An approach to assessing vulnerability of forest communities to climate change using impact models and expert elicitation

Background/Question/Methods

Climate change has the potential to alter forest composition and productivity through direct changes in temperature and precipitation and through the effects of physical and biological disturbance. Scientists and land managers use vulnerability assessments to determine which forest species and communities may be at risk under projected climate. To date, forest vulnerability assessments have tended to focus on coastal, alpine, or far northern regions with an emphasis on species over habitats or ecosystems. Vulnerability assessments have also generally relied on either quantitative modeling or a ranking approach based on qualitative information. As part of a broader effort to develop management responses to climate change in the Central Hardwoods, Northwoods, and Central Appalachians regions, we developed a method for assessing forest vulnerability that combines multiple quantitative models and expert elicitation from scientists and land managers. A range of potential futures was assessed using statistically downscaled results from combinations of two emissions scenarios and two global climate models, which were used to forecast regionally-specific climate trends and to set boundary conditions for three forest impact models. In each region, a panel of local experts in forest ecology and management examined the range of possible futures to determine potential vulnerability of forest community types to climate change. Uncertainty around each vulnerability determination was assessed using a confidence rating based on the amount of evidence and agreement among evidence.

Results/Conclusions

We developed vulnerability ratings for 6-12 forest community types in each of the Central Hardwoods, Northwoods, and Central Appalachians regions. Which community types were most vulnerable to climate change varied by region. For example, mesic hardwood forests were considered to have high vulnerability in the Central Hardwoods region, but moderate vulnerability in the Northwoods. Floodplain forests, by contrast, were considered to have relatively high vulnerability in the Central Hardwoods region, and lower vulnerability in the Northwoods. These contrasting vulnerabilities were due to regional differences in how climate change may affect system drivers, stressors, and dominant species and the capacity of a forest community to adapt to climate change. Results from these assessments are currently being used to develop climate change adaptation actions in each region. In addition, the assessment methodology can be used in different regions or for finer-scale vulnerability determinations of forest communities.