Killing Cormorants Won't Help Salmon, Says Government’s Own Study

The Army Corps' controversial plan to cull thousands of birds to save endangered salmon populations may be as futile as opponents have claimed.

For Columbia River salmon, life is full of perils—every year, after hatching in the river's headwaters, juveniles must pass through a gauntlet of predators and dams as they migrate hundreds of miles to the sea. Near the end of the journey, just after the 155 million or so survivors swim through the final dam and taste the salty freedom of Oregon’s Columbia River Estuary, they find themselves confronted with a 13,000-pair colony of Double-Crested Cormorants nesting at East Sand Island. Those hungry birds can eat as many as 20 million salmon smolts in a single year.

Typically such predation would be chalked up to the circle of life and met with a shrug. But because many of these salmon populations are listed under the Endangered Species Act, the cormorants have found themselves in the center of a tussle. In March the U.S. Army Corps of Engineers approved a plan—with the support of U.S. Fish and Wildlife—to reduce the size of the cormorant colony to 5,600 breeding pairs through a combination of scaring the birds off, oiling eggs, and shooting adults. In response, five conservation groups, including the Audubon Society of Portland, filed a lawsuit to prevent the killing, citing the lack of evidence that culling cormorants will help salmon populations. 

"From the beginning, we've felt that cormorants are being scapegoated," says Bob Sallinger, conservation director of the Audubon Society of Portland. "You can kill all the cormorants on the Columbia River and it's not going to recover the salmon." 

Despite the ongoing lawsuit, the cull began in May, and so far 158 adult birds have been killed and more than 5,000 nests oiled.

The conflict is a tricky one because it hinges on an unanswered question: Are cormorants eating unhealthy or healthy salmon? This matters because very few salmon survive their 3-4 years at sea before they return to spawn, and the young smolts that head into the ocean at a health disadvantage—such as injury, disease, or parasites—are pretty much fated to die, either becoming food for a hungry predator or starving to death. If cormorants pick off these unhealthy smolts that otherwise would die at sea, they don't do additional damage to salmon populations (this is called "compensatory mortality"). But if the smolts they eat each year are strong and healthy enough that they would have been able to survive their time at sea and return to spawn, the cormorant predation does additional harm (called "additive mortality").

The Army Corps plan assumes that all cormorant-induced mortality is additive. "By the time these fish get that far down in the estuary, they pretty much have an equal chance of surviving to be adults," says fisheries biologist Ritchie Graves of NOAA Fisheries, West Coast Region. "We don’t think fish that have gotten that far are damaged in any significant way."

New evidence released by court order suggests otherwise. A draft Fish and Wildlife analysis made public earlier this month examined 11 years of data to see whether there was a relationship between cormorant predation and the return rates of salmon. Since cormorant predation varies year to year, the report was able to assess whether the amount of fish eaten by cormorants ultimately affected the number of salmon that returned to the Columbia to spawn—and it found no relationship between the two factors.

 "The data indicate that efforts to reduce predation by double-crested cormorants are unlikely to have an effect on Snake River steelhead abundance or productivity," concluded Fish and Wildlife Service biologist Steve Haeseker, who authored the report. (The Snake River is the largest tributary of the Columbia River.)

Spokespeople for the Army Corps and Fish and Wildlife declined to comment, citing ongoing litigation.

Additional evidence supports the theory that cormorants eat unhealthy salmon. In one study, researchers from Oregon State University tagged steelhead in the Snake River and recorded whether the fish appeared to be sick or injured. After the steelhead migrated out to sea, the researchers visited the cormorant colony to collect the tags from fish they consumed. They found the cormorants ate both healthy and unhealthy salmon in roughly equal amounts, indicating that at least some of the deaths-by-cormorant are compensatory.

Still, the USFW analysis is not the final word. For one thing, it’s a draft and hasn't undergone scientific peer review. For another, the question it grapples with may be impossible to definitively answer, according to NOAA Fisheries' Graves. "You're trying to pick out this tiny little signal in this vast ocean of noise," he says. "We're not saying that compensatory mortality is zero; we're saying that we don’t have a good way to estimate it and we don’t think it's very high."

But Sallinger argues that whether or not the draft study provides a full explanation of what the animals are up to, it’s fishy that it was kept secret. For months, in court and outside of it, the plaintiffs requested any data or reports on compensatory mortality, but they never saw this one, he says. It wasn't until the judge ordered the defendants to share the analysis that they did so.

"All that time, they knew they had this analysis from their own biologist saying that this isn't going to have an impact," Sallinger says. "They had an obligation to disclose and discuss that document. They didn’t have the right to simply bury it."

For more on the East Sand Island cormorant population, watch this short documentary filmmaker Judy Irving made in collaboration with the Audubon Society of Portland.

Correction: An earlier version of this story misstated the proportion of tagged unhealthy salmon consumed by cormorants, according to the study from Oregon State University. We’ve updated those numbers.


Stay abreast of Audubon

Get updates about our conservation work and how to help birds.