In collaboration with the University of Missouri and Iowa State University, this project advanced efforts to understand and accommodate uncertainty by applying to Missouri River sturgeon population dynamics the tools of multi-scale climate models and hierarchical Bayesian modeling frameworks, linking models for system components together by formal rules of probability. While a complete climate prediction may be intractable at this time -- for instance, the climate projections may not incorporate land use changes and solar fluctuations into the boundary conditions -- we proposed a framework to quantify known uncertainty that is also flexible enough to adapt to advances in climate predictions. A key advantage of the hierarchical approach is that it incorporates various sources of observations and includes established scientific knowledge, and uncertainties associated with each. This work is critical for monitoring effects of climate change since lakes and streams have been identified as sentinels of environmental change. The proposed hierarchical modeling approach should help to account for these uncertainties, in particular variability of relevant climate conditions across temporal and spatial scales, so forecasts of community or population response to a given climate change scenario include realistic measures of uncertainty. The approach combined available data from numerous agencies (e.g., US Geological Survey, US Environmental Protection Agency, the National Oceanic and Atmospheric Administration, and the US Fish and Wildlife Service) and models (e.g., North American Regional Climate Change Program and sturgeon population, movement, and bioenergetics model) across a variety of scales. This work provides the framework for describing the potential consequences of global climate changes on large riverine ecosystems, alerting decision-makers to the most likely consequences, and producing a new suite of indicators of large riverine ecosystem change and health. The project also provides explicit means for scaling results up or down multiple hierarchical levels and associated uncertainty. The goal was to evaluate the potential distributional changes in an ecological system, given distributional changes implied by a series of linked climate and system models under various emissions/use scenarios for evaluation of management options for coping with global change consequences and a powerful tool for assessing uncertainty of those evaluations.
