COS 108-5
Invasive pathogen trumps natural selection thwarting evolutionary rescue

Thursday, August 8, 2013: 2:50 PM
L100G, Minneapolis Convention Center
Shawn T. McKinney, US Geological Survey, Maine Cooperative Fish and Wildlife Research Unit, Orono, ME
Diana F. Tomback, Integrative Biology, CB 171, University of Colorado Denver, Denver, CO
Background/Question/Methods

Evolutionary rescue (ER) is a theoretical process by which genetic adaptation allows a population to recover from near extinction following rapid environmental change. ER requires mutations (existing or de novo) that confer resistance to the new environmental stressor and, following population decline, selection that leads to renewed adaptation and subsequent population growth. Simulations, modeling, and laboratory experiments have addressed populations of one to a few species in environments with a single stressor.  What is the potential for ER in a natural population impacted by multiple stressors and compelled by higher-level community interactions? We examined how interspecific interactions vary across a geographic gradient of stress-induced mortality, and how differential interaction strength affects the likelihood of population persistence. Whitebark pine (Pinus albicaulis) forests occur throughout the central and northern Rocky Mountains, but are declining rapidly from Cronartium ribicola, an invasive fungal pathogen causing white pine blister rust, and the native mountain pine beetle (Dendroctonus ponderosae). We quantified rust infection, beetle infestation, tree mortality, forest structure and composition, seedling regeneration, rate of predispersal seed predation by red squirrels (Tamiasciurus hudsonicus), and the probability of seed dispersal by Clark’s nutcracker (Nucifraga columbiana), the pine’s obligate seed disperser, to form models of persistence likelihood.

Results/Conclusions

We found that habitat context profoundly influenced the strength of interspecific interactions and the likelihood of community persistence. As tree mortality increased from a novel combination of stressors, cone production declined, the rate of squirrel seed predation increased, and the probability of seed dispersal by nutcrackers declined. Few seeds remained for avian dispersal, greatly decreasing the likelihood of dispersal for seeds with putative rust resistance. Forests comprising unsuitable squirrel habitat (climax whitebark pine) experienced significantly lower seed predation compared to mixed-conifer forests; however, seed dispersal remained highly unlikely when cone production was below a threshold range. A lower probability of seed dispersal was associated with fewer regenerating seedlings across all forest types. Therefore, a key criterion of ER— that the growth rate of resistant types increases over time— cannot be realized. Our findings suggest that ER is improbable in whitebark pine communities because the severity of environmental stressors coupled with higher-level trophic interactions—particularly the obligate dependency on nutcrackers for seed dispersal—prevents natural selection.  It is said that extinction under environmental change is a race between demography and adaptive evolution. Without management intervention, it will be difficult to prevent extirpation of whitebark pine populations experiencing high mortality.