the film forum library tutorial contact DONATE

Latest NMFS Analysis Backs Up Late Spring Barging Strategy by Bill Rudolph NW Fishletter, December 14, 2011

Young sockeye migrating out of Redfish Lake were reported to do well this year as well, with a
72% survival rate for smolts to Lower Granite Dam, compared to the long-term average of 47.3%.

A recent analysis of smolt-to-adult returns produced by NMFS scientists suggests that transporting juvenile salmon from lower Snake River dams for two weeks in May would produce more adult returns.

Federal District Judge James Redden was never convinced the practice was effective, and since 2006 had called for more spill at dams, reducing the number of fish routed to barges.

The latest federal analysis was presented at the Corps of Engineers' week-long annual research review in Walla Walla in late November. It compared smolt-to-adult returns between barged and inriver migrating salmonids both before and after the court-ordered spill regime was implemented.

The federal scientists found that the geometric mean of the model-averaged ratio [T:M] of transport SARs [smolt-to-adult return rates] to SARs for inriver migrants were 1 or above from April 20 to the end of May between 1998 and 2005. For 2006-2008--when the new spill regime was in place--the ratios were 1 or above for all the species examined after April 30 through the end of May.

In simple terms, the results suggest that more fish would have returned as adults if they were barged later in the season. This outcome has held up across all species, both wild and hatchery spring/summer chinook, and wild and hatchery steelhead.

For instance, the analysis showed that wild Snake chinook transported in 2008 returned at nearly twice the rate as inriver migrants in May, when barging began that year. In 2006, when barging began earlier, the results showed that inriver migrants did better until May 1.

In 2008, the feds' analysis showed that some transport SARs for wild steelhead reached the 5-percent range, while inriver SARs for the same period approached 2 percent.

In 2007, near the end of June, barged SARs for wild steelhead released above Lower Granite Dam were in the 4-percent range, 10 times or more the SARs for inriver fish.

Another NMFS analysis at the research review of preliminary survival estimates of juvenile salmonids through the hydro system found that only 35 percent of PIT-tagged wild chinook from the Snake were barged in 2011, along with 41 percent of the hatchery chinook, 36 percent of the wild steelhead and 38 percent of hatchery steelhead. Before the new spill regime was in place, it was common to barge 60 to 90 percent of the spring chinook, and up to 90 percent or more of the steelhead.

This year's high spring flows reduced PIT-tag detections, thanks to a powerhouse outage at Little Goose Dam and pulled turbine screens at Bonneville Dam stemming from trash problems. Adverse river conditions also reduced detections by the estuary trawl group in the lower river.

Despite the huge flows this spring, inriver survival of spring chinook--both hatchery and wild--from the Snake River trap above Lower Granite Dam to below Bonneville Dam was only about average. According to the preliminary results, chinook survival was estimated at 48.3 percent, compared to the 1995-2011 average of 49.3 percent.

But steelhead was another matter. Juvenile steelhead (hatchery and wild) were estimated to have a 59.2-percent survival rate to below Bonneville Dam, compared to the long-term average of only 41.8 percent.

Young sockeye migrating out of Redfish Lake were reported to do well this year as well, with a 72-percent survival rate for smolts to Lower Granite Dam, compared to the long-term average of 47.3 percent.

Parr released into the lake the previous fall had only a 12.6-percent survival rate.

The NMFS analysis said the steelhead survival rate over the past three years has been the highest in the agency's time series and is likely due to relatively high spill rates at dams, the increased migration rate, and the addition of more surface passage structures at the dams. Also, with more inriver migrants, overall predation on steelhead may actually be reduced.

But the analysis also pointed out that the increased survival for juvenile steelhead did not necessarily lead to an increase in smolt-to-adult return rates.

Erin Richisky, from Vancouver, B.C.-based Kintama Research, reported on 2011 ocean survival results for acoustic-tagged spring chinook. For the first time in the past six years, transported fish fared worse than inriver migrants between Astoria and the north end of Vancouver Island. Only 1 percent of the transported smolts were estimated to make it up the coast, while about 4 percent of the inriver migrants were estimated to reach that receiver array. All groups showed similar minimum survivals from below Bonneville Dam to the mouth of the river, over 80 percent.

From Astoria to Sand Island, the inriver migrants showed no losses, but only 93 percent of the transported fish (plus or minus 6 percent) made it that far.

From Sand Island to Willapa Bay, minimum survival for inriver migrant Snake chinook was 23 percent, and 27 percent for Columbia inriver migrants, but only 14 percent for the Snake transported fish.

From Willapa to the north end of Vancouver Island, survival for the inriver migrants ranged from 22 percent to 25 percent, and 9 percent for the barged fish.

Canadian researcher David Welch said they will look into the question of why the transported fish did so poorly this year as part of a report on ocean projects funded by BPA.

The survival difference between barged and inriver migrants has been one of the most complicated issues that researchers have tried to unravel. Once PIT tagging became an acceptable practice, it still took years for federal scientists to realize the comparative survival between the two strategies varied not only from year to year, but from week to week in the same migration season.

A presentation by University of Washington researchers Jim Anderson, Jennifer Gosselin and PNNL's Kenneth Ham pointed out directions for future research.

The arrival time of fish in the hydro system and travel time through it is important, they pointed out, and the data is pretty clear about that.

They said some of the most important factors relating to the "differential delayed mortality [D]" also carry high uncertainty, such as the size of fish at migration. This factor may play a big role in their future survival because larger ones survive better. (Welch's expensive, acoustic tags need larger fish to carry them--130 mm or above--but half the migration is made up of smolts smaller than that.)

Dam operations are also important, said Anderson, along with conditions in the estuary and predators there. Ocean conditions also play a large role in determining D.

Of moderate importance to D, the researchers said, are fish physiology, disease, barging conditions, and conditions above the hydro system. But current assessments of these factors are based on data and conclusions that are highly uncertain.

On the other hand, they said, adult straying of steelhead was also moderately important, and the data was pretty clear about that.

Of lesser importance, they said, barging conditions, lower river conditions and predators, and adult straying of chinook all carried low uncertainty in data and conclusions. Higher uncertainty in data and conclusions were found in survival estimation and tagging effects.