Climate adaptation planning faces three central types of uncertainties: How much climate change will occur? How will ecosystems respond to a given level of climate change? How will people respond to resultant changes in ecosystems? We address these issues by using an agent-based modeling system that simulates the interactions and feedbacks among climate change, vegetation succession, wildfire and land-use decisions within an alternative futures planning framework. Through this framework we explore a fully crossed set of 8 contrasting future scenarios that vary across three dimensions in a 815 km2 landscape in Oregon’s Willamette Valley: 1) two climate change scenarios, 2) two land use scenarios that accommodate a projected doubling of human populations over the next 50 years, and 3) two land management scenarios in which landowners are encouraged to reduce fire hazard by conventional thinning or through restoring fire-adapted oak ecosystems. The biophysical modeling components link the FlamMap fire model with a state-transition model of succession driven by downscaled projections from the dynamic global vegetation model MC1. The human-decision component uses a survey of local landowners to parameterize agent decisions in the spatially explicit model Envision, which includes socio-economic evaluators that provide feedbacks from landscape change to landowner behaviors.
Results/Conclusions
Under the MIROC A2 climate scenario, little change in wildfire behavior was projected and wildfire risk remained relatively minor. Under the Hadley A2 scenario, significant increases in area burned and wildfire intensity were projected, and the risk of an extreme fire event, although still low, increased dramatically. Under both scenarios, the greatest increase in wildfire risk was due to dispersed development patterns, which put more people and homes in jeopardy in the wildland-urban interface. This points to an important dilemma for planners and citizens: fireproofing the landscape under either fire hazard management scenario required large financial investments across broad geographic areas because wildfire location is highly stochastic. Relying on conventional thinning strategies was less expensive in the short term, but has few other societal or ecological benefits. Restoring fire-adapted ecosystems actually increased the area burned and rate of fire spread, but allowed fire to move through the landscape with less risk to people and structures. Furthermore, it substantially increased population viability for sensitive wildlife that use these vanishing oak ecosystems. These results emphasize the importance of developing climate adaptation strategies that confer societal benefits even if climate change impacts do not turn out to be as severe as projected.