North Pacific Anadromous Fish Commission Technical Report #18. . . is the summary of what scientists found so far from ocean expeditions to the North Pacific in the study of Pacific salmon.
Virtual Conference on Winter Ecology of Pacific Salmon and Results from the Two Gulf of Alaska Expeditions April 20–22, 2021 in Canada and USA April 21–23, 2021 in Japan, Korea, and Russia
The highlights of this extensive report are as follows:
1 - pink salmon are much farther south in the winter and not competing for food with sockeye as scientists thought. This is important as these scientists thought that hatchery pink salmon were affecting sockeye salmon abundance.
2- coho and chum salmon were observed to be in schools. One large coho school had populations of coho from Oregon to Alaska. This is amazing as they had to find each other.
3- there were not many predators and this is important as scientists considered that predators were controlling abundance.
4 - steelhead live in the top few meters of the ocean and thus are affected more than other salmon by ecosystem changes.
5 - the abundances of BC chum salmon were very small possibly providing a method of forecasting adult returns several years ahead of the return.
6 - juvenile Fraser River sockeye salmon in their first ocean winter can be way out in the middle of the Pacific and much farther west than previously thought. You can read about all of this in the report.
Dick Beamish
The NPAFC technical report contains this as a high-level summary:
1. The 2019 and 2020 winter expeditions to the Gulf of Alaska were the first Canadian studies of this kind since the late 1960s and the first to include an international team of researchers. All researchers volunteered their participation and agreed that all data would be publicly available (see at https://iys.hakai.org/dataset). The two expeditions were the first high seas Pacific salmon research to be privately funded and organized.
2. The research applied a variety of methodologies to study the salmon and their ecosystem:
a. DNA stock identification (including first trials of on-board analyses, Deeg et al. 2021);
b. Environmental-DNA sampling (Deeg et al. 2021);
c. Pathogens and health (Deeg et al. 2022);
d. Body composition studies to examine fish condition (Waters et al. 2022);
e. Use of stable isotopes to examine dietary overlap between species (see Espinasse et al. 2020); and,
f. A comprehensive array of oceanographic measurements.
There was evidence that coho salmon can form large schools in the winter with fish from populations ranging from Oregon to Alaska.
3. The first estimates of the total population of Pacific salmon within the survey areas were made using the method developed by Russia and applied to our data by A. Somov (TNIRO, pers. comm.). When estimates were adjusted for equal sampling areas, the total population estimate was essentially identical at 54.95 million salmon in 2019 and 55.2 million salmon in 2020, over an area of 697,500 km2. It was fortuitous that the studies occurred during an unpreceded marine heat wave in the Gulf of Alaska. These abundance estimates were made at a time in 2019 when there were historic poor salmon returns to British Columbia in the fall of 2019, and in 2020 when there was a collapse of the salmon catches in all countries resulting in a total commercial catch equal to low catches in the early 1980s.
4. Extensive oceanographic survey (Pakhomov et al. 2022) plus our biological sampling enabled us to begin to associate winter environments with salmon abundance, distribution and diet. A major objective of the expeditions was to understand how climate and ocean conditions that affected growth in the coastal ocean was related to salmon survival in the first ocean winter. Pacific salmon species are different and behave differently in the winter ocean, therefore decreasing the interspecific competition between them (Radchenko 2022). Ross and Pena (2022) provided an historical perspective of the ocean environment in 2019 and 2020 based on Canada’s longest time series of Pacific Ocean observations along Line P and at Canada’s Weather Station Papa (50oN, 145oW). In total, sampling along Line P now provides 75 years of detailed oceanographic data (Pena and Bograd 2007).
5. Our collaborations between countries demonstrated the value of international cooperation, particularly in studying Pacific salmon in an open environment the breadth of the North Pacific Ocean. These surveys were highly informative and demonstrated the utility of trawl nets to sample Pacific salmon in the deep-water marine environments, but there are certainly further questions to address. Principle in them is the effectiveness of the trawls to representatively sample the fish community in the surface layers, including the species composition of salmon and in the broader fish community. It was notable that trawl catches did not include many salmon predators or competitors in both the 2019 and 2020 expeditions. In 2022, a charter vessel will utilize Japanese research gillnets to compare catches with trawlers fishing at similar times and places.
