This section of the Cohen Commission report on troubles with British Columbia's Faser River sockeye salmon run explores evidence (page 77) that volcanic eruptions have boosted populations by fertilizing large plankton blooms. The full report, “The Uncertain Future of Fraser River Sockeye,” is available at http://www.cohencommission.ca This is posted for the Dot Earth blog in relation to controversial efforts by the Haida Gwaii first nation to fertilize runs artificially by distributing iron dust at sea. More soon on Dot Earth: http://j.mp/dotironfert Scientific American: http://j.mp/sciamironfert Vancouver Sun: http://j.mp/vanouvsunironfert

1. some sockeye populations (e.g., Chilko Lake) is
low. The bloom in 2007 was the latest on record
since 1998.668 Although there are likely no Fraser
River sockeye postsmolts in Queen Charlotte
Sound during April, the lag until their arrival
in June and July can allow time for their prey Growth in the North Pacific
base, which depends on phytoplankton, to
develop.669 The distribution and movement of immature
The summer of 2008 was the opposite. Sea Fraser River sockeye salmon at sea is the least
surface temperatures along the North American understood of all life history phases. Stock-specific
coast were cool following what was the coldest movements of Fraser River sockeye in the open
year in the Gulf of Alaska since 1972, and these ocean are unclear, but there is some evidence that
cool anomalies persisted along the coast through different sockeye stocks are in different places in
September. Migrating sockeye in 2008, once the offshore. Dr. Welch testified that sockeye from
leaving the coastal straits, would have had a the Nass, Skeena, Fraser, and Columbia rivers
very different thermal experience during their and Rivers Inlet are spatially separated during the
outmigration compared with 2007. This evidence month of June.678 Dr. Timothy Parsons, professor
is consistent with the expert report by Thomson emeritus at the University of British Columbia and
and others.670 honorary research scientist with DFO, testified
During the evidentiary hearings, that radio isotope testing has shown that different
Dr. McKinnell testified that there is a correlation stocks of salmon go to very specific locations in
between these unusual sea surface temperatures the Gulf of Alaska, and that populations of Atlantic
and wind patterns on the one hand, and Fraser salmon are distributed in different geographic
River sockeye survival on the other, though that locations in the North Atlantic.679
does not establish causation.671 Further work Dr. McKinnell testified that the period between
would be required to understand causation.672 when Fraser River sockeye are migrating northward
He said that, in attempting to explain the poor along the continental shelf and when they appear
2009 return, it is also necessary to explain other in deep water is one of the least-understood periods
observations from that time period, including for these animals, in part because logistically dif-
double-the-average returns of sockeye to the ficult winter sampling is involved.680
Columbia River, better-than-expected returns Several of the Commission’s technical reports
of sockeye to Barclay Sound, and record high examined, and witnesses testified about, the
returns to the Harrison River.673 One needs to stressors during the sockeye’s residence in the
develop a model that somehow satisfies all North Pacific Ocean that may have caused or
these concurrent observations – and placing the contributed to the recent decline. I summarize
mortality of the 2007 age-one smolts in Queen these discussions below.
Charlotte Strait / Sound has the possibility of
doing so.674 Predation
Dr. McKinnell also testified that there is
doubt whether a 2009–10 El Niño / La Niña event In Technical Report 8, Predation, authors
influenced the 2010 return, as postulated by Dr. Christensen and Dr. Trites stated that salmon
Exhibit 1303 (Thomson and others, 2011) and sharks migrate between Hawaii and Alaska.
