Big question is why

After years of mainly sitting on Cook Inlet beaches, commercial setnet salmon fishermen were back on the water this week as returning sockeye offered a lesson in how little we know about the survival of salmon at sea.

Back in January, the Alaska Department of Fish and Game forecast a 2025 return of 6.93 million of these three- to five-year-old fish based on the number of spawners entering the river in 2020 to 2022, the number of smolt going to sea, and the average of past survival at sea.

As always, of course, the agency couched the forecast with the warning that “salmon fisheries
are inherently uncertain and (forecasts) are primarily used to gauge the general magnitude of expected runs and guide early-season management strategies.”

And thus the broader forecast “range” gave the return an 80 percent probability of falling between 5.41 million and 8.45 million. The actual return has already exceeded that window, and sockeye are still storming back to the Kenai River, the Inlet’s biggest sockeye producer.

As of yesterday’s count, almost 3.7 million sockeye had safely escaped into the river – two and a half times the in-river goal of a maximum of 1.4 million – and the total, in-river return was projected to top 4 million, even with the setnetters now busy picking off late returnees.

Meanwhile, the Cook Inlet harvest of sockeye had as of today surpassed 4 million; known sockeye returns into other Inlet index streams, chief among them the Kasilof River, had reached 1.2 million; and harvests in so-called “personal-use fisheries” at the mouths of several Inlet rivers were expected to have topped 400,000.

All of this adds up to a known return of at least 9.6 million sockeye, but Fish and Game calculates an estimated 17 percent add-on for returns to more than a dozen “unmonitored” sockeye systems feeding the Inlet.

Those add-ons would bring the total return to somewhere over 11.2 million, approaching twice the forecast and close to 65 percent more than the top of the range for the number of fish expected to return.

The only thing that explains this is unexpected survival at sea, one of the many vagaries of nature.

It must here be noted that while the sockeye run has exploded and the return of late-run Chinook to the Kenai looks is if it might achieve its minimum goal of 15,000 of the big fish for the first time five years, the picture is not nearly so rosy for salmon at the north end of the Inlet.

The Chinook return to the Deshka River, a tributary to the Alaska Range draining Susitna River watershed, can only be described as a disaster. The Deshka was once famous for it’s Chinook production.

Between 40,000 and almost 60,000 of the big fish Alaskans call king salmon were returning to the river at the start of the new millennium, and as late as 2020, nearly 19,000 came back.

The count this year is less than a tenth of that with the run basically over.

Why is impossible for anyone to know. Salmon survive in a world of death that starts the moment eggs are laid. Some of them go unfertilized. Some of them fail to hatch even though fertilized. Some of the alevins hatched from the eggs die before they wiggle out of the gravel.

The surviving fry start perishing in large numbers as food for other fish and birds. Those that survive to become smolts go to sea where they face the threat of death from even more predators and competition for food with a variety of other fish, including other salmon.

In-stream mortality for all salmon his known to be high. So, too, nearshore mortality. But millions of young salmon still make it out to sea.

Into the unknown

And there they enter what Bill Templin, the director of commercial fisheries research for Fish and Game, has accurately described as a “black box.”

Survival at sea can make a whale of a difference in how many salmon return to Alaska and of what species. And humans know so little about this aspect of the salmon life cycle.

For decades, this lack of knowledge was largely ignored by fishery biologists intent on maximizing salmon production carefully scrutinizing what happened to the fish in freshwater, and then watching them go to sea with the thought that whatever happens in the ocean stays in the ocean.

This laissez faire view of ocean survival only began to change late in the 20th century as first the Japanese, then Alaskans and finally the Russians got into the business of farming the sea with hatchery-spawned fish.

This forced fishery biologists to begin to consider a couple of ecological realities, the first being that all natural, wild systems have a “carrying capacity.” This concept dates back to the early 20th century and Paul Errington, a trapper turned ecologist, and Aldo Leopold, a destined-to-be-famous conservationist. 

They came to recognize that in trying to “manage” wildlife it is possible to have too much of a good thing. The Kaibab Plateau in Arizona became the classic example after then-President Theodore Roosevelt established Grand Canyon National Game Preserve to protect “the finest deer herd in America.”

Hunting was banned. Predators were killed. A deer herd that had once numbered 4,000 reached 100,000. By then, however, the deer were devouring anything and everything edible in a desperate and futile effort to survive.

Sixty-thousand were destined to die over the course of two winters in the 1920s and the population began a long decline as the range on which the deer depended began a slow recovery. Leopold’s history of what happened would forever alter how wildlife biologists and later fisheries biologists thought about managing wild resources.

Later came the theory of “tipping points” that could link small ecosystem changes to big ecosystem consequences.

