A visit to Norway
http://farmedanddangerous.org/uploads/File/blog_files/A_visit_to_Norway.pdf
Neil Frazer, PhD
Professor of Geophysics, University of Hawaii at Manoa, Honolulu, HI 96822
neil@hawaii.edu
May 27, 2010
On May 20, 2010, the Norwegian corporation Cermaq held its Annual General Meeting
in Oslo Norway. Cermaq does business in Canada through its subsidiary Mainstream,
which owns 3 hatcheries and 28 sea cage sites on the coast of British Columbia, and
through its subsidiary EWOS, which manufactures feed for farmed fish at its plant in
Surrey. According to its 2010 annual report, Cermaq has 350 employees in Canada.
Perhaps one third of those employees work for EWOS. The government of Norway owns
40% of Cermaq, and the Norwegian people are very conscious of the environment. As the
trade commissioner at the Canadian embassy in Oslo phrased it, “Norwegians strive to be
greener than any other country in the world.” A small group from North America
attended the Annual General Meeting to request that Cermaq/Mainstream remove its seacage
sites from the migration routes of wild Pacific salmon in British Columbia.
The group consisted of Bart Naylor and Don Staniford, of the Pure Salmon Campaign,
Darren Blaney of the Homalco First Nation, Nicole Mackay of Mackay Whale Watching,
and me. Prior to the meeting, Bart, Don, Darren, Nicole and I met privately with the CEO
and Directors of Cermaq to discuss events in BC. (Full disclosure: my travel expenses
from Victoria were partly paid by the Wilderness Tourism Association of BC and the
Pure Salmon Campaign.)
The private meeting with the CEO and Board was constructive. The Board appeared to
know it had trouble—Cermaq lost a lot of money in Chile in 2008—and it was open to
understanding events in BC. The meeting began with a promotional video by Cermaq that
used the word sustainable at least ten times, and mentioned ‘coordinated area
management’ (simultaneous fallowing and/or simultaneous treatment for lice) as though
it was a new development instead of something that independent scientists have been
urging salmon farmers to do for many years. The corporate director for public relations
also mentioned it with some pride, as though all of the problems of the past were now
solved.
Don Staniford showed the Board Damien Gillis’ video of the “Get-Out Migration” that
had culminated in Victoria on May 8. Darren Blaney spoke movingly about the effects of
the Church House salmon farm on the local clams (gone within a few years), and Nicole
Mackay spoke eloquently about the effects of salmon scarcity on orcas in the Broughton
Archipelago (with reduced numbers of salmon, orcas must now spend so much time
foraging that there is little time left to socialize and mate). I pointed out that the Victoria
rally shown in the video had been attended by four to five thousand people, a Victoria
policeman’s estimate. I also pointed out that The Pure Salmon Campaign’s goal of
removing two farms from Okisollo Channel was regrettably modest. In my opinion
2
nearly all of them should be removed, not only for the safety of the Fraser sockeye, but
also to let that area recover from the obvious eutrophication.
Shortly before the Annual General Meeting, we took part in a demonstration in front of
the building where the meeting was to take place. Several other environmental
organizations from Norway were present, along with a few Norwegian businessmen
concerned for wild salmon. I held a sign written in Norwegian that said “Closed
containment now,” and stood with the others next to a large Norwegian flag with a few
sea lice painted on it. Those lice might as reasonably have been painted on a Canadian
flag, but we did not have a Canadian flag with us.
At the Annual General Meeting, Bart Naylor’s shareholder resolution that the Board of
Directors be elected every year, rather than every two years, was approved by a majority
of independent shareholders, but it failed to pass because the manager of the Norwegian
government’s 40% share of Cermaq voted against it. Bart, Darren, Nicole and I each
spoke for about three minutes, and Don showed a minute or two of the video from
Victoria. An employee of Mainstream Canada smirked and snorted during our
presentations, in glaring contrast to the ever-courteous Norwegians. I suppose she was
present in Oslo because she had escorted to Norway an Ahousat chief that had recently
allowed salmon farms into his nation. That Ahousat man didn’t speak at the meeting
other than to verbally attack Nicole afterward. I missed his harangue because an elderly
shareholder, one of the founders of Norskdata, wanted to tell me about his own scientific
days.
Prospering Cermaq
There is no need to feel sorry for Cermaq/Mainstream shareholders, or for those of
Marine Harvest and Grieg. Removal of salmon farms from the migration routes of wild
Pacific salmon will reduce North American farmed salmon production, but will only
temporarily reduce the profits of those three large Norwegian companies that control over
91% of farmed salmon production in British Columbia. To see this, consider events
following the recent disease crisis in Chile. When Chilean production dropped, farmed
salmon suddenly became a luxury instead of a commodity in the markets served by Chile.
