Last week we talked about the importance of spill for out-migrating kelts and smolts in the Columbia Basin. Increasing spill is only one method employed to enhance downstream survival of smolts. As early as the mid-1950s smolts were loaded onto barges and moved downstream past the dams through the lock system. While fisheries managers experimented with this tactic early on it did not become common practice until the mid 1970s, when steelhead populations in the Snake and Columbia Rivers dipped to historically low levels.
A scientific study looked at the survival rates of fish that were released from barges at two different locations downstream of Bonneville Dam from 2006 to 2008. The first site was just downstream of Bonneville Dam at Skamania Landing, while the second was some six miles upstream of the river mouth near Astoria.
Smolt barges at Lower Granite Dam. Photo: Tony Grover, NW Power and Conservation Council.
The study found that releasing fish near Astoria significantly reduced fish and avian predation compared to the upstream release near Bonneville. Predators were able to consume 17-26% of the smolts released at Bonneville compared to only 2-5% at Astoria. Interestingly, despite the higher predation of smolts released at Bonneville the smolt-to-adult survival rate (commonly abbreviated as SAR) was not significantly different between release locations across the study years.
To figure out why the lower predation rates were not translating to more steelhead making it back to their home rivers the researchers looked at what is known as conversion rates. As adult steelhead migrate upstream they are counted at all of the Columbia dams — the decrease in numbers at each dam due to mortality or movement into tributaries is known as the conversion rate.
Researchers below Lower Monumental Dam. Photo credit: NOAA
Why are conversion rates important? Because they can help researchers understand where fish go missing. For example, scientists implanted PIT tags in barged smolts. Tracking of those fish past dams indicated that 20-22% fewer fish released at Astoria returned to Lower Granite Dam on the Snake compared to fish released at Bonneville. So even though barging to Astoria improved smolt survival, return rates of adults depended on where you dropped the kids off.
Further analysis found that both wild and hatchery steelhead released at Astoria were straying to the John Day and Deschutes Rivers at rates more than 50% greater than for fish released at Bonneville. Steelhead often find refuge in the cold water of tributaries during warm periods, but these values were for permanent strays that never made it back to cross Lower Granite Dam where they had been trapped as smolts.
While the exact mechanism is not known it is clear that barging steelhead smolts all the way to the estuary was causing fish to get lost on their trip home. So despite the advantage of bypassing the large colonies of terns and cormorants in the lower Columbia it did not give the fish an advantage in numbers of returning adults. Furthermore, this practice put many out-of-basin hatchery fish onto the spawning grounds in some of the best wild steelhead rivers in the Columbia Basin (such as the John Day).
This is a classic example of the importance of funding both short and long term monitoring work. Not infrequently, there are unintended consequences of well-intentioned efforts. The better we understand the biology of wild steelhead and how our management practices influence them, the better we will be able to manage this complex species. And in the long run that will translate to more stable populations and better fishing opportunity.