Rick Zollman stands at the edge of a rectangular, concrete pool, peering into the water below. Tens of thousands of juvenile chinook salmon rush toward him, their speckled backs and silver bellies glistening in the afternoon sun.

Zollman waves and smiles at the fish as they leap from the water to greet him, conditioned to expect food when they sense his presence.

Each of the 18 pools — or raceways — at northeast Oregon’s Lookingglass Hatchery holds roughly 65,000 juvenile chinook, totaling nearly 1.5 million fish.

The fish were hatched here in January from parents collected in one of five regional rivers, then transferred to the raceways in spring. They’ll remain here for a year, growing and maturing until ready for release into the wild.

Shaded by towering lodgepole and Ponderosa pines, Lookingglass Hatchery sits along Lookingglass Creek, not far from the Idaho border in the historic homeland of the Nez Perce Tribe.

The Nez Perce have exclusive fishing rights to Lookingglass Creek, one of the Tribe’s traditional fishing spots. For centuries, Nez Perce families have gathered here to harvest salmon returning from the Pacific.

The Tribe uses the hatchery to restore the area’s natural population of wild chinook, in the hopes that they may one day reach levels that support consistent harvest.

In a controversial practice known as “supplementation,” Lookingglass managers take mature wild fish from the area’s streams and spawn them at the hatchery.

The goal is to ensure that the fish released from the hatchery are from the same genetic lineage as the wild stock, so they can return to spawn naturally.

Many scientists and conservationists have pointed to hatcheries as a contributing factor to the demise of wild salmon stocks in the Pacific Northwest.

But for tribes like the Nez Perce, whose culture is bound to salmon, hatcheries may be all that’s preventing their way of life from disappearing.

To supply fishing grounds while minimizing the effects of hatcheries on endangered wild salmon, tribal-operated hatcheries are employing experimental methods like supplementation to restore wild fish populations in the rivers.

“With hatcheries, they’re not a solution, they’re a tool,” says Zollman, who works for Nez Perce fisheries but is not a tribal member himself. “The idea is that we still have fish spawning so our grandkids can go watch them, and still be able to catch fish and have them on the table.”

At Lookingglass, the spring chinook conservation program operates for rivers in the Grande Ronde and Imnaha river systems.

Lookingglass is one of five hatcheries among the 33 operated by the Oregon Department of Fish and Wildlife that has a conservation program. Like Lookingglass, the others — Cascade, Irrigon, Umatilla and Wallowa hatcheries — each have tribal co-management.

The Lookingglass program uses supplementation — removing wild fish from rivers and integrating them into hatchery broodstock — to produce the next generation of salmon.

Chinook spawned at the hatchery eventually return to their natal streams as adults to spawn naturally, producing offspring that are genetically and behaviorally indistinguishable from wild-origin fish.

The National Oceanic and Atmospheric Administration regulates hatcheries that take endangered salmon populations — such as Columbia River spring chinook — for broodstock.

To prevent an overabundance of hatchery-reared fish on the spawning grounds, which generally have greater return numbers than wild-origin fish, Lookingglass managers employ an elaborate system of weirs (fish traps) to maintain a healthy ecosystem balance.

In what’s referred to as the “sliding scale” method, Lookingglass managers use the weirs to select how many of each type of fish — hatchery-reared or wild — reach the spawning grounds. The number of a given year’s wild returns determines the number of hatchery fish allowed to reach the spawning grounds.

Lookingglass managers also use the weirs to adjust the number of wild-origin fish taken for broodstock based on that year’s wild returns. During years when wild returns are low, more hatchery-origin fish are collected for broodstock, so as to not interrupt the wild chinook population’s recovery.

Once collected, broodstock are spawned at the hatchery, and their offspring are incubated, then transferred to early-rearing tanks.

Once the young fish reach a few centimeters in length, they’re segregated based on the rivers their parents originated from, which prevents biological connections from being compromised.

After about a year of maturing in the raceways, the fish are trucked to acclimation sites (small pens near the spawning grounds in their home rivers) where they spend their final four to six weeks before release.