testimony of Dr. Beamish.675 This is because the They are reported to feed primarily on Pacific
high return for 2010 was evident in the 2009 test salmon in spring and summer. However,
fisheries where the jacks (three-year-olds) were Mr. McFarlane testified that salmon sharks are
seen in unusually high abundance.676 He ex- opportunistic and episodic feeders, and a lack of
plained that “[w]e interpret this to be an indica- diet information makes it difficult to link salmon
tion that the high abundance of the return in shark predation and changes in sockeye popula-
2010 was established at least a year earlier than tion.681 Although abundance trends are very
the time when the 2010 return occurred.”677 limited, there is an indication that abundance
75

2. has increased in recent decades, an occurrence In the North Pacific marine environment,
that means the predation impact on Fraser River long-term climate change trends are difficult to
sockeye salmon may have increased as well. detect because conditions are strongly related
The researchers concluded that the only way to both inter-annual and inter-decadal modes of
to reliably evaluate if salmon sharks have had climate variability:
an increasing impact would be to gather more
information about their open-ocean abundance Inter-annual variability is related to El Niño
and abundance trends.682 During the eviden- Southern Oscillation events, which occur
tiary hearings, Dr. Christensen testified that the every two to seven years and persist for up
salmon shark was at the top of their list of Fraser to 1.5 years. Typically, El Niño events lead to
River sockeye predators.683 warm sea surface temperature in the waters
Blue sharks are much more abundant than of the west coast of North America, and since
salmon sharks, but since their population has the 1970s, El Niño events have become more
not increased in recent decades, it is unlikely frequent. By contrast, La Niña events result in
that abundance trends can explain the Fraser cooler waters, and such events have become
River sockeye decline, even if it may have con- less frequent.688
tributed to it. Dr. Christensen and Dr. Trites also Inter-decadal variability in the climate
discounted the role of the Pacific sleeper shark in of the North Pacific Ocean has been
the decline.684 described by indices such as the Pacific
Sablefish are opportunistic feeders known to Decadal Oscillation (PDO), which typically
consume sockeye salmon, but sablefish in British persists for 20 to 30 years. Warm sea surface
Columbia and in the Gulf of Alaska have been in temperatures over the eastern North Pacific
decline since the late 1980s. For that reason, the Ocean characterize the warm or positive
researchers concluded that it is not likely that phase of the PDO, whereas opposite sea
sablefish would be a major factor in the decline of surface temperature patterns characterize
Fraser River sockeye salmon.685 the cool or negative phase. The PDO was
The researchers also considered daggertooth, predominantly in the positive phase between
walleye pollock, and arrowtooth flounder, all of 1977 and 1997 and, since 1998, has exhibited
which could exert predation pressure on sockeye more frequent alternations, lasting three to
salmon. However, owing to inadequate information four years. Sea surface temperature in the
about abundance and abundance trends, they Gulf of Alaska has increased by about 0.25°C
could not conclude that any of them has been a fac- per decade since the 1950s. It is now 1.5°C
tor in the decline of Fraser River sockeye salmon.686 warmer than 60 years ago and 0.5°C warmer
than 20 years ago. However, such observed
Climate change warming has been attributed mostly to the
positive phase of the PDO. In contrast to
In Technical Report 9, Climate Change, authors warming trends, both salinity and pH of the
Dr. Hinch and Dr. Martins stated that the abundance North Pacific Ocean has been decreasing in
of sockeye salmon has closely tracked decadal-scale recent decades.689
fluctuations in sea surface temperature over most of
the past 300 years. These fluctuations have been well Only a few studies have explored the rela-
documented during the past century and linked to tionship between temperature and survival of
major climate-driven changes in the marine environ- immature sockeye salmon in the open ocean. A
ment occurring every 20–30 years. The abundance of 2009 study found that survival of Alaskan sockeye
Fraser River sockeye began to increase dramatically was positively correlated to sea surface tempera-
at the end of the 1970s, reaching historic high abun- ture during all the years of ocean residence, while
dance in the early 1990s. Subsequently, abundance a 1991 study found that the survival of Fraser
and productivity began to decline to recent low River sockeye was negatively correlated to sea
levels, in coincidence with the exacerbation of the surface temperature in their last few months
long-term warming trend of the global climate.687 in the open ocean.690 The authors of Technical
76

3. Report 10, Production Dynamics, reported that fish in 1958. He explained why this affected only the
increased sea surface temperature in the location younger 2008 fish, not the 2009 fish:
of early ocean residence for smolts of a given
stock is associated with increased productivity Our take on that is that what you have taking
in Alaska.691 The researchers concluded that place is a massive bloom of diatoms which are
increased temperatures in the Gulf of Alaska absorbed very quickly by the zooplankton. The
over the past two decades have possibly resulted zooplankton will be rather small zooplankton
in lower survival of Fraser River sockeye during and they will be consumed much more eas-
open-ocean residence.692 ily by the young adolescent salmon than the
During the evidentiary hearings, Dr. McKinnell larger 2009 salmon which are still waiting for
testified that Intergovernmental Panel on Climate something big to come along. They’ve already
Change projections for future climate are difficult to gone through the stage where they were eating
represent in terms of the finer-scale climate, such as small prey. They’re a year older.697
climate changes that will occur in British Columbia
and what the response of the marine ecosystem will Dr. Parsons acknowledged that these are really
be in the northeastern Pacific Ocean.693 quite hypothetical answers to a question which is
somewhat speculative, but worth recording as a
Marine ecology possible mechanism for the 29 million Fraser River
sockeye that returned in 2010.698 Dr. Irvine raised
Technical Report 4: Marine Ecology several questions about Dr. Parsons’s hypothesis.
