LB 9-101 - Modeling the combined effects of climate change, demography, and fire on Californian plant population dynamics

Background/Question/Methods:

Many plant species have limited dispersal ability relative to the projected rates of climate change and could face population decline or extinction within the next century. Climate change also interacts with other stressors such as altered fire regime and land use change, threatening California plant communities with the potential of substantial biodiversity loss. Species distribution modeling can estimate species’ range loss or expansion with climate change, but these statistical techniques often overlook the ecological processes that drive species range shifts, such as dispersal, life history, or fire response. We tested the relative importance of these factors by combining species distribution modeling and demographic population modeling to better estimate the threat of climate change on three iconic Californian plants (oak, pine, and ceanothus) with different life histories. Species distribution models predicted the initial range and potential future habitats (under four climate scenarios), while density-dependent stage-structured population models projected species’ population dynamics. Fire uniquely affects the species’ survival and fecundity, so we simulated the species’ dynamics under a range of changing fire regimes, modifying future fire return intervals and the effect of fuel accumulation on fire probability.

Results/Conclusions:

Our simulations showed that these species are expected to decline under each combination of climate change scenario and fire scenario. In particular, simulations with the high emissions Hadley climate model projected a substantial midcentury decline in habitat suitability. Though the total habitat recovered, populations did not fully recover because population abundance lagged behind. Moreover, although total habitat suitability increased in several simulations, species could not reach the full extent of these potential range expansions, even with high dispersal distances. Both oak and pine species are resistant to moderate intensity fires, but we found that increasing fire frequencies and fuel-dependence of fires reduced the final population sizes of these species. In contrast, ceanothus is an obligate seeder that requires fire to germinate, so simulations in which fire was frequent (but not too frequent) generally resulted in less dramatic population decline due to climate change. Our results suggest that combining species distribution models, demographic models, and the effects of fire might improve our ability to predict species’ range loss (or expansion) under cumulative or interacting threats and how best to manage them.