PERSIST: Predicting evolutionary rescue of a species in space and time

An increasingly important biological challenge is to understand how physiological variation in populations will evolve in response to rapid climate change. Evolutionary theory predicts that populations should be locally adapted to their environment if spatial variation in selection is not overwhelmed by gene flow. Anthropogenic climate change appears to be disrupting local adaptation and increasing the frequency of maladaptation on an ecological timescale, but for surprising reasons. As part of a working group, I developed mathematical models showing that climate change should not make local populations less fit than foreign populations on average, but it should increase the variance in local advantage. This is because climate anomalies sometimes reinforce existing local adaptation and sometimes counteract it. Reinforcing and counteracting anomalies on average cancel one another out, but increase the variance in outcomes, which results in more frequent local maladaptation. A global data synthesis of 149 reciprocal transplant studies over 40 years reveals the variance in local adaptation has increased dramatically, resulting in more frequent local maladaptation ( Bontrager et al. In review). My roles in this collaborative effort include leading the mathematical modeling and co-leading Bayesian statistical analysis in Stan.

Next, we are directly testing whether climate change is disrupting local adaptation and if populations are adapting to changing climate on an ecological timescale. We are using the California native wildflower Mimulus cardinalis as a model system. Using a field reciprocal transplant during an anomalously warm year, we found strong local adaptation in the South, but not in the North. This is consistent with the pattern from our metaanalysis in which warm anomalies reinforce local adaptation in lower latitude populations and counteract local adaptation at higher latitudes.

By disrupting the fit between organisms and their environment, climate change may cause populations to decline and go extinct. Adaptive evolution can rescue a declining population by increasing its growth rate. Although adaptation and population trends (demography) are typically studied separately by evolutionary biologists and ecologists, respectively, they are fundamentally connected by fitness. In declining populations, individuals’ fitness is generally too low to replace themselves, but adaptive evolution can boost fitness as long as there is heritable variation for critical traits. The major question is whether adaptation to changing climate will be faster than the rate of demographic decline. I am co-PI on an NSF award with Seema Sheth, Jeff Diez, Lluvia Flores-Renteria, and Jay Sexton to test whether natural selection during the mega-drought between 2011 and 2016 in the western United States increased population growth rate most in populations that started with the greatest heritable variation in drought-related traits and fitness. Parameter estimates from this field experiment will inform whether evolutionary rescue is possible for other species at risk from changing climate.