Evaluating Uncertainty from Different Sources for Population Viability Analysis Under Climate Change Scenarios – What Does Management Need to Know? a Case Study of Threatened Chinook Salmon

NOAA-Fisheries uses life cycle models to inform management decisions for ESA-listed species by quantifying population viability under different management and climate scenarios. To incorporate climate change into population projections for Snake River spring/summer Chinook salmon, we employ life stage-specific sensitivity to environmental conditions. A crucial component of this analysis is to describe uncertainty related to climate projections, uncertainty in biological sensitivity to these forcing factors, and uncertainty related to other aspects of population modeling. To characterize these various sources of uncertainty, we fit the biological sensitivity parameters in a Bayesian framework to 9 populations, each drawing information from multiple data sources, and explored over 100 climate scenarios for freshwater and a full spectrum of conditions. For all populations, the uncertainty in future ocean projections outweighed all other sources. We identified “tipping points” for each population, at which population viability declined sharply with a relatively small deterioration in ocean survival. The next crucial factor was the relative population sensitivity to temperature versus flow. Uncertainty in climate projections from downscaling methods, GCM or emissions behavior was much less important for management decisions for most populations; however, a subset of populations might either increase or decline depending on future patterns in precipitation.