When Stephanie Schmidt became Alaska’s Yukon River fishery research biologist in January 2012, she knew that all was not well along the sinuous length of the famed river. Chinook salmon numbers had been dwindling since 1998, and as a result, commercial harvest of the fish — also called king salmon for their immense size and sumptuous meat — was frequently halted. That didn’t help: The 2010 run was the second-worst in recorded history. 2011 was only a few fish better.

Faltering returns also hurt the Yukon’s subsistence fishermen, who catch chinook with nets and fish wheels to feed their families through the long Subarctic winter. But past hardships paled in comparison with 2014, their most difficult season yet. Earlier this spring, Schmidt and her colleagues at the Alaska Department of Fish and Game considered the dismal projections — an estimated 64,000 to 121,000 fish, pitiful compared to historic runs, which averaged 300,000 as recently as the mid-’90s. Then they made the painful decision to close the Yukon chinook fishery to everyone for the entire summer.

“I don’t want to say it was a no-brainer, because it was very difficult to know that people weren’t going to get the fish they depend on,” says Schmidt, a member of Wisconsin’s Brothertown Indian Nation who’s passionate about helping subsistence fishermen. “But when we crunched the numbers, it was clear: If we want to have any hope of meeting our objectives for escapement” — the number of fish that must survive to perpetuate the stock — “we couldn’t have any harvest.”

What makes the closure so frustrating is that researchers don’t know why the region’s chinook have vanished, or how to bring them back. In the Lower 48, such declines often have obvious causes: dams that impede migration, say, or the destruction of spawning grounds. On the relatively pristine and almost entirely undammed Yukon, though — and on other Alaskan rivers where chinook have nose-dived, like the Kuskokwim, south of the Yukon — the usual suspects don’t apply.

The king salmon’s disappearance is an ecological mystery as cryptic as it is alarming, and for the villages that dot the banks of the 2,200-mile-long Yukon — remote Alaskan and Canadian towns in which cash and employment are scarce and a gallon of gas can cost more than $6 — the salmon’s absence hasn’t just eroded fishing culture, it’s strained local health and prosperity. As one Alaskan put it: “I didn’t realize how much I depended on (the salmon) until I didn’t have it.”

Seven years ago, it seemed clear what was plaguing salmon: Alaskan pollock, the ubiquitous flakey white species featured in fish sandwiches at fast-food joints like McDonald’s. Although the Marine Stewardship Council considers the billion-dollar industry —- America’s largest fishery —- sustainable, it’s impossible to haul 1.1 million metric tons of fish from the Bering Sea each year without landing a few others that you don’t mean to kill. In 2007, a decade into the salmon crisis, pollock boats scooped up a whopping 121,000 chinook as bycatch in their trawls; up to two-thirds would have returned to Western Alaska to spawn. Yukon salmon fishermen who had seen their own catches slashed were furious. “(The pollock trawlers’) difficulty is big, big money on one side, and small people on the other,” railed one Canadian fisherman.

To its credit, and with a hearty push from regulators at the North Pacific Fishery Management Council, the pollock industry responded with a suite of reforms. Many vessels voluntarily avoided bycatch hotspots and began using nets that allow strong-swimming chinook to escape while leaving weaker pollock ensnared. The industry developed incentives to reward the best performers and punish the worst offenders. And to ensure honest accounting, the fleet expanded the presence of fisheries observers, on-deck biologists funded by industry but managed by the federal government.

In subsequent seasons, bycatch indeed plummeted: Last year, pollock boats took just over 13,000 chinook. The fleet doesn’t necessarily deserve full credit for that reduction — fewer salmon in trawls might simply mean there are fewer salmon in the sea. Still, Stephanie Madsen, executive director of the At-Sea Processors Association, which represents a large portion of the pollock fishery, is sure the fleet’s actions have helped. “Nobody likes restrictions, but these guys can adapt,” Madsen says. “We’re confident that behavior’s changed, and we think those changes have resulted in lower bycatch.”

Less bycatch, however, hasn’t produced more Yukon chinook — casting doubts about whether the pollock fishery was ever really the problem. Salmon runs remain around 200,000 fish below historic levels, a discrepancy much greater than bycatch can explain. “The numbers just don’t add up,” says Daniel Schindler, a professor of aquatic and fisheries sciences at the University of Washington.