The 148 Page Technical Report #18 (condensed below) begins with the following:
"The abundance estimates of all Pacific salmon in the survey area in 2019 and 2020 were made using the methods of Russian scientists (Volvenko 1999, 2000). The estimates of 54.95 million fish in 2019 and 51.3 million in 2020 were similar but much lower than expected. However, commercial salmon catches in British Columbia in 2019 and 2020 were at historic lows and catches in Southeast Alaska were at low levels in 2019 and 2020. Chum salmon returns to Japan also continued their decline in 2019 and 2020. In addition, there was a basin-scale collapse of all Pacific salmon commercial catches in 2020 with the total catch by all countries declining to levels of the early 1980s. Thus it was possible the unexpected low abundance in our survey catches was a consequence of extremely low abundances in the Gulf of Alaska in the winters of 2019 and 2020. At the same time, there was an unprecedented marine heat wave in the Gulf of Alaska from 2014 to the end of 2019 (Bond et al. 2015; DiLorenzo and Mantua 2016; Cornwall 2019; Suryan et al. 2021). It is unclear if the unique appearance of the marine heat wave was responsible for the low abundances of salmon, but it should be clear that there needs to be a much better understanding of the factors affecting the ocean survival of Pacific salmon if we are to be responsible stewards of Pacific salmon in a future of rapidly changing ocean ecosystems as occurred from 2014 to 2019."
https://npafc.org/wp-content/uploads/technical-reports/Tech-Report-18-DOI/Technical-Report-18.pdf
Page 12 "Chum salmon O. keta are the most abundant salmon in the North Pacific Ocean. Some experts placed chum salmon second (e.g., Fukuwaka et al. 2007), thereby missing the point of existence of several year classes spending mainly three-four winters at sea. Chum salmon are also the most “domesticized” salmon, considering the proportion of hatchery-originated stocks in the total species abundance. Japan led in chum salmon hatchery propagation and fishery harvest before the mid of the last decade. Then, unexpected, a decline of the chum salmon run happened, which is more frequently explained by the climate warming and deterioration of environmental conditions for salmon at the southern edge of their areas. Total chum salmon harvest in the North Pacific gradually declined since 2015."
Page 14 "The 2019 Gulf of Alaska survey clearly showed sockeye salmon adherence to subarctic waters with two-layer structure. Nine of ten sockeye salmon catches larger than 1 fish/hour occurred northwards of latitude 52ºN, where the surface mixed layer was notably thinner - about 50 m, than the area southwards with the mixed layer depth below to 100 m (Pakhomov et al. 2019). This circumstance can impact the increase of sockeye trawl catches both due to a salmon distribution density increase throughout the area of better food conditions and due to the shrinking of the vertical distribution range. It is known that sockeye salmon are good divers, and this salmon prefers to feed near the thermocline, where vertically migrating zooplankton are accumulated due to a concentrating effect of any physical border. Such feeding behaviour can partially explain the notable prevalence of sockeye salmon night-time catches before daytime catches in our winter surveys. Biologically, sockeye salmon are more adapted to feed visually at dusk. They possess bigger eyes, adapted for visual detection of food at dusk and have many midwater zooplankton and nekton prey in their diet. In addition, sockeye salmon display a similar pattern of daily redistribution in the freshwater period in big lakes (Clark and Levy 1988)"
Page 15 "In February - March 2009, coho salmon catches were larger than in the trawl survey in the Gulf of Alaska in 2019. In the spring of 1991, coho salmon were captured at the southern limits of that survey area only. In early February 2019, during the R/V’s Professor Kaganovskiy passage to Vancouver, coho salmon were found in catches of only three out of nine trawl hauls in a small amount. It looks like quickly migrating coho salmon can be found here and there in offshore central and eastern North Pacific during winter.
At the same time, coho salmon can be found year-round throughout the North American shelf. Despite the proportion of coho abundance in the Strait of Georgia and the oceanward side of Vancouver Island changes between autumn and winter, notable numbers of coho salmon dwell in the near-coastal zone even in the most severe season. Along the western coast of Vancouver Island, coho salmon can be found in significant abundance in all winter months (Beacham et al. 2016). This is why coho salmon are always considered as the salmon species with a mostly near-coastal distribution in winter."
Page 15 "Chinook salmon are a relatively rare salmon in the trawl catches in upper pelagic layer. Juvenile and immature Chinook salmon keep themselves dispersed and were mostly captured by trawl in single numbers later in winter and spring. Few Chinook salmon were captured in the Gulf of Alaska in February - March 2019, in April - May of 1990, and in February of 2006 until R/V Pacific Legacy captured almost 30 fish at the southwestern coast of the Vancouver Island in April 2020. In the western North Pacific, juveniles of the first ocean winter predominate in catches in autumn and early winter. In October, they start migrating from the shelf water domain to deep-water areas with a general south-eastern direction. Near the eastern Kamchatka Peninsula coast, the peak of Chinook salmon outmigrant abundance is reached in November while most of them migrate further to the Subarctic Current area in December and January."