Prices jumped, and salmon farmers not directly affected by the epidemic enjoyed greatly
increased profits.
Cermaq aside, it is quite possible that the economy of BC would also be improved by
making farmed salmon a luxury rather than a commodity. Recall that Cermaq employs
only 350 people in BC, less than 12% of its global workforce. If farmed salmon were a
luxury rather than a commodity, the industry would more closely resemble the wine
industry, which contributes many jobs to the BC economy; and the 21,000 direct jobs in
BC’s wilderness tourist industry would be safe. For reference, the BC Salmon Farmers
Association claims 2,800 direct jobs in salmon farming.
Intelligent companies always seek to increase profit rather than revenue, provided that a
reduction in revenue does not cause the size of the company to shrink relative to that of
3
its competitors. If the removal of farms from migration routes in British Columbia does
not penalize one company more than another—in other words, if the percentage decrease
in production of each company is the same—the companies will accept the necessary
changes after only perfunctory protest. Deciding which farms to remove is not difficult
science, and the farms can be shut down within six months with relatively little financial
hardship to the companies concerned. Remember we are talking about large multinational
companies. Cermaq’s 2009 cash flow from operations was C$180 million. Only 33% of
its revenue comes from Mainstream, its fish-farming subsidiary, and only 25% of
Mainstream’s revenue comes from its sea-cage sites and hatcheries in BC.
Why the best scientists are angry
Peer-reviewed studies show that farm-fostered sea lice reduce the productivity of wild
salmon populations. Each autumn, larvae from lice on in-migrating wild adult salmon
infect farm salmon. Six months later, larvae from lice on the farm salmon infect outmigrating
juvenile wild salmon. Pink salmon fry are especially vulnerable because they
enter the ocean weighing about a quarter of a gram and they do not have scales. In areas
without salmon farms, infection rates of pink salmon fry are negligible compared to rates
in areas with farms. Sea lice are classified as parasites because they increase the mortality
rates of their hosts in well-understood ways.
Despite the simplicity of the mechanism, and the many peer-review studies that support
it, there is confusion in Canada because of a small group of scientists in Canada’s
Department of Fisheries and Oceans (DFO). Those scientists have an explicit duty to
support government policy, and the current Minister of Fisheries and Oceans favors
salmon farming. This is not a recipe for good science. Judging from their papers, this
group lacks basic training in the population dynamics of host-parasite systems, and their
intention is to mislead their readers. In plain English, their papers appear to have been
written mainly for propaganda purposes. In Norway, by contrast, government scientists
now accept that sea lice from salmon farms are responsible for declines of wild salmon
and sea trout.
Dr. Larry Dill, a world-renowned Canadian ecologist, has described DFO’s Minister and
senior bureaucrats thus: “They are either extremely ignorant, misinformed, or they are
lying to us.” Dr. Daniel Pauly, easily the world’s most eminent fisheries scientist, agrees
with Dr. Dill. As Dr. Dill noted in his testimony to the Parliament of Canada’s Standing
Committee on Fisheries and Oceans, the science supporting declines of wild salmon due
to salmon farms is about as certain as science can ever be, and it is wrong for decision
makers to delay or deny by seizing on the fact that nothing in science is ever 100%
certain. Scientists who have studied the situation, and are not compromised by being on
the payroll of the salmon farming industry or a captive government agency, agree that
farms must be removed from the migration routes of juvenile Pacific salmon in British
Columbia if those salmon are to be preserved.
I share the feelings and concerns of Dr. Dill and Dr. Pauly. Moreover, the same lies that
were used to deceive the people of BC are now being used to deceive countries that
4
haven’t yet given their coasts over to sea cages. The propaganda generated by that small
group of scientists at DFO has not only created confusion in Canada; it has created
confusion around the world.
One of the ways in which the bureaucrats at DFO have controlled this issue is by tasking
unqualified people to deal with it. The sea lice group at DFO would not be bad scientists
if they stuck to what they know, but they haven’t had the courage to admit that they don’t
have the necessary mathematical training. To fish pathologists and old-school fisheries
scientists, the sea lice issue is a mystery, but to anyone with training in mathematical
ecology, it is easy. Remember there are only two important questions: (1) Does adding
farm fish to the ocean cause sea lice levels to rise? (2) Do increased levels of sea lice
cause wild fish to decline? Both questions can be answered with a Yes, using basic
ecology or basic physics, without ever putting a farm fish in the water. Anything other
than a decline of wild fish would be a miracle.
