A river can grow all the young salmon in the world, and it will not matter if the ocean cannot feed them.
The sea has a carrying capacity — a limit to how many salmon and steelhead it can carry to adulthood — and that limit is not fixed. It rises and falls with the health of the whole food web, from the krill on up. Real recovery means tending that entire web, not just protecting fish on paper. Two things, both of our own making, are pushing the ocean's capacity down.
Predators and prey rise and fall together, in cycles that run for years. When the fish run thin, the animals that eat them — seals, sea lions, even whales — go hungry, and their numbers ease back until the prey recovers. That fluctuation is natural and healthy. We are not here to declare war on predators; the goal is balance, with the human thumb taken off the scale.
The trouble is that we have put our thumb on the scale in two places — concentrating predators where they never used to pile up, and stripping the food web out from under everything at once.
Start with what is at stake at the top. The Southern Resident killer whales — an endangered population of roughly seventy animals — eat almost nothing but Chinook salmon. Their decline tracks the Chinook's, animal for animal. They are not a problem to be managed; they are starving for our salmon. Every fish we bring back is food in their mouths.
Farther out, the gray whale tells the same story from the seafloor. Gray whales farm the bottom, rolling into the mud to filter out the tiny shrimp-like amphipods that make up most of their diet — and they, too, have been hit hard by a die-off in recent years as their food has faltered. Orca or gray, the whales are not overgrown. They are hungry, at the end of a food web we are pulling apart. Feed the web, and you feed the whales.
The first distortion is close to home. Seals and sea lions were protected in 1972 and recovered strongly — a genuine conservation success. But the dams, locks, and fish ladders we built now funnel migrating salmon through narrow chokepoints, and the predators have learned to wait there. Along the coast, the Chinook eaten by seals and sea lions has risen roughly tenfold since 1970 — by 2015 it equaled everything the resident orcas ate, and about twice what all fisheries caught combined. At Bonneville Dam, sea lions have taken up to 44% of a spring Chinook run in a single season.
Where a specific, learned problem animal is eating an endangered run at a man-made chokepoint, the answer is targeted, humane removal — carried out under federal permit and co-managed with the Columbia River treaty tribes, as it is today. Not a cull. A scalpel. And here is the part that matters most: every Chinook that does not get eaten by an overconcentrated sea lion at a dam is a Chinook that can make it home — or out to the starving orcas. Managing the pileup and feeding the whales are the same act.
The young fish are hit long before they ever reach the ocean. On the Columbia, eleven years of research out of Oregon State found something almost nobody talks about: for Upper Columbia steelhead, fish-eating birds killed more young fish than every other cause combined — the dams included. Caspian terns and double-crested cormorants together eat roughly one in five of all the juvenile salmon and steelhead that reach the estuary. Steelhead take it worst, because they travel high in the water column — right where the birds hunt.
And it is, once again, a problem of our own making. The great bird colonies swelled on dredge-spoil islands we built, and the slow reservoirs behind the dams strip young fish of the cover and current they need — turning a fast run to the sea into a slow crawl past a wall of beaks. The good news is that careful, targeted management works: moving and limiting the Caspian tern colonies cut their toll on steelhead by about 70% since 2020. The warning is what happens when it's done bluntly — a botched cormorant cull in 2015 scattered the colony onto the Astoria–Megler Bridge, where the birds now eat more salmon than before. With a scalpel, predator management is one of the fastest levers we have. With a sledgehammer, it backfires.
The second distortion is bigger, and it works at three levels at once. Industrial fishing — above all the great factory trawlers — does not just catch fish. It takes the food web apart from the top, the base, and the bottom.
Trawl nets sweep up the very fish we are trying to save. More than 1.7 million king salmon have been recorded as trawl bycatch since 1991 — hauled up and discarded as a byproduct of chasing other fish.
Krill are the foundation of the whole marine web, feeding the small fish that salmon, seabirds, and whales all depend on. The US West Coast banned commercial krill harvest in 2009 to protect it — but elsewhere the base is still being mined.
Gear dragged across the seabed tears up the bottom communities — the amphipod beds that gray whales and much of the food web feed on. What takes decades to grow can be flattened in a single pass.
The krill story is the clearest warning — and the clearest proof we can do better. Off our own coast, we saw the danger early and banned commercial krill harvesting outright in 2009, before a single fishery took hold, precisely because krill feed everything above them.
The rest of the world has gone the other way. In 2024 the Antarctic krill catch hit a record — over half a million tonnes — and the fleets that take it (largely Norwegian, Chinese, and South Korean) concentrate their nets in a few narrow straits that are exactly where hundreds of thousands of penguins, seals, and whales gather to feed. Most of it becomes fish-farm feed and diet supplements. The international body meant to manage it has stalled two years running while the biggest players push to catch more. We banned it at home for good reason. The reason has not changed — only the ocean it is playing out in.
This is the same answer we give on dams: do not just ban it — build something better. The goal is not to end commercial fishing or take protein off anyone's table. It is to end the collateral damage while keeping the catch.
Better ways already exist. Selective and terminal fisheries that harvest where target stocks concentrate, traps and pots, hook-and-line, weirs, and modern bycatch-reduction gear can all put fish on the dock without strip-mining the web to do it. The whitefish protein a trawler chases can be caught other ways; the bycatch, the flattened seafloor, and the mined-out forage cannot be caught back. Trawling is an industry long overdue for the kind of innovation that contributes to restoration instead of working against it — and reforming it lifts the ocean's capacity for salmon, steelhead, and whales alike, at little cost to most of us.
We hold this to an honest standard. The Alaska pollock fishery that most trawlers serve is the largest in the country by volume, a real source of affordable food and thousands of jobs, and it is reasonably well managed on the fish it actually targets. The problem is the collateral — the bycatch, the forage, the seafloor — not that people eat fish. And bycatch is one major pressure on salmon, alongside habitat, climate, and marine survival, not the single cause of their decline.
The same restraint applies to predators. Seals, sea lions, and whales are recovered native animals, not vermin; the case for managing them is narrow, humane, and co-managed, and it applies only where our own structures have thrown the balance off. Our argument is for reform and innovation matched to the evidence — not a slogan, and not a war on wildlife.
The ocean's capacity is the thread that ties this whole collection together. It is the same limit that warns against flooding the sea with too many hatchery pinks, and the same reason passage and habitat only pay off if there is a living ocean waiting on the other side. Tend the web — keep predation in balance and reform the harvest that strips it — and the fish we plant have room to survive, the runs rebuild, and the whales get fed.
It is active recovery, applied to the sea: intervene and tend the system, rather than manage its decline. 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.