the film
Economic and dam related articles

Study Documents ESA-Listed Fall Chinook
Migration Changes

by Barry Espenson
Columbia Basin Bulletin - December 10, 2004

Some young Snake River fall chinook salmon may be making the best of a dam-caused situation, effectively shifting the timing of their migration to the ocean and thus creating a second distinct life history type within the same fish stock.

Evidence is building that a significant portion of Snake River fall chinook annually shift from the stock's historic pattern of migrating to the ocean as subyearlings just a few months after being hatched in the spring. A study led by the U.S. Fish and Wildlife Service's William P. Connor shows that 51 percent of the adult hatchery chinook and 41 percent of the wild fish sampled from 1998-2003 were reservoir-type fall chinook -- fish that apparently dallied in Lower Granite and other reservoirs over the summer, fall and winter before migrating as yearlings.

The results of that study were recently accepted for publication in the Transactions publication of the American Fisheries Society. Connor said official publication will likely be next spring. Joining lead author Connor in the research were John G. Sneva of the Washington Department of Fish and Wildlife, Kenneth F. Tiffan of the U.S. Geological Survey, R. Kirk Steinhorst of the University of Idaho and Doug Ross of WDFW.

The realization that there are two distinct life histories -- ocean- and reservoir-type -- could prompt both fish and hydro management changes in attempts to recover the Snake River fall chinook stock. The wild fish are listed as threatened under the Endangered Species Act.

Ironically, summer flow augmentation is probably helping the life history type most barely thought existed -- the late migrating reservoir-type. The cool releases of water from Dworshak Dam flow directly into Lower Granite reservoir during July and August and, more recently, into September.

The augmentation is designed to provide cooler migrating conditions for what was believed to be a mostly ocean-type fall chinook stock. But the practice likely also helps provide late summer thermal refuges for the reservoir-type Snake River fall chinook as they feed and grow, not yet ready for their flight to the ocean.

"Given the lack of thermal refuge in the contemporary spawning areas, mortality of these later migrating fish would be high without summer flow augmentation," according to the research paper. "Therefore we believe that the reservoir-type juvenile life history is a successful response to the large-scale changes in historical habitat that has been enabled or at least enhanced by summer flow augmentation."

The divergence is more than likely a response to the water temperature and climatic conditions than it is an actual genetic adaptation.

"I think that has a lot to do with environmental conditions," Connor said. He said he believed that the number of fall chinook that choose to leave as subyearlings or linger to grow in reservoirs likely varies depending on the conditions where they hatch and rear that particular year. But it does appear that certain locations produce more ocean-type and others more reservoir-type reservoir conditions.

Prior to the 1960s, the primary Snake River fall chinook spawning area was in a river reach between Swan Falls Dam and Marsing just across the border in southwest Idaho. That spawning option was blocked off when the first of three dams in Idaho Power's Hells Canyon complex was completed.

According to the Connor paper, the temperature and productivity of that stretch promoted an ocean-type juvenile fall chinook life history. The warm, again relatively, conditions speed incubation to give the fish an early start and enhance productivity so the rearing fish grow more quickly to the point they are ready to migrate.

In the pre-dam era on the Lower Snake, the young fall chinook smolts had a relatively clear and rapid course toward the ocean once they were ready and water temperatures began to reach uncomfortable levels. Sampling done in the 1950s in the areas where Lower Granite and Little Goose dams now stand showed that the entire chinook smolt run had passed by the end of June, before river flows had drop to summer levels and temperatures warm appreciably. When water temperatures increase past a certain level, predation on the young salmon increases and their growth slows.

Sampling done during the 1990s showed that less than half of the juvenile fall chinook from the lower Snake and Clearwater rivers had filed past Lower Granite by the end of June. With upstream passage blocked in Hells Canyon, spawning shifted principally to the reach between Hells Canyon dam down to the upper end of Lower Granite Reservoir and in lower reaches of the Clearwater, Imnaha, Grande Ronde and Salmon rivers. The lower Snake reach conditions most closely resemble that historic Marsing spawning reach.

"With the exception of the Snake River upper reach, these 'contemporary' spawning areas are cooler during incubation and less productive during rearing than the historical spawning area and they produce young fall chinook salmon that progress through juvenile life stages later than was observed in the historical spawning area," according to the research paper.

The data collected during the 1990s indicates that on average only 2 percent of the wild fall chinook salmon juveniles from the Snake River reach above the Salmon river exhibited the reservoir-type history as compared to 53 percent of the juveniles from the much cooler lower Clearwater.

"Therefore, it is likely that the majority of the spawners that have been ocean-type juveniles were destined for the warmer contemporary spawning areas, whereas the majority of the spawners that had been reservoir-type juveniles were destined for the cooler contemporary spawning areas," according to the research paper.

It is believed the reservoir-type Snake River fall chinook are making a large contribution to the spawner return in large part because they leave the freshwater at a larger size and thus survive better during their saltwater entry and maturation. Wild and hatchery juveniles collected in the spring of 1997 as yearlings were nearly twice the length of broodmates captured the previous summer as subyearlings beginning their outmigration.

", we speculate that the large size and early spring migration timing of the reservoir-type fall chinook salmon was advantageous to survival to saltwater as well as smolt-to-adult return rates," according to the paper. That belief is based on data showing that hatchery fall chinook released as yearlings survive better during their outmigration and have shown smolt-to-adult return rates sometimes 10 times higher than subyearling releases.

The paper notes that "the existence of reservoir-type fall chinook salmon juveniles does not support the decision to dewater the six juvenile fish bypass systems in the Snake and Columbia river during the winter." Those devices are shut down with the belief that few if any juvenile fish are moving through the system, leaving turbines as the only passage option. The paper cites one study for stream-type spring chinook that shows 8 to 14 percent blade-strike mortality for turbine passage while another study shows mortality of 1 to 2 percent for the bypass systems.

"The need for this expensive recovery measure should be evaluated by identifying the primary reservoirs used by reservoir-type fall chinook salmon, the extent of this use, the passage timing at the dams that form the reservoirs and the actual rates of mortality attributable to turbine blade strike," the paper advises.

Evidence of significant numbers of returning spawners from reservoir-type juveniles also has important implications regarding studies examining the efficacy of transporting Snake River fall chinook via barges to the Columbia estuary. Those studies involve monitoring for the fishes' in-river progress downstream and comparing smolt-to-adult return rates with those of transported fish.

"A strong assumption inherent in this study design is that the in-river treatment group will be composed of ocean-type fall chinook salmon juveniles. We showed that this assumption will be violated when PIT-tagged fall chinook salmon subyearlings select the reservoir-type life history, move seaward during the winter when the juvenile fish bypass system are not operated, or move seaward during the spring when high levels of spill facilitate dam passage without detection. Those study designs need to be adjusted in include scale pattern analysis to identify each fishes' life history type."

Barry Espenson
Study Documents ESA-Listed Fall Chinook Migration Changes
Columbia Basin Bulletin, December 10, 2004

See what you can learn

learn more on topics covered in the film
see the video
read the script
learn the songs
discussion forum
salmon animation