Independent scientists say Canadian researcher David Welch correctly analyzed data to show that the Snake River's low returns of Chinook salmon are similar to the rate of returns for many Chinook runs from California to Alaska—including Chinook runs in undammed rivers.
But Welch went too far when he used that data to conclude there is little evidence for dam-induced delayed mortality and for limited potential to improve salmon runs through actions in freshwater, the Independent Scientific Advisory Board said in its June 29 review.
"Both conclusions are speculation and not directly supported (or refuted) by the analyses reported in Welch et al. (2020)," the ISAB wrote.
Welch is an independent scientist at Kintama Research Services in British Columbia, and was lead author of the paper, "A synthesis of the coast-wide decline in survival of West Coast Chinook Salmon," published in the October 2020 journal Fish and Fisheries.
Funded by BPA, the study compared smolt-to-adult return rates (SARs) of 123 Chinook salmon runs from central California to southeast Alaska and found that average rates for hatchery fish in all regions except for coastal Oregon subyearlings are now 1 percent or less (CU No. 1978 ).
Welch went on to question the Columbia Basin's focus on freshwater recovery efforts. "At the broadest level, the major implication of our results is that most of the salmon conservation problem is determined in the ocean by common processes," the study says. "Attempts to improve SARs by addressing region-specific issues such as freshwater habitat degradation or salmon aquaculture in coastal zones are therefore unlikely to be successful. Given the importance of these conclusions, we call for a joint systematic review by major funding agencies to further assess the broader consistency and comparability of SAR data with our findings."
Welch's research was disputed in an analysis by the Fish Passage Center, which said his study compared "apples to oranges" and that the data presented do not support the conclusions that ocean conditions—and not hydroelectric dams—are causing serious declines in North Pacific Chinook runs (CU No. 1983 ).
In its review of both Welch's study and FPC's critique, ISAB said that Welch raises important issues, and agrees that further study is needed to better understand ocean survival. "The Welch et al. (2020) paper adds to other evidence for the need to further investigate SAR values across populations and to continue investigating oceanic and freshwater contributions to low SARs as a critical uncertainty in the basin," the review states.
However, the scientists wrote, Welch's efforts to investigate those questions faced several problems, including unknown data comparability, the inability to isolate the freshwater component of the SARs, a lack of statistical rigor in the analyses, a vagueness in stated conclusions and the fact that the SARs analyzed are primarily from hatchery runs.
"[T]he ISAB supports the idea underlying the Welch et al. (2020) paper of analyzing SAR estimates from many populations, but conclusions beyond general statements of broad patterns and trends require appropriate treatment of data and statistical methods," the independent scientists wrote.
The ISAB found that some of the problem is with the data itself. Over time, there have been changes in tagging and measurement methodology that add "substantial uncertainty" when examining trends over long time spans.
One major issue is that much of the SAR data in the Columbia Basin is from passive integrated transponder tags (PIT-tags), while many other basins track salmon using coded wire tags. ISAB pointed out that coded wires are recovered from adults, but do not offer data on mortality until salmon are harvested. PIT-tags offer information on mortality during migration through hydroelectric projects, but are not recovered through fishing and do not include fishing losses.
In addition, 94 of the 123 populations Welch studied were hatchery runs, which rely less on freshwater habitat than natural-origin Chinook, and the analyses offers limited information about the survival or early stages of wild runs, the ISAB notes.
Those and other uncertainties make Welch's conclusions about the effectiveness of freshwater survival speculation rather than conclusions supported by data, the ISAB said.
The ISAB also defended a goal for salmon runs in the Columbia Basin to reach SARs of 2 to 6 percent, as recommended in the Northwest Power and Conservation Council's Fish and Wildlife Program.
"Welch et al. (2020) question whether the Fish and Wildlife Program's objective of 2-6 percent SARs is attainable and realistic," the ISAB notes. But the study does not consider comparative survival study reports that have found SARs for Chinook frequently exceed 1 percent, and sometimes 2 to 4 percent, and that SARs greater than 2 percent resulted in generational increases in abundance. "The ISAB concluded that 2-4 percent [SAR values] provide a readily measured, first-order objective for restoring stocks," the independent scientists wrote.
The ISAB said that Welch's conclusion that freshwater habitats are not likely to improve salmon returns is based on the current low marine survival. "They do not address the potential for ocean regimes to improve and the value of restored freshwater habitat for recovery of Chinook populations during periods of improved ocean conditions. Ocean conditions are cyclical and climate change is not the only factor that determines trends in ocean productivity," the ISAB wrote.
The board's review offered several suggestions for improving Welch's study.
"Welch et al. (2020) recommend that more research be conducted on the causes of marine survival," it said. "The ISAB concurs that there are very good reasons for better understanding conditions in the marine environment and their potential effects on survival. Information on marine survival is helpful so that decision-makers are aware of the return on investment from hydrosystem operations and habitat actions, and can inform the life cycle modeling that evaluates population responses to habitat improvements."
It also concluded that "the current approach for management rightly focuses on understanding and testing actions that are under management's control and can therefore be manipulated." A better understanding of marine survival will offer better guidance for those actions and help develop realistic expectations from specific management decisions, it said.