Unfortunately basic ecology doesn’t tell us how much the wild fish will decline. For that,
one can do a type of mathematical calculation that fewer than a hundred biologists in the
world would understand—or one can perform experiments by actually farming fish in
sea-cages. Since most fisheries biologists prefer experiments to mathematics, it is no
surprise that the experiment has now been done. Every sea-cage in Norway, Scotland,
Western Ireland, British Columbia is part of that huge experiment, and we now have the
answer to the question: How much do wild fish decline?
Not surprisingly, the answer is this: If you put just a few farm fish in the water (a few
mom-and-pop farms) wild fish decline hardly at all, but if you put a lot of farm fish in the
water, wild fish decline a lot. For technical reasons that are difficult to explain here, wild
fish decline at an accelerating rate as more farm fish are added. In other words the loss of
wild fish caused by the second farm in an area is more than the loss of wild fish caused
by the first farm in that area. Similarly the loss of wild fish caused by the third farm is
greater than the loss of wild fish caused by the second farm, and so forth, until the wild
fish are completely gone.
The decline of local wild fish caused by the farms doesn’t happen overnight. The way it
happens is that every year, on average, there are fewer wild fish. Unfortunately, wild
populations are difficult to census, and they are so variable from year to year that a
decline can be hard to discern until decades of data have accumulated. The parties
responsible for the decline inevitably use the delay in the decline to point the finger
elsewhere, as tobacco scientists used to do with lung cancer.
DFO’s distraction science
Salmon sea cages in coastal areas provide a dramatic example of the effect discussed
above because transfers of parasites between farm and wild are synchronized in time. As
noted earlier, when adult wild salmon migrate past salmon farms toward their rivers each
fall, they give whatever diseases they have to the farm fish. The farm fish unintentionally
culture those diseases over the winter. When the juvenile wild salmon swim past the
5
farms in the spring they get the diseases of their parents back from the farm fish—several
months earlier than would happen under natural conditions. The infection pressure on the
wild juvenile salmon is very great because the farm salmon are large in number and size.
Sea lice are a well-studied example of this because they are easy to see on a fish.
That disease transfer mechanism isn’t difficult to understand, but the sea lice group at
DFO tries hard not to understand it, and the natural human propensity for wishful
thinking has made their work popular with the salmon farming industry and with
governments that want to have both farmed and wild fish. The DFO group has tried to
discredit the disease transfer mechanism in three main ways: (a) by promoting unlikely
alternative sources of the sea lice on juvenile wild salmon: (b) by attempting to show that
lice don’t lower the life expectancy of their juvenile hosts: and (c) by distracting the
public from well-established population-level effects. I’ll consider the alternative source
hypothesis first.
By trawling a large area of ocean, one DFO scientist managed to find 35 wild salmon in
April–May, then hinted that this very small population could be the source of the lice on
the juvenile wild salmon, ignoring the many millions of farm salmon located just a few
miles from the wild salmon rivers. In another paper, he suggested that returning adult
wild salmon are the source of the lice on the juvenile wild salmon, a proposition that
could only make sense if the adult wild salmon returned several months earlier than they
actually return.
A second DFO scientist has worked very hard to create an impression that three-spine
stickleback are the source of the lice on juvenile wild salmon. He’s very careful not to
say exactly that—it’s enough that his work create confusion about the issue. None of the
21,000 lice he found on 1,300 sticklebacks were egg-bearing females, and almost all of
the lice were juvenile lice stages. Since he sampled the stickleback for lice in spring, and
lice only survive a few months, the lice on the stickleback couldn’t possibly have
originated with adult wild salmon the previous autumn. The lice almost certainly
originated with the millions of farm salmon near where he sampled the sticklebacks.
Later he attempted to have salmon lice reproduce on stickleback in his laboratory,
without success.
A responsible scientist would have done a spatial analysis of the lice on sticklebacks to
see whether lice burdens on the sticklebacks were greater near farms than distant from
farms, as on the juvenile wild salmon sampled by Marty Krkošek and Alexandra Morton.
This DFO scientist has not analyzed the spatial pattern of lice infection on stickleback,
nor has he done so with his samples of juvenile wild salmon. In his 2006 paper on
stickleback there were no salmon farms on his map of the Broughton Archipelago. The
paper was published in a journal whose editor is unlikely to know that the Broughton
Archipelago has over 20 salmon farms.