Page 16 "Chinook salmon were found throughout the whole outer shelf and upper continental slope from the Olyutorsky Bay to the Bristol Canyon (Radchenko and Glebov 1998). Chinook bycatch in the walleye pollock fishery is an important fisheries management problem. Despite all measures undertaken, including fishery restriction in areas of the most frequent salmon occurrence, Chinook salmon bycatch still exceeds 15,000 fish per year including salmon from every major North American stocks (NPMFC data, https://www.npfmc.org/bsaisalmon-bycatch/salmon-bycatch/)."
Page 16 "Based on diet studies, the Chinook salmon near-bottom distribution can be related with its major prey distribution. In summer, Chinook salmon of the first and second marine years mostly consume juvenile squid dwelling in the upper pelagic layer while older salmon hunt for larger squid, keeping themselves in the near-bottom realm. Chinook salmon measurements in the Bering Sea showed that pelagic trawl catches mostly consisted of younger salmon while the bottom trawl catches were for older and larger fish (Radchenko and Glebov 1998). This is also evident in an example of Chinook salmon behaviour with the Data Storage Tag. In its second winter, this fish started diving deeper and spending more time in at depth (Myers et al. 2016). In the Strait of Georgia, overwintering Chinook salmon mostly prey for forage fish and present an important prey for residential orca whale populations (Riddell et al. 2018)."
Page 17 "The ocean distribution of salmon is complex and variable, depending on spatio-temporal scale and synergies among heredity, environment, population dynamics, and phenotypic plasticity (Myers et al. 2016). Recently, new information was collected in scarcely studied regions and seasons, and summary reviews were published on the marine life phase of Pacific salmon (Shuntov and Temnykh 2008, 2011; Beamish 2018; Pakhomov et al. 2019; Somov et al. 2020). Nevertheless, considering salmon wintering in high seas, scientists remain puzzled and often say contradictory things (as shown by Shuntov et al. 2017). In this review, species-specific distinctions in salmon ecology are highlighted that determine differences in species distributions, migrations, feeding habits, and trophic relationships and should be taken into account during planning of further salmon studies in the winter ocean."
Page 17 "The cherry salmon migration pathway from the western Kamchatka coast lays across the southern Sea of Okhotsk where it can occur in relatively high abundance up to 15–21 fish per hour (Shuntov and Temnykh 2011). In winter, as well as during spring - summer migrations, cherry salmon feed intensively. The basis of their diet are small fish including Japanese anchovy, capelin, sand lance, juveniles of arabesque greenling, walleye pollock, as well as squid and amphipods (Shuntov and Temnykh 2011). Despite cherry salmon staying mostly in the marginal seas, it also can be preyed upon by pelagic predatory fish of the ocean realm. In August 1994, we had a rare observation of cherry salmon preyed upon by daggertooth. A daggertooth specimen with two cherry salmon in its stomach was caught in the upper layer in the Sea of Okhotsk off the Northern Kurile Islands (Radchenko and Semenchenko 1996)"
Page 42 "Keywords: lamprey, salmon shark, dogfish shark, marine mammal, daggertooth, wounds, scars Predation on Pacific salmon (Oncorhynchus spp.) at sea remains one of the big “unknowns” for salmon marine ecology (Pearcy 1992; Beamish 2018). The answers to basic questions such as who are the major predators (fish, birds, marine mammals), whether predation is random or selective (e.g., based on size or health), or how it varies in space and time are all poorly understood. Perhaps the biggest challenge to understanding predation is discovering predators “in the act” of preying on salmon by finding salmon remains in predator stomachs. "
Page 49 "Together, chum, pink, and sockeye showed lower Relative Infection Burden (RIB) in the GoA compared to coastal British Columbia (BC) (Fig. 1a). In contrast, RIB in coho was higher in the GoA than in coastal waters (Fig. 1a), although the number of infectious agents as well as their diversity was lower in the GoA for all species (Fig. 1a, b). This suggests that the higher RIB in coho in the GoA is due to the higher loads of VER, Loma sp., and I. hoferi. "
Page 56 "Results of this study demonstrated substantial annual differences in cephalopod abundance and composition in winter period in the Gulf of Alaska. For example, in 2019, B. borealis was the most frequently caught squid (~81%) which also dominated by weight (~60%). In 2020, A. felis was the most abundant (~47%), however O. borealijaponica dominated by weight (~58%). Squid distribution and abundance differences most likely were related to oceanographic conditions (0.33 ℃ cooler in 2019; developing spring bloom in 2020; change in water movements) and methodology (shallower vertical net opening in 2020). Cephalopod larvae occurrence in Juday nets (B. borealis, G. onyx, J. diaphana, T. borealis) confirm that these species use the winter in the GoA as a spawning and nursing ground. Applying molecular techniques showed better catch-eDNA reads agreement in 2020, when an addition of bleach deactivator (sodium thiosulfate) was introduced into the sampling protocol. Although, results of eDNA analysis faced a few challenges (e.g., the need for an improved references database - sequence data and taxonomy), it did provide very promising results and indications that it can greatly improve verification of species identification and monitoring in the future. Overall, squid winter studies proved to be a valuable addition to salmon studies."
