Fixing One River Confluence Could Save This Salmon Run and Hurt the One Next Door

Two spring Chinook Salmon populations spawn in neighboring streams in the same northeast Oregon watershed. Model the same river restoration fix for both, and a new study predicts it would help one and hurt the other. Restoring the historical confluence between Catherine Creek and the Grande Ronde River is predicted to raise smolt survival 69 percent for Catherine Creek fish, and cut it 38.4 percent for smolts coming down the Grande Ronde River itself.

The Catch

Read that gap as a caution before anything else. Those percentages come out of a model, a prediction of what restoring the historic channel configuration would likely do, not something anyone has actually watched happen on the river. The rest of the study carries a similar asterisk. Researchers tracked predation using tags recovered from bird colonies, and they're upfront that this only counts tags the birds actually carried back to the nest, so real predation on these smolts is probably higher than what got measured. This is also two adjacent tributaries in one river basin over one specific stretch of years, not a general rule for how every Pacific Northwest river would respond to the same kind of restoration.

The Study

USGS researchers radio-tagged 924 spring Chinook Salmon smolts, 857 of wild origin and 67 raised in hatcheries, across Catherine Creek and the Grande Ronde River in northeast Oregon. They tracked the fish every spring from 2011 through 2017, March through June, as they emigrated downstream through what the study describes as hydrologically altered natal streams. Using the tag data, the team calculated survival reach by reach and matched it against water discharge, fish size, current velocity, and how much pressure came from fish-eating birds. They then applied a predator-prey model to estimate how far, on average, a smolt swam before running into a predator, before finally testing what restoring the historic stream confluence would do to the numbers.

What They Found

Catherine Creek smolts died at a high rate inside their home stream, then survived at close to 100 percent once they reached the Grande Ronde River. Grande Ronde River smolts showed the mirror image: low mortality upstream, then high mortality through the Grande Ronde Valley. Across both groups, survival rose with higher discharge, bigger body size, and faster current, and fell as bird predation increased. One detail stood out on the hatchery side: wild-origin Catherine Creek smolts swam farther between predator encounters than hatchery smolts from the same stream, even though the hatchery fish were bigger and moved faster. Bigger and faster didn't mean safer.

What This Means on the Water

I've never fished for salmon. It's the one major species in my own experience bank that's an honest blank, so I'm not going to pretend I've got a felt sense for what a Grande Ronde smolt run looks like from a boat. What I can say is that this study reshaped how I think about habitat fixes generally. My instinct, before reading this, would have been that restoring a river to something closer to its old, natural shape is close to a universal good. This study is a reminder that a river is a connected system, not a series of independent fixes, and a change that helps fish in one stretch can shift the problem downstream rather than solve it. That's the same logic behind why a single dam removal or one stretch of new woody habitat doesn't automatically fix a whole watershed, something I touched on when I covered a Wisconsin habitat study a few months back. It also lines up with a broader pattern I keep running into covering Chinook Salmon specifically. I wrote about the Yukon River Chinook collapse back in July, a totally different river and a totally different cause, and between the two studies the throughline is the same: there's rarely one clean lever to pull for salmon recovery. This is one modeled scenario on two Oregon streams, not a verdict on river restoration everywhere, but it's a good reason to want to see the actual before-and-after data before calling any single restoration project a win.

Citation

Favrot, S. D., Perry, R. W., & Kimbrel, S. W. (2026). Biological, environmental, and geomorphological factors influencing reach-specific survival of spring Chinook Salmon smolts upstream of the Columbia–Snake River hydrosystem. North American Journal of Fisheries Management, 46(4), 711. https://doi.org/10.1093/najfmt/vqag035