sm-220-pink-salmon-3502674

A pink salmon is shown near a dam near Washington. The new study found that when sea lice populations were high among farmed fish, there was much higher mortality among wild salmon that swam near fish farms compared to those who did not swim near fish farms. (Dean J. Koepfler/The News Tribune/Associated Press)

A new study has found that sea lice infections at B.C. fish farms are linked to increased mortality among wild pink and coho salmon, contradicting the conclusion of a previous study of pink salmon that found no link.

"Our works suggests that when one looks at all the available evidence there is indeed good reason to minimize lice on farmed salmon and the potential for their transmission to wild fish," said Brendan Connors, a Simon Fraser University researcher who co-authored the study to be published this week in Proceedings of the National Academy of Sciences.

Sea lice are parasites that infect many salmon species and some studies have suggested that young wild salmon can become infected with sea lice as they swim past salmon farms, where large numbers of salmon are kept in a confined space.

Some studies have also suggested that when young wild salmon are infected with large numbers of sea lice, the infection can have a negative impact on their health, behaviour and survival.

Study vs. study

The new paper published by Martin Krkosek included criticism of the paper by Gary Marty last December, which concluded that pink salmon were unaffected by sea lice from fish farms.

salmon-krkosek-071213

Sea lice are parasites that infect many salmon species, including this juvenile pink salmon. ((Courtesy of Alexandra Morton/Science))

Krkosek said that while Marty’s paper claimed to have analyzed data going back 60 years, only the data from 2000 to 2009 were used to determine statistically whether sea lice affected pink salmon populations.

He said his own paper used data from the 1970s to separate the role of other sources of variation among salmon populations from the role of sea lice.

"Without having those control data sets in there, the effect of the sea lice is lost in all the variation," he said. "You don’t see it."

Marty did note in an FAQ on his web page that, from 1975 to 1977, the number of pink salmon returning to spawn was far lower in the Broughton Archipelago than nearby, despite the absence of fish farms at that time. He said that using early population data as controls was not appropriate due to a lack of disease data for those years.

For his part, Marty had criticisms of Krkosek’s paper. In particular, he questioned its use of data from individual salmon populations.

In an interview, he said his own study aggregated the data because there was no way to know where each population migrated to and which farms they were exposed to.

He added that he assumed sea lice patterns were the same before and after 2000, when farms started keeping statistics. Krkosek’s paper removed that data from the analysis and treated it as "missing data."

Marty said the main difference between the two papers is the assumptions they make to deal with unknowns. He added that Krkosek and his colleagues re-analyzed almost the same data and he doesn’t think they saw "a huge, significant difference" in the end, despite their different conclusion.

Salmon farms and sea lice are among the factors being examined by B.C.'s Cohen Commission, which is trying to explain the possible reasons for the declines in the population of Fraser River sockeye salmon over the past decade.

In general, very few of the salmon hatched in freshwater rivers, which swim out to the ocean as juveniles, ever return to their hatching grounds as adults to spawn, Connors said. They die from a variety of causes, such as being eaten by predators.

"The question is even if some sea lice come from salmon farms and infect fish and possibly even kill them, does that really matter in the grand scheme of things?" Connors said. "And that's what management and policy really care about that."

This past December, a study led by University of California researcher Gary Marty, published in the same journal,  found the survival of wild pink salmon populations appears unrelated  to the number of lice found on farmed fish nearby in the Broughton Archipelago, which is located off the B.C. coast, just to the east of northern Vancouver Island.

Marty and his colleagues said they analyzed 10 to 20 years of fish farm data and 60 years of pink salmon data to arrive at that conclusion.

"That was a really important paper. It received considerable attention," said Connors, adding that Marty was the first researcher to gain access to fish farm data on sea lice infections.

Marty's paper suggested that growing salmon in closed containers (instead of net pens open to the ocean) and other costly measures to minimize sea lice transmission to wild fish would not benefit wild salmon populations.

He noted that he and co-authors Sonja Saksida of the non-profit Centre for Aquatic Health Sciences in Campbell River, B.C., and Terrance Quinn of the University of Fairbanks, Alaska, said their study was not funded by governments, environmental groups or industry.

80 per cent higher mortality

This more recent study, co-authored by Connors and led by Martin Krkosek, a Canadian researcher at the University of Otago in New Zealand, used Marty's fish farm data along with population data for a wider geographic area. It also included an extra species of salmon in addition to pink salmon — coho — and a larger time span, from 1970 to 2009. (Fish farming in the Broughton Archipelago began in the 1980s.)

The study received funding from the federal government and two non-profit environmental groups — the Watershed Watch Salmon Society and Save our Salmon Society.  

It found that when sea lice populations were high among farmed fish, there was much higher mortality among wild salmon that swam near fish farms compared to those that did not swim near fish farms, Krkosek said.

When sea lice populations were low, there was no difference in mortality between wild fish either near or far away from salmon farms. However, he acknowledged that there was a wide variation in the effect of sea lice depending on their levels.

In general, regardless of the presence or absence of sea lice, four out of five juvenile salmon die within their first couple of months in the ocean, Krkosek said: "Everything eats them."

But his study shows that among the one in five survivors, up to four-fifths of them may die from the influence of sea lice from fish farms.

Connors said that suggests that fish farms should continue to take measures to prevent the transmission of sea lice to wild fish. Such measures include using chemicals to kill sea lice and  emptying pens along wild salmon migration routes when they are migrating.

"Management and policy measures designed to protect from salmon farms will in fact have benefits for pink and coho salmon as well as coastal economies and the coastal ecosystems that depend on the salmon," he said.

Other researchers on Krkosek's team included Alexandra Morton, a researcher at the Raincoast Research Society who has been a vocal critic of fish farms, along with scientists from Simon Fraser University, the University of Alberta and the University of Washington in Seattle.

Connors is scheduled to testify before the Cohen Comission on Aug. 25, 26 or 29. Marty is to appear on August 31, and Saksida on Sept. 6.