Drawing the Line in Salmon Genetics

At the Yakama Tribe’s innovative experimental hatchery, WSG researchers show what a difference a wild broodstock can make.

On the east side of Cle Elum, Washington, between Interstate 90 and the Yakima River, sits a tribal salmon-rearing facility that encapsulates the past and present of hatchery management in the Pacific Northwest. Now, thanks both to foresight that was controversial in its time and to Washington Sea Grant-sponsored research that looks beyond its time, this unique salmon hatchery may also chart a future for the region’s prize fish.

The Cle Elum Supplementation and Research Facility, operated by the Yakama Nation’s Yakama Klickitat Fisheries Program, was approved in 1982 under the Northwest Power Act. Its original mission was a common one for hatcheries, to produce more salmon (spring Chinook, in this case) that would swim out to sea, mature and return for fishermen to catch. The original plan was to use returning hatchery-bred fish as broodstock for the next generation, maintaining a population genetically segregated from wild fish.

By the time the facility opened in 1997, however, fisheries thinking had diverged from this approach. Scientists feared that genetically segregated populations would lose diversity and fail to adapt, as wild fish must, to changing conditions. “Worse than that,” recalls Yakama Fisheries Research Manager Dave Fast, “when one hatchery was short of brood eggs it would just transfer them from another” — removing the stock further yet from its wild roots.

The Cle Elum facility instead adopted an “integrated” approach — using wild broodstock to produce hatchery fish matching the river’s natural population as closely as possible. As a research facility, it investigated differences between the natural environment and hatcheries as well as genetic changes in the two stocks. But some scientists thought they were missing an opportunity. “A perceived weakness of the integrated program was that we were inadequately monitoring genetic change,” says Craig Busack, a NOAA Fisheries Senior Fish Biologist who was assigned to the Yakama Klickitat project as a Washington Department of Fish and Wildlife (WDFW) employee in the 1990s. A scientific review panel suggested that integrated and segregated lines be reared under identical conditions and the genetic outcomes be compared.

That idea met initial resistance from hatchery operators. Raising segregated lines was and remains controversial; critics warned that the “domesticated” fish thus produced could breed with and weaken wild stocks. “It was assumed that the integrated approach was the better way to go,” recalls Busack.

But raising and comparing parallel stocks could test that assumption and quantify the benefits of using wild broodstock. The Cle Elum hatchery’s multiple-raceway design and location just above Roza Dam, where workers could count and separate the fish, made it uniquely suited to conduct such experiments. Its operators agreed to undertake the expanded mission.

Kerry Naish, a WSG-supported evolutionary geneticist at University of Washington’s School of Aquatic and Fishery Sciences (SAFS), had just the genetic tools and just the graduate researcher,

Charlie Waters, UW graduate researcher.

Charlie Waters, to carry out the experiment. Nearly a decade earlier, Ken Warheit — supervisor of WDFW’s Fish Health and Genetics Programs and collaborator on the Cle Elum project — had more limited tools with which to compare the two broodstocks. In the intervening years, he noted that genetic technology had progressed “even further than we imagined.” Then Naish and her SAFS colleague Jim Seeb mapped the Chinook salmon genome,1 and another of Naish’s graduate students, Marine Brieuc, identified markers for an important fitness trait.2 Waters used these tools to track genetic divergence over four generations in integrated and segregated lines bred from the same wild population at Cle Elum.3

The results were striking — divergence from the source stock was minimal in the integrated line. But it was rapid and pronounced in the segregated line, largely due to genetic drift (random loss of genetic diversity). Still supported by WSG, Waters is now working to determine which fitness traits are most susceptible to genetic change, and to what degree using natural broodstock can prevent it. He notes that teasing out such comparisons would be extremely difficult or impossible using lines drawn from different stocks in different hatcheries. “You’d have so many confounding factors — different histories, population sizes, environments,” he explains. “Cle Elum is a unique opportunity because the two lines were derived from the same wild population.”

Some confounding factors intrude even at Cle Elum, despite the common base stock and shared environment. Because the hatchery is also dedicated to producing fit fish for harvest, its integrated line is much larger than its segregated line, making the latter more susceptible to genetic drift. Still, says Waters, after accounting for that difference, the integrated line shows more diversity than could be explained by its greater size.

What does all this mean to salmon production? Despite the genetic differences, “We’ve seen no significant difference in survival between the integrated and segregated stocks,” says WDFW’s Warheit. “I would expect it will be a longer time before we see differences. We really have no idea how long it will take. This is all set up as a big experiment.”

1 Siple MC (2013) A code of many colors: the salmon genome revealed. Sea Star Autumn:4-5.

2 Brieuc MSO, Ono K, Drinan DP, Naish KA (2015) Integration of random forest with population based outlier analyses provides insight on the genomic basis and evolution of run timing in Chinook salmon (Oncorhynchus tshawytscha). Molecular Ecology 24:2729-2746.

3 Waters CD, Hard JJ, Brieuc MSO et al. (2015) Effectiveness of managed gene flow in reducing genetic divergence associated with captive breeding. Evolutionary Applications 8:956–971.