If you wanted to design the perfect test of "just leave the salmon alone," you would build the Olympic Peninsula.
The Hoh, Queets, Quinault, and Quillayute are the last great wild-steelhead rivers in the Lower 48 — the ones anglers travel across the country to fish, the rivers held up as what a salmon river is supposed to look like. Most of their watersheds sit inside Olympic National Park. No major dams. Almost no development. Cold, clean water running out of a temperate rainforest, with only a light touch of hatchery influence. By every measure the hands-off philosophy holds dear, these rivers should be thriving.
They are not. And that is exactly why this is the sharpest warning we have.
Every other place in this collection comes with a confound. The Columbia has its dams. The Delta has its water wars. Puget Sound has its cities and its shorelines. When those runs fail, there is always something concrete to point at — a wall of concrete, a diversion, a paved-over estuary.
The Olympic Peninsula has none of that. This is as close as the real world offers to a controlled experiment in passive recovery: take the wildest, most protected, least-altered salmon rivers on the map, manage them the hands-off way — by regulation and restraint, with the habitat left largely intact — and see what happens over thirty years. The habitat here is not the variable. It is the constant. Which makes what happened impossible to blame on a dam.
Across the four largest Peninsula basins, wild winter steelhead have fallen roughly 42% from the early 1990s to the most recent five-year period — and managers have been forced into closures that would have been unthinkable a generation ago on these rivers.
Combined wild winter-steelhead abundance across the Hoh, Queets, Quinault, and Quillayute, early 1990s compared with 2018–2022 — the four rivers that were supposed to be the safe ones.
Olympic National Park closed the Queets to steelhead for 2024–25 and again for 2025–26 — two consecutive years — citing critically low forecasted returns. On the Hoh, the season was cut to a narrow winter window, wild fish release-only.
The federal government's own scientists concluded in November 2024 that Olympic Peninsula steelhead face a moderate risk of extinction — on the very rivers held up as the healthy ones.
Here is the part that captures the whole problem. In November 2024, NOAA Fisheries completed a status review and found that Olympic Peninsula steelhead face a moderate risk of extinction. That is the government's own alarm bell, rung by its own biologists.
In February 2026, NOAA declined to list them under the Endangered Species Act.
Sit with that. The agency looked at the wildest steelhead rivers in the country, said they carry a real risk of blinking out — and then declined to give them protection. "Not listed" does not mean "not in trouble." It means the system will not act, even on a warning it wrote itself. That is passive recovery in a single decision: measure the decline, document the risk, and manage the paperwork while the fish keep sliding.
If the habitat is intact — and here it genuinely is — then the forces driving these fish down have to lie elsewhere in the life cycle. And they do.
The prime suspect is the ocean. These fish leave clean rivers in good shape and die at sea, in the years when marine survival collapses and the food web thins. Layered on top is the slow arithmetic of cumulative harvest across a fish's whole journey, and the predation that mounts when runs are already thin. And on the Hoh, there is a stressor no streambank project can touch: the glaciers feeding the river are receding, reshaping its flows and temperatures as the climate warms.
Not one of those is a habitat problem you can fix by fencing off a riverbank and waiting. They are whole-life-cycle problems — and they are invisible to a recovery philosophy that treats "protect the river and step back" as the entire toolkit.
Active recovery on the Olympic Peninsula does not mean flooding these wild rivers with hatchery fish. It means refusing to manage only the one thing — the habitat — that is already in good shape, and finally working the parts of the life cycle that are failing: rigorous, science-based harvest management; attention to marine survival and the forage base that feeds these fish at sea; predation management where it is thrown out of balance; and, for a run pushed critically low, the willingness to use careful active production as a bridge until it can stand on its own.
This is not hypothetical here. On the Quinault, the Quinault Indian Nation is doing exactly this for its blueback sockeye — rebuilding lost spawning habitat with engineered logjams, holding hard harvest discipline, and running a hatchery supplementation program, all at once, on a run that isn't even federally listed. The people closest to these fish are already managing the whole life cycle. The philosophy that failed these rivers is the one that says they shouldn't have to.
This is not a story with a villain. A large share of the decline is driven by ocean conditions that no manager fully controls, and honest people can disagree about how much of the rest is harvest, predation, or climate. Managers have acted — the closures on the Queets and Hoh are real, painful decisions that put the fish first, and they deserve credit for making them. And the anglers and advocates who prize these wild rivers are right to: they are irreplaceable, and keeping them wild is a goal we share.
Our argument is narrower, and harder to wave off. It is simply this: on the rivers where the habitat is as good as it gets and the fish are declining anyway, protection alone provably did not hold the line. That is not an indictment of anyone's intentions. It is a fact about what "leave them alone" can and cannot do.
Somebody can always argue away the Columbia, the Delta, or Puget Sound — blame the dams, the farms, the sprawl, and insist the fish would be fine without them. Nobody can argue away the Olympic Peninsula. There is nothing here to blame but the approach itself.
The wildest, most protected steelhead rivers in the country were handed the ideal conditions passive recovery says produce healthy fish — and they fell about 42% while their own government watched, measured the risk, and declined to act. If hands-off recovery could work anywhere, it would work here. It didn't. That is the whole case for active recovery, written on the one river system where every excuse runs out.
See the rest of the record, read about how we work, or get involved.
The figures on this page are drawn from public, authoritative records:
Questions about our facts and figures? Contact us.