Process-based, Bayesian methods to advance understanding of forest change and inform adaptive management

Malcolm Itter, PhD

Date: June 25, 2020
Time: 3 pm (s.t.)
Place: zoom. Please register by sending an email to

Forest ecosystems face an uncertain future under global change. Rapid shifts in climatic conditions and natural disturbance regimes are already subjecting regional forest systems to novel environmental conditions to which they are not adapted. Historic forest management approaches, designed to emulate regional disturbance regimes, may no longer sustain critically-important forest ecosystem services under these novel environments. Adaptive forest management seeks to maintain forest function and sustain these services by promoting ecosystem resistance and resilience to novel conditions in the short term, and by facilitating the formation of robust, complex adaptive systems in the long term.
Developing effective adaptive management strategies requires testing potential approaches to determine their impacts on forest demographic processes in the face of changing conditions. While adaptive management experiments provide empirical observations of the efficacy of management strategies, the time frame over which results are observed (decades to centuries) often limits their utility for informing current management decisions. Instead, managers rely on forest simulation models capable of forecasting conditions under alternative management strategies to test their efficacy and inform management actions in the present. Scientific understanding of forest demographic processes and their response to weather and disturbance are the foundation of these models. Yet, the lack of understanding of forest responses to novel, no-analog conditions results in significant uncertainty and variability in forecasts of forest demography. I will present Bayesian methodologies to assimilate experimental and observational data with process-based models to advance functional understanding of complex ecological processes, refine their approximation, and better predict future forest change. Further, I will describe the use of these methodologies within Bayesian decision theoretic frameworks allowing for informed adaptive management decisions in the face of uncertainty.