Now let’s look at the notion that lice do not lower the life expectancy of juvenile wild
salmon. It is well established in the scientific literature that lice feed on the exterior
mucus layer and skin of fish, consuming metabolic resources and providing a point of
6
entry for secondary infections, and that they cause fish to act in ways that make them
more visible to predators and thus more likely to be eaten. All three mechanisms are
guaranteed to lower life expectancy. As Dr. Poulin pointed out at the sea lice meeting,
“That’s why we call them parasites.”
The DFO stickleback scientist has also confused the life expectancy issue by doing
laboratory experiments in which juvenile pink salmon were exposed to lice larvae. What
happened is that most of the lice fell off his fish. The lack of mortality of the pink salmon
might have been related to the duration of exposure being very short compared to the
exposure duration of juvenile pink salmon out-migrating through the Broughton
Archipelago, and it might have had something to do with the fact that the fish weren’t
infected under natural conditions. When independent scientists did a similar experiment
using naturally infected fish, keeping the fish in the same waters in which they were
caught, the naturally infected fish died at a much higher rate than the uninfected fish even
when they were protected from predators.
Now let’s look at the population level question. In plain English, the suggestion is that
maybe the little fish that died from sea lice would have died anyway, so the number of
returning adult wild salmon isn’t reduced by their deaths. It’s not an absurd proposition,
because if there is a shortage of food, having too many young salmon competing for it
could cause many of them to starve. However, a strong population-level effect from lice
had already been established in the scientific literature, both generally and for pink
salmon stocks in the Broughton Archipelago. Unable to confront the issue head on, and
without the mathematical training necessary to even try, the senior DFO sea-lice scientist
wrote a paper entitled “Exceptional marine survival of pink salmon that entered the
marine environment in 2003 suggests that farmed Atlantic salmon and Pacific salmon can
coexist successfully in a marine ecosystem on the Pacific coast of Canada.” I’m
convinced that title was changed after review—I can’t imagine any serious reviewer
passing on it—but as a propaganda instrument it is perfect. The paper is fundamentally
unscientific in a number of ways.
First of all, survival wasn’t exceptional. (Here I use the word survival to mean intergenerational
survival, the ratio of returning adults in the daughter generation to the
escapement of the parent generation. Escapement is the number of fish in the parent
generation that entered the stream to spawn.) All population models for pink salmon have
the feature that survival increases with decreasing parent population, and the parent
generation (the pinks that returned in 2002) of the pinks that went to sea in 2003 and
returned in 2004 was the smallest it had been since 1960, so one would expect high
survival. For example, the pinks that went to sea in spring of 1961 and returned in 1962
had even higher survival than the pinks that went to sea in spring of 2003.
The second unscientific thing about the paper is that marine survival is unknown. To
estimate marine survival, you must divide survival by several other numbers, including
eggs-per-spawner and egg-to-fry survival. Values for egg-to-fry survival are as low as
5% for streams without spawning channels and as high as 57% for streams with spawning
channels. If you want a high estimate of marine survival, you pick an egg-to-fry survival
7
of a few percent, and if you want a low estimate of marine survival, you pick an egg-tofry
survival appropriate for a stream with a spawning channel. At least 40% of the pink
salmon in the Broughton Archipelago come from streams with spawning channels, so it
would have been appropriate to use a high egg-to-fry survival or at least some kind of
average egg-to-fry survival. Instead, the author picked 5.6%, which is only 1.3% more
than the lowest egg-to-fry survival in the source he cited. The effect was to inflate his
estimate of marine survival by a factor of at least 4.
The third unscientific thing about the paper is that the title-as-conclusion fails to warn the
reader that conditions in the Broughton in the spring of 2003 were nearly ideal. The BC
government had mandated a fallow of farms along the migration route. Other farms held
only smolts. In the prior six months many farms in the Broughton had been emptied of
farm salmon because of an epidemic of IHN virus. Sea lice infection pressure in the
Broughton was thus about as low as it could be without totally shutting down salmon
farming. Is this mentioned in the paper? The fallow is mentioned, even in the abstract, but
not in a way that conveys its true extent.
You can see why this paper and his other papers on sea lice make entertaining reading.
The language is impeccably scholarly, there are no lies, and every base is covered. But
unless you are a scientist reading the paper carefully you would be tempted to conclude
from it that salmon farming isn’t having much of an effect on pink salmon. If you read
only the title, you might think that everything is very fine indeed. It’s an exquisite piece
of misdirection.