It’s here that fish internalize the rivers’ unique chemical and magnetic cues that will one day guide them home.

When Lookingglass began its conservation program in 1997, each of the area’s watersheds had only a few dozen fish returning to spawn.

At Lookingglass Creek, those numbers were in the single digits.

Now, hundreds of fish return to Lookingglass Creek each year — enough to support limited sport and tribal fisheries.

While year-to-year numbers fluctuate wildly, average annual returns in the nearby Lostine River now top more than 1,000, according to data provided by Zollman.

Salmon had a particularly prosperous year in 2010, when returns to the Lostine were close to 5,000. Half of 2022’s returns to the Lostine, which have yet to be fully counted, were wild-origin fish.

Supplementation represents a shift in hatchery management that began around the turn of the century.

But some scientists say these programs are risky. Studies have shown that interbreeding hatchery fish with natural-origin fish can negatively affect wild populations.

Salmon are biologically linked to the rivers they come from. Raising juvenile fish in an artificial habitat can make those fish less suited to natural environments, decreasing the chances that they return home to spawn.

This lack of biological fitness carries on to the hatchery fishes’ offspring, which can genetically weaken the local wild populations when the two interbreed, according to a recent report by the Washington Department of Fish and Wildlife.

“Hatchery fish are domesticated, and that difference is actually programmed into the genetics of the fish themselves,” says Jamie Glasgow, director of science and research at the Wild Fish Conservancy, a Washington-based nonprofit conservation organization. “If wild fish interact and spawn with hatchery fish, the next generation of offspring from that hatchery and wild pairing is much less likely to survive in the wild.”

Since the region’s first hatcheries were built in the late 1800s, the majority of hatchery programs have operated under an agricultural model of fish production.

This approach relies on the sheer volume of fish produced to sustain runs and support fisheries without consideration for habitat restoration or the fishes’ genetic fitness, says Jack Stanford, a retired professor and fisheries ecologist at the University of Montana.

“There’s this mantra out there that you can replace lost catch because of the demise of wild fish with hatcheries,” he says. “And it does not work.”

The net result is the entire Pacific Northwest salmon fishery being reliant on a system that may be contributing to the decline of the very fish it’s intended to save.

“It’s like we’re trying to save this patient, but we’re standing on their throat while we’re doing it,” says Glasgow.

Nevertheless, some members of the Columbia River Plateau Tribes view hatcheries as essential to keeping ancient traditions alive.

They see supplementation as necessary to not only saving the fish from extinction, but to keeping salmon in the rivers and streams in tribal homelands that once served as sacred fishing grounds.

The lives of the Indigenous people who inhabit the plateaus and valleys of the Columbia River Basin once completely revolved around salmon. The seasonal returns of salmon to natal streams are integral to their cultures.

“We’re a salmon people,” says Joe Oatman, a member of the Nez Perce Tribe and director of the Harvest Division of its fisheries program. “Our whole identity and our whole view of the world revolve around salmon. And to be salmon people, we need to have salmon in the rivers.”

Lookingglass is considered to be among Oregon’s more successful conservation hatchery programs.

While numbers are nowhere near enough to sustain a fishery robust enough to supply the Nez Perce year-round, the program has prevented the region’s spring chinook salmon population from disappearing entirely.

Zollman says that for the foreseeable future it’s unlikely numbers will reach a point where the hatchery program is no longer needed, given the many factors contributing to the fishes’ mortality that are beyond his control.

But in terms of giving salmon a fighting chance at survival, Zollman is confident the program is working.

For Oatman, the fact that there are still fish in these rivers at all is a sign of a successful supplementation program.

“It’s more than just catching a few fish to bring home,” says Oatman. “It’s about finding a place where we can pass on these traditions that have been there for countless generations.”

Cole Sinanian is a journalist based in Eugene. Columbia Insight, based in Hood River, is nonprofit news site focused on environmental issues of the Columbia River Basin.