Page 64 "Sustainable use of salmon resources is based on accurate forecasting of stock abundance. In Kamchatka, forecasting of the Pacific salmon stock abundance is largely based on the Ricker model, i.e.,stock-recruitment (Ricker 1954; Shepherd 1982), or standard sibling model that forecasts abundance of a given age-class for a given year based on the abundance of the previous age-class in the previous year (Peterman 1982). Autumn surveys in the southwestern Bering Sea and Sea of Okhotsk study juvenile salmon seaward migration. Trawl catches from these surveys are used as empirical data for pink and chum salmon in the model “juvenile fish at sea-offspring returns”.
Forecasts also utilize data on juvenile salmon abundance in rivers. These data were obtained by enumerating salmon juveniles in traps at selected streams in Eastern and Western Kamchatka. Ecological indicators can play an important role in making decisions about the catch forecast. Math modeling is always limited by a number of predictors and by the method’s errors. Fish abundance is the main and often the only predictor used in forecasting. In such a case for Pacific salmon, multi-factor characteristics of the environment is hardly considered that reduces the forecast informativity (Urawa et al. 2016)."
Page 72 "Introduction - A major objective of the two expeditions was to improve the understanding of the winter ecology of sockeye salmon in the open ocean with a specific focus on fish originating from British Columbia (BC). Methods and preliminary results are summarized in two reports by Pakhomov et al. (2019) and Somov et al. (2020). In this report, we repeat some of the catch results from these two reports, but focus on the highlights as they relate to new analyses and interpretations with a focus on sockeye salmon originating from the Fraser River in BC.
Most sockeye salmon spend 1 to 3 years in fresh water before migrating into coastal areas in the spring. They move offshore in the summer and winter where they remain for 1 to 3 years before returning to natal rivers to spawn in the late summer and fall (Burgner 1991; Farley et al. 2018). Average annual commercial catches of sockeye salmon by all countries in recent years from 2000 to 2020 averaged 16.5% of the total weight of all commercially caught Pacific salmon. Since 2000, the commercial catch of sockeye salmon by all countries has an increasing trend with an average from 2000 to 2020 of 148,904 MT. Over the period from 2000 to 2020, Alaska catches averaged 109,537 MT, Russian catches averaged 32,933 MT and Canada (BC) averaged 6,427 MT. However, in BC, the recent catches in 2019 and 2020 were 2.4% and 2.7% of the average catch from 2000 to 2020. Thus, as commercial catches were increasing in Alaska and Russia, they were declining in Canada with a major collapse in 2019 and 2020. There is no clear understanding of the cause of this collapse; however, the first two expeditions are providing new information on the ocean residency of sockeye salmon that will help direct future research.
General results
The Gulf of Alaska expeditions were conducted from February 19 to March 17, 2019 and March 12 to April 6, 2020. In 2019 fishing was conducted by the Russian research vessel Professor Kaganovskiy (Pakhomov et al. 2019) and in 2020 the fishing was conducted by the Canadian commercial trawler Pacific Legacy (Somov et al. 2020). There were 64 sets completed in 2019 and 52 sets completed in 2020, however sampling locations varied (Fig. 1) due to a combination of weather and a requirement to refuel mid-trip during the 2020 expedition. In both years most sets were conducted south of 52oN and west of 137.5oW (2019 - 50%; 2020 - 65%; southwest quadrant; Table 1). In 2019 the northwest region of the survey area (north of 52oN and west of 137.5oW; northwest quadrant) had 37.5% of the sets with no sets in the northeast (north of 52oN and east of 137.5oW; northeast quadrant; Table 1). In comparison, in 2020 there were only 2% of the sets in the northwest quadrant but 31% of the sets in the northeast quadrant. Additionally, in 2020, 25% of the sets were conducted in the southeast region of the study area (south of 52oN and east of 137.5oW; southeast quadrant) whereas in 2019 13% of the sets were in this region (Fig. 1; Table 1)."
