Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Salmon study sparks row over dams

Results dismissing link to fish mortality are called into question.

Free-flowing water might not aid chinook salmon as much as some believe. Credit: M. Conlin/V&W/

Research that argues dams have no direct effect on the migration of juvenile salmon is roiling waters in the US Pacific Northwest.

The study, published this week in PLoS Biology, uses a new way to tag fish to compare the heavily dammed Columbia River system with the free-flowing Fraser River to the north in British Columbia, Canada. After hatching upriver, salmon smolts fared equally well on their journeys down the two rivers to the ocean. But that's where the agreement ends.

Salmon migration remains at the heart of an ongoing court battle, which pits environmentalists who want to see some of the dams removed against federal agencies charged with maintaining the dams, providing power and protecting salmon. Thirty-one federal dams on the Columbia River and its tributaries collectively provide some 60% of the region's electricity.

The Bonneville Power Administration, based in Portland, Oregon, has spent billions of dollars making the dams more friendly for fish, and the latest plan, issued on 5 May, proposes further modifications to dam operations. Taken at face value, the new research might suggest that the investment is paying off and that other factors are suppressing the 13 salmon populations protected under the Endangered Species Act.

But environmentalists and several salmon biologists pounced on the study, suggesting that industry funding might have biased the results. These critics question the value of comparing the two rivers and say that the study doesn't even address what many think is the dams' biggest effect: stressed smolts dying after they reach the ocean.

Everybody assumed that we would simply see much higher mortality in the dammed system.

Still, the study explicitly acknowledges these limitations. Lead author David Welch, founder of Kintama Research in Nanaimo, British Columbia, and an adjunct professor at the University of Victoria, has received nearly US$4.2 million from Bonneville Power for his work, although his largest funder is the Gordon and Betty Moore Foundation. However, he says that his funders had no role in vetting the study. "Everybody assumed that we would simply see much higher mortality in the dammed system — we didn't find that," he says. "We don't want to oversell it and say it definitively answers these questions, but it's pretty damn surprising."

But two of Welch's co-authors, Shaun Clements and Carl Schreck, who did their work at Oregon State University in Corvallis, are concerned that the study is being oversold. Clements says that a press release issued by PLoS Biology with the headline "Dams make no damn difference to salmon survival" goes too far. "This study doesn't take dams off the hook," he says. It may point to problems in the Fraser rather than clearing dams on the Columbia, he adds.

Sound trackers

Tagging studies typically rely on passive sensors implanted in the fish, but the sensors have a short range and can be counted only as the salmon are funnelled past receivers at the dams. In contrast, Welch's company uses powered transmitters that send acoustic signals to receivers, allowing the researchers to track the salmon in free-flowing rivers and along the coast.

The team tagged hatchery spring chinook and wild steelhead on the Fraser River from 2004–06 and then compared those results to other tagging work done in the Columbia. In 2006, the team also tagged hatchery spring chinook fish on the Columbia's Snake River tributary and tracked fish as they navigated eight major dams to the ocean. The numbers of surviving fish varied widely, Welch says, but were statistically indistinguishable between the river systems (D. W. Welch et al. PloS Biol. 6, e265; 2008).

Michele DeHart, manager of the Fish Passage Center in Portland, which tracks salmon statistics, says the comparison is "seriously misleading". She points out that the Fraser River has its own problems, including an infestation of bark beetles, which could affect water quality, and that federal dam operators have been under a court order since 2005 to help juvenile salmon by spilling water over the dams during spring and summer migrations.

Ed Bowles, the top fish biologist for the Oregon Department of Fish and Wildlife, says the results merely underscore the notion that delayed mortality is the real killer. He points to work showing that the number of salmon returning upriver to their original spawning grounds — years after their initial migration to the ocean — is several times higher for fish that navigate fewer dams in the lower reaches than for populations that must traverse eight dams to reach the sea (H. A. Schaller and C. E. Petrosky North Am. J. Fish. Mgmt 27, 810–824; 2007).

The additional energy and stress required to move through the dammed system seems to inhibit survival in the ocean, he says. "This paper doesn't really do anything to dispel that working hypothesis."

Welch says he is now looking into the issue of delayed mortality by tracking the same salmon populations after they enter the ocean and head up the coast to Alaska.


Related links

Related links

Related external links

Fish Passage Center

Kintama Research

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Tollefson, J. Salmon study sparks row over dams. Nature (2008).

Download citation

  • Published:

  • DOI:


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing