocks in western North America, 1972-92. In R. J. Beamish (editor), Climate change and northern fish populations, p. 245-260. Can. Spec. Publ. Fish. Aquat. Sci. 121.

Ishida, Y., S. Ito, M. Kaeriyama, S. McKinnell, and K. Nagasawa. 1993. Recent changes in the age and size of chum salmon (Oncorhynchus keta) in the North Pacific Ocean and possible causes. Can. J. Fish. Aquat. Sci. 50:290-295

Kaeriyama, M. 1989. Aspects of salmon ranching in Japan. Physiol. Ecol. Japan. Spec. Vol. 1:625-638.

McLain, D. R. 1984. Coastal ocean warming in the Northeast Pacific, 1976-83. In W. G. Pearcy (editor), The influence of ocean conditions on the production of salmonids in the North Pacific, p. 61-86. Oregon Sea Grant Program, Oregon State Univ., Corvallis. Publ. ORESU-W-83-001.

Nickelson, T. E. 1986. Influences of upwelling, ocean temperature, and smolt abundance on marine survival of coho salmon (Oncorhynchus kisutch) in the Oregon Production Area. Can. J. Fish. Aquat. Sci. 43:527-535.

Pearcy, W. G. 1992. Ocean ecology of North Pacific salmonids. Washington Sea Grant Program, University of Washington Press, Seattle, 179 p.

Pearcy, W. G. In press. Salmon production in changing ocean domains. In D. J. Stouder, P. A. Bisson, and R. J. Naiman (editors), Pacific salmon and their ecosystems: Status and future options, p. 331-354. Chapman and Itall, New York.

Peterman, R. M. 1984. Density-dependent growth in early ocean life of sockeye salmon (Oncorhynchus nerka). Can. J. Fish. Aquat. Sci. 41:1825-1829.

Reimers, P. E. 1973. The length of residence of juvenile fall chinook salmon in Sixes River, Oregon. Res. Rep. Fish Comm. Oregon 4(2):1-43.

Roemmich, D., and J. McGowan. 1995. Climatic warming and the decline of zooplankton in the California Current. Science 267:1324-1326.

Rogers, D. E. 1980. Density-dependent growth of Bristol Bay sockeye salmon. In W. J. McNeil and D. C. Himsworth (editors), Salmonid ecosystems of the North Pacific, p. 267-283. Oregon State Univ. Press, Corvallis.

Rogers, D. E. 1984. Trends in abundance of northeastern Pacific stocks of salmon. In W. G. Pearcy (editor), The influence of ocean conditions on the production of salmonids in the North Pacific, p. 100-127. Oregon Sea Grant Program, Oregon State Univ., Corvallis. Publ. ORESU-W-83-001.

Trenberth, K. E., and J. W. Hurrell. 1995. Decadal coupled atmosphere-ocean variations in the North Pacific Ocean. In R. J. Beamish (editor), Climate change and northern fish populations, p.15-24. Can. Spec. Publ. Fish. Aquat. Sci. 121.

Ware, D. M. 1995. A century and a half of change in climate of the NE Pacific. Fish. Oceanogr. 4:267-277.

Table of Contents