It was not clear to him why the older salmon
The authors of Technical Report 4, Marine Ecology, (2009 return year) would not have consumed this
considered the hypothesis that a volcanic eruption additional prey, as nutrient sources were ex-
in the Aleutian Islands in August 2008 led to the tremely important to them because they would be
widespread deposit of volcanic ash (including accumulating a lot of mass before beginning their
iron) in the northwestern Pacific. The theory is that return migration.699 Also, it was to be expected that
this deposit enhanced productivity of chlorophyll the 2010 return would be strong, since the 2008
in mid- to late August, benefiting the 2008 smolt smolts were migrating northward during a very
year and explaining the strong 2010 Fraser River strong La Niña, with cold waters.700 Dr. Parsons
sockeye return. The authors doubted that this responded that the impact of the volcanic ash may
enhanced productivity of chlorophyll benefited have been different for different sockeye stocks,
the 2008 smolts because the fish would have been since radio isotope testing has shown that different
migrating along the continental shelf, where iron is stocks of salmon go to very specific locations in
not normally the limiting nutrient – so its addition the Gulf of Alaska.701 Although satellite imagery
would not have produced much benefit to this seems to indicate that the chlorophyll (represent-
cohort. To have directly benefited this cohort, the ing presence of phytoplankton) was distributed
enhanced biological production would need to throughout the Gulf of Alaska, that does not mean
be entrained in the pelagic food web and stored that it was necessarily evenly distributed.702
through, or made available to it over, the winter. McKinnell and others concluded that there
However, zooplankton biomass in 2009 was near was a strong, abrupt, and generally persistent shift
the average.694 to warm sea surface temperature anomalies along
Dr. Parsons testified about the possible impact the continental shelf in late June of 2007. However,
of the 2008 volcanic eruption on Kasatochi Island, since most southern stocks were likely using this
Alaska, on Fraser River sockeye.695 There is no doubt migratory route during 2007, and non-Fraser
that an enormous diatom bloom was generated in stocks did not suffer incremental mortality, it
the Gulf of Alaska and that zooplankton increased seems unlikely that this coast-wide phenomenon
by a factor of three, which he said benefited sockeye was the cause of incremental mortality of the
in the Gulf of Alaska.696 A similar event occurred Fraser River stocks.703
in 1956, when a volcano erupted in Kamchatka in The report noted that lack of observations of
eastern Russia, resulting in a return of 20 million salmon at sea, at relevant time and space scales,
77

4. severely limits the ability to draw firm conclusions sockeye.709 None of Dr. Peterman’s evidence was
about their fate.704 It concluded that, for most contested by DFO.
Fraser River sockeye populations (where smolt I also heard in hearings on the marine envi-
abundance is not observed), the relative contribu- ronment that the interactions between hatchery
tions of freshwater and marine effects on survival and wild salmon is a substantial issue in fishery
can only be assumed. Fry-to-adult survival trends science, and that there is extensive literature on the
tend to favour a marine origin for the decline, potential interactions for pink, chum, chinook, and
because they share a common ocean but not a coho.710 Dr. Beamish stated that there is evidence of
common lake. These populations also share a hatchery–wild interactions among various salmon
common means of egress to the sea, although this species, although whether there could be a long-
life history stage is rarely examined in detail.705 term substantial reduction in production is less
According to the researchers, the sea provides clear among the scientific community.711
only limited amounts of food for growing sockeye Dr. Peterman explained that in the North
salmon. Fraser River sockeye were smaller when Pacific Ocean there is considerable potential for
the total abundance of sockeye in the Gulf of indirect interactions between wild and enhanced
Alaska was greater, but this is not a universal salmon.712 Across pink, chum, and sockeye
truth.706 During the habitat enhancement and salmon, 22 percent of adults in the North Pacific
restoration hearings, Dr. Peterman testified that Ocean are of hatchery origin, and plans exist to
the body size of a given age of adult sockeye further increase annual hatchery releases, particu-
salmon decreases as abundance of competitors larly in Russia and Alaska.713 Although relatively
increases.707 few sockeye are produced by hatcheries, wild
McKinnell and others said that there is some sockeye appear to interact with pink salmon.714
evidence that the mean fork length* of Fraser River Dr. Beamish testified that there is the potential,
sockeye was significantly smaller in brood years particularly for chum salmon that are enhanced
that matured in odd years. Because the odd / even in Asia and pink salmon enhanced in Alaska, to
cycle of abundance of pink salmon in the Fraser is have a density-dependent effect on Fraser River
potentially a source of competition for Fraser River sockeye in the Gulf of Alaska as a result of the large
sockeye returning the same year, it is reasonable to numbers of enhanced fish released into the
postulate that a reduction in mean size in odd years same area.715
is a consequence of competition for food with pink Dr. Peterman described several mechanisms for
salmon during the period of overlap in the Gulf interactions between wild and enhanced salmon.
of Alaska.708 Competition for food can occur between wild and
enhanced salmon because their diets overlap and
Interaction between wild and they are thought to generally pass through feeding
enhanced salmon areas at similar times and places.716 Food supply
in the open North Pacific Ocean has diminished
Dr. Randall Peterman, professor at the School of as a result of feeding largely by pink salmon.717
Resource and Environmental Management, Simon Also, predation-induced mortality on wild juvenile
Fraser University, and Canada Research Chair salmon can be increased because of the attraction
in Fisheries Risk Assessment and Management, of predators to high abundances of juvenile salmon
testified about potential interactions between wild driven by large hatchery releases.718 Dr. Peterman
†
and enhanced fish. In his view, increasing fish noted, however, that high total abundance of
densities in the North Pacific may have negative hatchery plus wild juveniles could also increase the
impacts on wild stocks, including Fraser River survival rate of wild juvenile salmon co-migrating
* Fork length is measured from the tip of the snout to the end of the middle caudal fin rays.