Effects on BC universities
The propaganda generated by that small group of scientists at DFO has also had a
pernicious effect on university science. If you look carefully at BC’s three largest
universities you will find at each of them at least one major scientist with a minimal
understanding of ecology and a large grant for research in aquaculture. Aquaculture is a
full-employment program for disease specialists with PhDs, and those who do not
understand ecology have difficulty understanding the difference between salmon farming
and other types of aquaculture. That is one reason why BC is about to lose Martin
Krkošek, one of its finest scientists, to New Zealand. Marty, as he is called, is the young
ecologist who greatly expanded the sea-lice sampling program begun by Alexandra
Morton in the Broughton Archipelago. By all measures of scientific worth (scholarly
publication, teaching, presentations at meetings and demonstrated ability to secure
funding) Marty is head and shoulders above anyone else at his career stage. In both sea
lice science, and ecology generally, he towers above the sea lice scientists of every
country in the world—yet neither UVic nor UBC nor SFU hired him when they had
faculty positions for which he was more than qualified.
A few weeks ago I attended Sea Lice 2010, a biennial scientific meeting about sea lice,
which this year was held in Victoria. A UVic biology professor, the director of a research
centre there, gave a presentation in which he cheerfully confessed to not understanding
ecology. During the discussion period he wondered out loud whether sea lice really cause
8
mortality rates of fish to increase. Robert Poulin, the eminent parasite ecologist and
evolutionist, responded dryly, “That is why sea lice are classified as parasites,” and the
discussion moved on. As you might have guessed, this UVic professor has a large grant
for research in aquaculture that is irrelevant to the important questions mentioned above.
The real purpose of his grant, although he may not realize it, is to distract attention from
the important issues. He is doing ‘distraction science.’
Are neurotoxins the answer?
In BC’s Broughton Archipelago the salmon farmers have temporarily arrested the decline
of pink salmon by using the neurotoxin emamectin benzoate to reduce sea lice levels on
their fish, but this is not a desirable long-term solution because neurotoxins can have
unintended effects in the environment, and one of the hard lessons of the 20th century is
that very small concentrations of xenobiotic chemicals can have important effects that
aren’t immediately obvious. Also, sea lice could develop resistance to emamectin
benzoate, as has happened in Norway. That is thought to be less likely in BC than in
Norway, because every autumn the gene pool of lice on BC farm salmon is refreshed by
lice from in-migrating wild salmon, but there is recent anecdotal evidence for resistance
at some BC farms. Finally, there are other diseases lurking in wait. Nature has a large
supply of them, and many of them make sea lice look benign by comparison. In a farmfree
environment those diseases are controlled by predators like seals and porpoises that
eat sick fish before they have a chance to spread their disease, but a farm system has no
such preemptive controls—it is like a bomb with a delicate fuse.
The Fraser sockeye
The surprising non-appearance of the Fraser sockeye in 2009 may be an example of the
bomb effect. Scientists who are not on the payroll of the salmon farming industry or its
captive government agencies agree that sea lice from the farms in the Discovery Islands
area are a possible cause, but it is impossible to say how big a part lice played. The dark
shadow over this picture is that salmon farming companies in BC seldom tell the public
when they have disease outbreaks other than lice. If the farms in the Discovery Islands
area had infectious haematopoietic necrosis (IHN) when the juvenile sockeye swam past
on their way to the ocean, the sockeye could easily have been wiped out. IHN is the main
reason sockeye are cultured only in fresh water, and it is a reasonable suspect because
there was an epidemic of IHN on salmon farms in the Broughton Archipelago during the
winter of 2002–2003. IHN is endemic in BC, but as with sea lice, it is the magnitude of
the infection pressure from farms that makes the difference.
A scientist from the sea lice group at DFO is now saying that the 2009 Fraser sockeye
returned in low numbers because when they were juveniles migrating north in Georgia
Strait they found little to eat. If the publications of this scientist regarding sea lice were
less misleading, I might be inclined to believe him, but given his record, I reserve
judgment. Don’t expect the Cohen Commission to discover what caused the loss of the
2009 Fraser sockeye unless they subpoena all corporate records relating to diseases on
salmon farms in the Discovery Islands. When I asked Dr. David Levy, the scientific
9
manager of the Cohen Commission, about this he replied that it is easy for the
Commission to extract information from DFO, because the Commission is also a federal
agency, but that it is more difficult to extract information from the BC government or
from the companies. In other words, don’t hold your breath. But even if the Cohen
Commission obtains those disease records there may still be no certain answer. Often
records aren’t kept. What is certain is that the farms