Still, down-and-out salmon fishermen continue to blame pollock. One group representing Native Alaskan villages threatened to boycott McDonald’s, and salmon stakeholders have repeatedly called for cuts to the pollock fishery’s allowable take of chinook. “Bycatch certainly isn’t the smoking gun,” says Diana Stram, a fishery analyst with the regional management council. “But until chinook turn around in Western Alaska, it’s going to be a concern anytime one is caught as bycatch.”

If bycatch can’t explain the salmon slump, what can? So far, scientists are flummoxed. Last year, a team of experts from universities and government agencies released a chinook research action plan, an 87-page document whose hypotheses only underscore the uncertainty about the cause of the collapse. One theory is that fishermen have caught too many large kings, causing the population to evolve in the direction of smaller, less fecund fish, while another is that fishermen haven’t caught enough chinook, resulting in excess competition among juveniles for food and habitat.

And several factors could be acting together. Schmidt uses the analogy of rivets in an airplane: Take out one or two and the plane might still fly, but the more rivets you remove, the greater the likelihood of a crash. “But there’s so little data available to analyze the different rivets,” she points out. “The Yukon is a huge system, and getting to tributaries where salmon are spawning and rearing requires extensive helicopter travel. We’re not on a road system where we can just drive up and put a boat in the water.”

The main rivet currently under investigation is the chinook’s early life, when young salmon weighing no more than a deck of cards migrate downriver and into the Bering Sea. According to Kathrine Howard, the state’s Arctic-Yukon-Kuskokwim regional research biologist, juvenile salmon require favorable conditions, including adequate food and proper water temperatures, when they reach the ocean. “If there’s a mismatch between when they’re migrating out and when the right conditions occur, that can mess everything up,” she explains.

Anthropogenic climate change, which has unsettled the timing of wildlife movements around the globe, is often blamed for such mismatches. Young Yukon salmon head to sea when river ice melts; if the ice breaks up too early, it could cause juvenile fish to migrate before the ocean has produced sufficient food to sustain them. But while Howard acknowledges global warming’s disruptive potential, she’s quick to add that there’s still not enough data to demonstrate a conclusive link.

For his part, Schindler, who helped spearhead the action plan, believes that natural fluctuations in North Pacific water temperatures may be responsible for the decline. As evidence, he points to the Columbia River Basin, where salmon runs have spiked in recent years, largely due to beneficial ocean conditions that have produced more food. When climate cycles boost salmon in the Lower 48, explains Schindler, they naturally quell Alaskan runs, and vice versa. The most important thing salmon managers can do, then, might also be the hardest: Be patient.

Of course, it’s hard to remain patient when your livelihood depends on harvesting salmon. In June 2012, a group of 23 Yup’ik gill-netters were arrested on the Kuskokwim, near the town of Bethel, 400 miles west of Anchorage, for defying a fishing ban. Although the judge sympathized with them, he still fined each violator $250. “It’s heartbreaking to hear stories about how fishermen are losing their culture, how they aren’t able to feed their families,” Schmidt says.

In Canada, the situation is even more dire. Under the terms of an international treaty, a minimum of 42,500 chinook are supposed to escape across the border each year, providing Canadian fishermen a chance to harvest and allowing salmon to reach their upriver spawning grounds. Recent runs, however, have failed to meet that guarantee. In 2012, Alaskan subsistence fishermen took around 26,000 chinook from the Yukon; Canadians were only able to catch 2,000.

If there’s a silver lining, it’s that the crisis has forced Native fishermen to become more involved in salmon management. “State and federal government has a seat at the table, but tribal governments do not,” says Natasha Singh, general counsel for the Tanana Chiefs Conference, a nonprofit that provides services for Interior Alaska Natives. In response, the Chiefs Conference and other Native groups have called for the formation of a Yukon River Intertribal Fish Commission, a body that would research and co-manage salmon alongside agencies. Even Native Alaskans who support the current closure want a voice in future salmon decisions. They hope to gain one, perhaps as soon as next year; as Singh, herself a longtime Yukon fisherman, says: “We need this immediately.”

This year, fortunately, the most alarming forecasts haven’t been realized: As of last week, around 130,000 kings had entered the river. That’s poor compared to past years, but enough, perhaps, to meet escapement goals; scientists should know for certain later this month. The better-than-expected run comes as a relief to Schmidt, but she worries that the mystery is no closer to being solved. “There’s been some great efforts to increase money and manpower for research projects,” she says. “But if we’re going to understand the broad picture on the Yukon, we’re going to need more.”