† Dr. Peterman was qualified as an expert in density-dependent effects on wild and enhanced fish populations for the purpose of this
hearing topic (May 2, 2011, pp. 10–11). He was also previously qualified as an expert in fisheries biology with expertise in fish population
dynamics and ecology and risk assessment during hearings on Exhibit 748 (Technical Report 10), which he co-authored for the
Commission (May 2, 2011, pp. 4–5). His curriculum vitae is Exhibit 749.
78

5. with hatchery fish because the abundance of prey So there’s a huge energetic demand on the
may satiate predators.719 maturing fish that does not exist for the im-
Where adults of wild and enhanced salmon mature fish, because they have to be able to
co-migrate through fishing areas, pressure is have enough resources to get from the Gulf
intense on managers to allow high harvest rates.720 of Alaska to fresh water, to swim up the river,
However, wild stocks generally have lower produc- to mate and produce gametes and everything
tivity (adults per spawner) than enhanced fish, so that goes along with maturation. That’s an
high percentage harvest rates targeted on enhanced energy intensive process.728
fish are known to eventually lead to overharvesting
and depletion of abundance of wild co-migrating Fraser River sockeye are captured in fisher-
stocks that are subject to those same harvest ies between Alaska and Washington State; their
rates.721 Finally, after adults leave the ocean, large availability depends on their migration route.
numbers of hatchery fish straying into spawning There are two migratory return routes – down
areas for wild fish can decrease biological diversity the west coast of Vancouver Island and through
and fitness of the wild stocks.722 Juan de Fuca Strait, or through Johnstone Strait
Dr. Peterman provided evidence that the body and the Strait of Georgia (the northern diversion
size at a given age of adult sockeye salmon de- route).729 (See also the discussion of pre-season
creases as abundance of competitors increases.723 forecasting in Volume 1, Chapter 5, Sockeye
He also explained how the survival rate of sockeye fishery management.)
salmon can decrease as the abundance of pink The percentage that follows the northern
salmon competitors increases, although he said diversion route varies from year to year. Dr. Welch
that there are only a few documented examples testified that, when the ocean temperature is at 10°C,
of this reduction in survival rate compared with the migration is almost entirely through Juan de
examples of reduced growth rate (reflected by Fuca Strait, but when the temperature increases to
adult body size).724 Dr. Peterman noted that the 12–13°C, 80 to 90 percent of returning sockeye come
concern about competition among wild and en- through the northern diversion route.730
hanced salmon for limited resources may become Several of the Commission’s technical reports
considerably more acute if the North Pacific Ocean examined, and witnesses testified on, the stressors
becomes less productive again (as it was before that may have caused or contributed to the recent
the mid-1970s).725 (For a discussion on DFO’s decline during the return of adult sockeye to the
management response to interaction between wild Fraser River. I summarize these discussions below.
and enhanced salmon, see Volume 1, Chapter 6,
Habitat management.)726 Predation
In Technical Report 8, Predation, authors
Return to the Fraser River Dr. Christensen and Dr. Trites reported that,
after harbour seals received protection under
In their fourth (or in some cases, fifth) year of theFisheries Act in 1970, their numbers increased
life, and after spending one-and-a-half years (or from approximately 9,000 to 108,000, with about
in some cases, two-and-a-half years) in the Gulf 40,000 of these in the Strait of Georgia. An
of Alaska, Fraser River sockeye leave the Gulf of analysis of 3,000 fecal samples collected from
Alaska and return to the Fraser River to spawn. 58 Strait of Georgia sites during the 1980s in-
During the evidentiary hearings, Dr. McKinnell dicated that harbour seals primarily ate Pacific
testified that the decision when to return appears hake (42 percent) and herring (32 percent), while
to be made in the winter preceding the year that salmonids comprised only 4 percent of the overall
the fish will mature; the decision has genetic and diet. Harbour seals appear to prefer chum and
growth components.727 From that time on, the coho salmon over sockeye or pink salmon.731 The
maturing sockeye needs to find enough food to researchers concluded that the numbers of seals
double its body weight in that last spring at sea. have been relatively stable in British Columbia
He stated: for the past decade and showed no changes that
79