Some scientists have since suggested that although hatchery salmon represent but a small percentage of the salmon in the Pacific, their addition to the ecosystem – and particularly the addition of hundreds of millions of small but fast growing pink salmon – boosted the total salmon population enough to hit one of these tipping points and cause what has been described as a “trophic cascade” that harmed other whales, seabirds and other species of salmon. 

The theory is much debated, as one would expect, in light of large financial investments in hatcheries, and nothing makes for better science – science being a never-ending search for knowledge fueled by heated debate – than people trying to protect their self-interests.

But nearly all ecologists agree tipping points are a reality, and all ecologists accept that ecosystems have carrying capacities. They also agree that these carrying capacities can be altered by environmental changes, the most obvious of those being weather and/or climate.

A warming Pacific led researchers to in 2018 report the North Pacific held more salmon than at any time in recorded human history. The issue then, and since, was that most of those salmon were smallish pinks, and it was theorized there were taking such a big bite out of the ocean’s food base that they were displacing larger, more valuable salmon and causing huge, year-to-year swings in salmon abundance as odd-year pinks triumphed over even-year pinks in Alaska. 

Survival at sea

Now there is some thought that pinks, or “humpies” as Alaskans tend to call them, have been knocked down a notch by a cooling of ocean waters that once favored them.

That something at sea has obviously changed cannot be ignored because the Kenai isn’t the only river on the West Coast of North American seeing an unexpectedly large return of sockeye.

Returns to the Fraser River in British Columbia, Canada are also unexpectedly high. Once one of the major sockeye producers on the West Coast, the Fraser has struggled through some grim years.

In 2024, only 26 early sockeye successfully spawned in the Stuart River, a major Fraser spawning tributary, the Pacific Salmon Commission reported.

This year?

“As of July 29, 734,400 Early Stuart sockeye had already passed through the lower Fraser River,” the Commission reported, adding that the Stuart hasn’t witnessed a sockeye return of this magnitude since the last century.

“It’s too early to determine exactly why the early Stuart sockeye run is so high this year,” the Commission added. “(But) some scientists are pointing to favourable, cooler ocean conditions due to an unusual three-year La Niña weather pattern between 2020 and 2023.”

There is a lot of speculation that a “cold tongue” of cool water in the middle of the Gulf of Alaska provided better feeding conditions for sockeye, which have been in a generally losing competition with pinks for years now. Warmer waters are generally accepted as more productive for the smaller species.

Along this line, it is interesting to note the pink return to Cook Inlet has gone bust this year. Fish and Game reports it as about a twentieth the size of what it was two years ago, putting this odd-year return somewhere in line with what one might expect to see in an even-numbered year.

But the thing is, odd-year pinks and even-year pinks are distinctly different fish, and it is the odd-year pinks that are dominant in the 49th state.

Only not this year. Pinks do appear to be more plentiful in Prince William Sound, where most of the production comes from commercial fishermen-controlled hatcheries, than in the Inlet, but the Sound harvest is now lagging about 9 million fish behind 2023 and if there isn’t a big harvest in the coming week, it could be lagging way, way behind. 

The preseason forecast was for a harvest of approximately 62 million, about 4 million more than in 2023. About 65 percent of the 2023 harvest had been caught by this date two years ago. About 46 percent of the 2025 forecast has been landed to date.

Pink returns aren’t the only salmon returns lagging in the Sound, either.

Strangely enough, while sockeye numbers exploded in the Inlet to the north and the Copper River just to the south of the Sound witnessed a decent, but short of Kenaiesque return, the sockeye return to the Main Bay Hatchery appears to have crashed.

Prince William Aquaculture Association forecast the hatchery would produce about 1 million sockeye with more than 263,000 earmarked for a “cost-recovery” fishery out front of the facility and the rest expected to be netted by Main Bay and Coghill area commercial fishermen. 

The state is now, however, reporting a harvest of less than a fifth of that, and the fishery is largely over. Fish and Game records say Coghill area fishermen caught about 109,000, and only 59,000 sockeye were swept up in the cost-recovery fishery. That’s not good news for a hatchery that was planning to net $2.2 million worth of sockeye to help cover its operating costs.

What happened will take time to sort out, however, because marine losses of immature salmon can sometimes be huge in nearshore areas and then significant in distant marine waters about which so much remains unknown.

Feeding zones

One of the things that scientists have, however, learned about salmon at sea is that not all of them feed on the same pastures.

After studying archived scales from 17 different stocks of sockeye, researchers from the University of British Columbia, Idaho State University, the Columbia River Inter-Tribal Fish Commission, and Russia’s Kamchatka Research Institute of Fisheries and Oceanography in 2020 reported that they were able to “define salmon locations during their last growing season before returning to their spawning grounds” for 13 that allowed “us to define feeding grounds with confidence.”

