Two recent studies demonstrate how dam breach paired with increased spill in the mid-Columbia would allow many Snake River spring/summer Chinook populations to reach various management goals
I am sitting on my parent’s porch on a Sunday afternoon in July when the neighbors stop by to say hi. “Oh! You are a fish biologist,” they say after chatting about the upcoming work week.
I know what the next question will be before they even ask it. “What do you think about the Snake River Dams––should they be removed?”
I know this question is coming because I live in Walla Walla, Washington, near the confluence of the Columbia and Snake Rivers, where people have deep connections to the irrigation, power generation and jobs that the lower Snake River dams provide.
Because of my background, my answer always focuses on the fish.
I tell them, “You can’t escape the math of it. The dams and their reservoirs kill so many fish that the populations cannot sustain themselves. And if we do not eliminate that cause of mortality, meaning breach the dams, the spring/summer Chinook populations will likely go extinct.”
Tracking fish through the dams
We know this because of the herculean tagging effort that has taken place in the Snake River for the past 30 years. Millions of juvenile salmon and steelhead have been tagged with PIT tags (akin to microchipping dogs and cats), and their survival through the Snake and Columbia River dams and in the ocean are monitored.
These data allow calculation of a metric called the “smolt-to-adult return”, or SAR, which represents how many ocean-bound smolts make it back as adults. It is an essential metric because it captures almost all dam-related mortality these fish experience.
This tagging work is led by the Comparative Survival Study (CSS), which is a highly regarded, collaborative program that includes membership from the federally funded Fish Passage Center, the US Fish and Wildlife Service, Columbia River Intertribal Fish Commission, as well as Idaho (IDFG), Washington and Oregon Departments of Fish and Game/Wildlife.
Analyses conducted by the CSS in 2017 and 2019 have estimated that dam breach would increase smolt-to-adult returns for wild spring/summer Chinook salmon at Lower Granite Dam (the dam furthest upstream) by an average of 1.58 to 1.91 timesa, depending on what additional water management actions were taken in the lower river.
These predicted survival benefits would bring smolt-to-adult survival at the Lower Granite Dam up from an average of 1.68% (from 1996-2015b) to an average of 2.7-3.2%.
What has not been known until this year is:
- Would this improved SAR survival be enough to make a meaningful difference for spring/summer Chinook populations?
- Would it lead to population growth and stabilized Snake River spring/summer Chinook populations?
- Or, even more ambitiously, could it lead to healthy and harvestable population goals?
New research points, yet again, to dam breaching
Two peer-reviewed scientific studies published this year tackled this question. And, importantly, despite different methods, data and study scales (across numerous populations vs. subpopulations), they both demonstrated the resilience of Snake River spring/summer Chinook populations and their remarkable intrinsic ability to rebound.
Together, these studies found that the increased SARs that are expected from breaching the lower four Snake River dams would not only lead to population recovery and meet Endangered Species Act (ESA) delisting goals in most cases, but populations in the best freshwater habitats could reasonably reach ‘healthy and harvestable’ levels.
One study by authors from the IDFG and the Nez Perce Tribe (Copeland et al. 2024) measured SAR survival rates needed to reach three distinct population goals:
- delisting from the ESA (a minimum goal),
- intermediate abundances (between low and high goals; serves as checkpoints for progress toward healthy and harvestable)
- and the healthy and harvestable abundance goal agreed upon by the Columbia Basin Partnership.
They looked at how many smolts were produced per adult female that spawned in each of 14 populations across the Snake River watershed. The 14 populations varied in how much human-caused impacts they have experienced (i.e., hatcheries, habitat degradation), which gave the authors insight into how current population status would affect the SARs needed to reach various goals in individual populations and in aggregate across the Snake River basin.
Not surprisingly, they found that populations in the wilderness regions of the Snake River basin needed a much lower SAR to reach the various goals compared to populations in human-impacted regions.
However, what was notable was that the SARs needed to reach delisting goals and stabilize the populations were far less than what is expected from breach in more than half (8 of 14) of the populations.
This indicated that the Snake River spring/summer Chinook basin-wide population (combined individual populations or ‘in aggregate’) could reach the abundances needed as a major component of delisting if the dams were breached and spill was increased in the mid-Columbia.
Middle Fork Salmon River study
A complementary study by Jacobs et al. 2024 estimated what SAR survivals would be needed for population growth across a wilderness watershed of the Snake River (the Middle Fork Salmon River).
This study evaluated outmigrating smolt counts and 24 years (and more than 18,000 observations!) of redd counts collected by the US Forest Service’s Rocky Mountain Research Station, IDFG, the Shoshone-Bannock and Nez Perce Tribes and the Forest Service to estimate the annual population productivity (i.e., smolts per redd). These productivity estimates were then used in a different population model to evaluate influences on productivity over time and across a range of watershed conditions.
This study identified the SAR survivals that would be needed to reverse population declines and increase population sizes in this system.
They found that SAR survivals of 1.8% would result in positive population growth in all 23 Middle Fork Salmon River segments studied. Notably, 1.8% is well below the SAR survival of 2.7-3.2% predicted from removing the four Lower Snake River Dams, meaning there is also ample room for realizing further growth in the future.
As with the Copeland et al. study, this indicates that dam removal combined with reasonable water management in the mid-Columbia would be enough to recover and rebuild spring/summer Chinook in the Middle Fork Salmon River.
We know how to fix the problem
Collectively, what we can learn from these two studies is that dam breach and adjacent water management alone would not lead to population recovery in the most severely degraded habitats, though they are essential for realizing the gains of ongoing habitat restoration and other management actions here. But they would, on their own, lead to population recovery for those that still have intact freshwater habitat, much of which falls within the Frank Church Wilderness, the largest contiguous wilderness in the Lower 48.
Importantly, given the anchoring effect of intact habitats the Snake River spring/summer Chinook would meet delisting abundance goals as a result of dam removal and improved water management. This is consistent with conclusions from the NOAA Rebuilding Report where the authors concluded that dam removal is the cornerstone of not only delisting populations but reaching the Columbia Basin Partnership’s ambitious healthy and harvestable goals.
Meanwhile, back on the porch
So, what I want to say on the deck in the afternoon sun is if we lose spring/summer Chinook in the Snake River basin, it won’t be because we didn’t know what to do nor how to fix the problem. It will be because we chose not to act on the information we had.
The data are clear; spring/summer Chinook salmon in the Snake River will go extinct without breach and, similarly, they could be recovered with dam removal and increased spill in the mid-Columbia.
It’s a straightforward math problem that we should stop avoiding before it’s too late.
Tell Congress to remove the lower four Snake River dams.
Footnotes
aestimate based on arithmetic means of SAR improvements relative to a no-action alternatives (NAA vs MO3 and MO34) in Table 2.4 of the 2019 CSS Report
bestimate based on arithmetic mean of SARs in Table 1.4 of Lawry et al. 2019