Hoping to engineer their way out of a salmon crisis, wildlife agencies are manipulating the natural rhythms of the species to an unprecedented degree in hopes of producing more fish.
California has long trucked most of its young hatchery salmon around the Sacramento-San Joaquin Delta to avoid losing them to pumps, poisons and predators. This year, under pressure from fishing groups, it will truck nearly all of them – nearly 17 million salmon smolts.
Federal officials will also truck about 10 percent of the salmon produced at their hatchery near Redding. They haven't trucked fish in more than a decade, and then only as a test. This year, those fish will ride nearly 200 miles to reach San Pablo Bay.
The hatcheries were created to replace spawning habitat eliminated by dams. But this year's changes are prompting important questions about how the hatcheries themselves affect salmon survival.
Fisheries experts have worried that trucking salmon around their home rivers breeds out the instinct that draws fish back to the right place to spawn – an instinct that defines the species.
"Are hatcheries supposed to be helping fish recover over time or just pumping out fish for fishermen to catch?" asked Rachel Barnett-Johnson, a fisheries biologist at the UC Santa Cruz Institute of Marine Sciences. "That's where the struggle lies in managing them now."
Barnett-Johnson published a study this month that found 90 percent of the salmon caught in California's ocean fishery are hatchery-raised, and only 10 percent are wild spawners. She analyzed growth patterns in fish ear bones, which look like tree rings.
Her results, based on a limited sample of fewer than 200 chinook, alarmed fisheries experts. Many assumed, again based on limited information, that wild salmon still make up 50 percent or more of the state's chinook population.
The fall-run chinook population is expected to hit a record low this year for reasons that remain unclear. Commercial and recreational ocean fishing have been closed for the first time in history, and river fishing is likely to be closed next month.
"We felt it was worth using some fish in this trucking experiment to see whether those fish might return at a higher rate than the fish we release on-site," said Alex Pitts, spokeswoman at the U.S. Fish and Wildlife Service, which runs Coleman National Hatchery near Redding. "There was interest in us trucking the entire production, but we didn't think that was a good idea."
The bulk of its 12.6 million chinook will still be released in the Sacramento River this year. That's because they begin life much farther from the ocean, so the risk of "straying" to other rivers as they return to spawn is greater, Pitts said.
Research has shown that trucking hatchery salmon doubles or triples their odds of reaching the ocean by avoiding threats along the way, said Alice Low, an environmental scientist at the state Department of Fish and Game.
To further improve the odds this year, all trucked salmon will be dumped first into net pens suspended in San Pablo Bay, where they will be protected from predators while adjusting to the shock of the Bay's warmer, saltier waters. After two to three hours, the salmon are freed.
"We do want to maximize the benefit to ocean fisheries and other fisheries," said Low. "Personally, I'm convinced that trucking isn't that bad of a practice."
She cited research that found fewer than 10 percent of Feather River hatchery salmon strayed to other locations.
Wild vs. hatchery-bred
Wild salmon are known to survive better in the wild and produce more young. But inbreeding with hatchery fish could weaken these traits. It could also dull the wild salmon's ability to find its way home.
Hatchery fish may also compete for food and habitat with wild salmon and disrupt wild breeding. Their abundance could also hide the possibility that truly wild fall-run chinook have become endangered.
For all these reasons, a joint state-federal commission in 2001 said state hatcheries should "consider" ending the trucking of fall-run salmon.
"The evidence seems pretty strong now that it's basically a hatchery-supported run," said John Williams, an environmental hydrologist at UC Davis who published a comprehensive scientific review of Central Valley salmon and steelhead in 2006. "We're in a deeper hole than we can get out of just by producing a few more hatchery fish."
To gather more information, 25 percent of all salmon produced by the state and federal hatcheries this year – about 8 million fish – are being tagged with a coded wire implanted in their noses. Their adipose fin – between the dorsal fin and tail – is also clipped off as a visual identifier.
The marking occurs in specialized trailers bought by the state last year for $1.1 million each.
On Thursday, 3-inch salmon smolts by the thousands were pumped through pipes and hoses into one of the trailers parked at the state's Nimbus Fish Hatchery on the American River. Cameras and computers sort the salmon by size into rubberized clamps that gently hold each fish just so.
In the blink of an eye, a cutter snips off the adipose fin, and a needle emerges like a dart to stab a coded wire into the salmon's nose. Then the fish is sucked away to await its truck ride.
The missing adipose fin tells a fisherman his catch was born in a hatchery. Anglers are urged to cut off the salmon's head and send it to a lab, where the wire tag is dissected and analyzed.
Each wire tag, only 1.1 millimeters long, is laser-engraved with tiny numbers. When read under a microscope, the numbers tell where the fish was bred. The state aims to recover as many as 50,000 tags per year to learn how many hatchery fish survived and where they ended up.
Eventually, this may also help biologists decide whether hatcheries make sense at all.
"We really need to know if we're doing the right thing," said Low.