Page 75 - "Ocean migration pattern of sockeye salmon from the Fraser River and other areas around the Gulf of Alaska: Juvenile sockeye salmon from the Fraser River and other rivers from Washington State to SE Alaska mostly spend two winters in the ocean (Burgner 1991; Farley et al. 2018). Ocean entry times vary, but in general, the coastal residence time is short, with juveniles moving offshore in summer and migrating quickly in a counterclockwise direction north. The counter-clockwise movement was proposed by French et al. (1976) and by Hartt and Dell (1986). The reasons for this interpretation were summarized by Burgner (1991) and Farley et al. (2018). More recent stock identification methods using DNA generally support this proposed interpretation of a rapid movement northward with residence in the Gulf of Alaska for two winters before returning to spawn in natal rivers (Tucker et al. 2009; Beacham et al. 2014); however, sampling for the work supporting this migration model was focussed within the coastal areas. "
Page 77 "There should be little doubt that it is time to understand where sockeye salmon are in the two years of their ocean residence and what might affect their distributions. As ocean warming events (Bond et al. 2015; Cavole et al. 2016; DiLorenzo and Mantua 2016) increase in frequency and other climate related changes in ocean conditions occur that affect sockeye salmon, this understanding becomes an essential contribution to forecasting returns and to understanding the future of sockeye salmon production in the Fraser River and other rivers."
"Conclusion We suggest that the surprising poor returns of sockeye salmon returns to British Columbia in 2019 and 2020 identify the necessity to expand research to provide a bigger picture and more complete understanding of the factors that regulate sockeye salmon abundance in the ocean. We suggest that the data collected in 2019 and 2020 should not be dismissed as being too small to be of use, but to be the first indication that more effort will provide new methods of forecasting and stewardship."
Page 119 "British Columbia and Washington State sockeye salmon indices of BC sockeye salmon systems are run reconstructions for major river systems to account for
differences in fishery removals over time; these indices include the Nass River, the Skeena River, Barkley Sound sockeye (southwest coast of Vancouver Island), and the Fraser River (Fig. 4). These reconstructions involve multiple sockeye populations within each watershed. To account for the cyclic dominance in Fraser River sockeye salmon and wide differences in the abundance between lines (Roos 1991), baseline averages and deviations were calculated within cycle lines. Each reconstruction is based on quantitative estimation of spawning abundance, total catches in fisheries along their adult migration routes, and were consistently conducted during this period.
For these BC indices, sockeye returns have been less than baseline averages in recent years: Nass River, -33 to -51% since 2016; Skeena River, -15% to -59% with the greatest reduction in 2019; in Barkley Sound, -46% to -72% with the largest deviation in 2019; and for Fraser River, a progressive decline from -42% in 2015 to -90% in 2020. In Washington State, sockeye indices were examined for Baker Lake (Skagit River, Puget Sound), Lake Washington (Puget Sound), and Columbia River. Trends in these sockeye populations each differ. Returns to Baker Lake have been rebuilding from hundreds of sockeye in the 1980s to ~20,000 during the past decade. Pattern of sockeye returns to Washington Lake (Fig. 5) in 2019 and 2020 were 89% and 86% less than their baseline average, and have similar trend to Fraser River sockeye. In the Columbia River, sockeye returns have been improving over the past decade but reductions were observed between 2017 and 2019, followed by an above average return in 2020 (Fig. 5). Other smaller sockeye populations in Washington State (Gustafson et al. 1997) were not assessed."
Page 128 "Coho salmon (Oncorhynchus kisutch) While the abundance of coho salmon was expected to be very limited compared to sockeye, pink, and chum salmon in the deep-water environments of the Gulf of Alaska, coho salmon were widely distributed and commonly caught in both 2019 and 2020 winter expeditions (Beamish et al. 2022 this volume). The distribution of coho salmon throughout the Gulf of Alaska was one of the most interesting observations."
Read the report in its entirety here: https://npafc.org/wp-content/uploads/technical-reports/Tech-Report-18-DOI/Technical-Report-18.pdf