Kenai stocks were not included in the study published as open research, but stocks from Kodiak Island and the Chilkoot and Chilkat rivers of the Alaska Panhandle were included, along with salmon from a number of Canadian and Pacific Northwest watersheds.

The results of the study were not what one might expect. Sockeye from the Wenatchee River in Washington State were, for instance, found feeding in the northern Gulf of Alaska just south of Prince William Sound with sockeye from the Okanagon River in British Columbia, Canada, only slight further to the south and near the coast of Southeast Alaska.

Kodiak sockeye, meanwhile, showed up west and well to the south of the more southern stocks with the fish from Kodiak’s Olga Creek concentrated at about the latitude of Seattle and even to the south of there at the very bottom of the Gulf.

“The estimated feeding grounds for stocks with spawning grounds south of the Aleutian Islands were all located in the Northeast Pacific itself, but with little overlap between the stocks,” the researchers wrote. “The Southeast Alaska stocks – Chilkat and Chilkoot – which have spawning grounds 10 kilometers (six miles) away from each other, showed well separated at-sea locations, with Chilkoot being distributed 10 degrees further west than Chilkat.

With those two, neighborly stocks found in the Gulf at near the latitude of British Columbia’s Moresby Island, about 105 miles southwest of Prince Rupert, 10 degrees of longitude would work out to about 425 miles of separation at sea for fish born six miles apart.

And “while Columbia River stocks have the most southerly spawning grounds,” the researchers wrote, “their high seas distributions were located in the northern sector of the northeast Pacific.”

Those young fish went north to Alaska to feed while the Alaska fish went south. And then there were the Rivers Inlet, British Columbia, stock that “was distributed further south than any other stock,” according to the study. 

The Rivers Inlet fish spawn within a few hundred miles of the Okanagan sockeye in British Columbia, but instead of heading north when they hit the sea, they turn south. They were found to be feeding south of the latitude of Portland, Ore., while the Okanagan fish were 700 miles or more to the north near the latitude of Ketchikan.

Though all these feeding grounds appeared to be linked in some ways to sea surface temperatures (SSTs), the researcher noted that the correlation with SSTs “did not always result in the identification of a single well defined foraging region.”

They also conceded there is no way of knowing whether the fish were using long-established feeding areas or whether these areas shift around from year to year. This is hard to assess, they wrote, “given that there are limited data available on high seas stock-specific locations. The question is probably even more essential for stocks with a strong pattern in age distribution (one predominant age-class), resulting in little mixing between cohorts from different brood years.

“Another aspect that should be investigated is whether there are differences in stable isotope values (a stock identifier) between early and late runs for the same stock. While for some stocks, adult salmon return predominantly occurs over a short time window, other stocks have two temporally separated runs identified and it is not currently known if high seas distributions differ between runs.”

All that is really clear at this point is that some sockeye – most notably those of Cook Inlet and the Fraser River – found some pretty green pastures at sea over the course of the last couple of years, but that others weren’t quite so lucky.

Along with the so-so run to the Copper, the sockeye return to Southeast Alaska was below the five-year average, and the Skeena River in northern B.C. is coming in near the five-year average but below forecast.

Overescapement

It’s one big puzzle, but one definitive thing can be said about the Kenai at least:

The fear of “overescapement” – the issue about which commercial fishermen have been haranguing the Board of Fisheries for years in an effort to put more fish in their nets instead of letting sockeye escape into the river – appears largely a strawman.

Commercial interests have long complained that the Kenai’s in-river goal of 1.4 million sockeye is too high, leading to over-crowded spawning beds or in-river competition between young fish that reduces production and thus diminish the return of adult sockeye.

But the annual number of in-river spawners for the parents of the sockeye now swarming back to the Kenai has averaged about 1.9 million, or at least that was the count before the harvests of in-river anglers were tallied.

The in-river, sport fishery now appears capable of cropping off 500,000 to as many 675,000 sockeye when the in-river returns go over 1.4 million. And that would reduce the number of in-river spawners to somewhere very near the top of an optimum escapement goal (OEG) for the river of  750,000 to 1.3 million sockeye.

Given how valuable the in-river fishery has become to the tourism economy of the Kenai, which booms in the summer and dies in the winter, the Alaska Board of Fisheries should probably be talking about raising the in-river goal rather than lowering it to increase the value of the sockeye resource.

Especially since it is well documented that anglers from the Lower 48, Europe and Asia will spend far more money per pound to catch an Alaska sockeye than they would ever even think about paying if they were buying it in a supermarket or from a fish monger.