Application of decision analysis to evaluate recovery actions for threatened Snake River spring and summer chinook salmon (Oncorhynchus tshawytscha)

There is uncertainty about the importance of various factors in explaining declines of chinook salmon (Oncorhynchus tshawytscha) populations in the Snake River basin of Oregon and Idaho. This uncertainty has prevented implementation of long-term recovery actions for these stocks. We used simulation models and decision analysis to evaluate three management actions for seven index stocks of Snake River spring and summer chinook salmon: (i) continue current operation of the Columbia River hydropower system, (ii) maximize transportation of smolts, and (iii) natural river drawdown (breaching) of four Snake River dams. Decision analysis provided a useful approach for including multiple hypotheses about population responses to environmental and anthropogenic factors, systematically assessing the importance of alternative hypotheses, and identifying risk-averse recovery strategies that meet survival and recovery goals over a wide range of uncertainties. We found that the most influential uncertainties were related to hypothesized causes of estuary and ocean mortality. Current monitoring provides limited information on survival in this life stage; carefully designed management experiments are more likely to generate useful information. Given that these uncertainties exist, drawdown was the most risk-averse action, meeting long-term survival and recovery goals over a wider range of assumptions than the other actions.