Evidence of local adaptation in the demographic response of American ginseng to interannual temperature variation.

Bioclimatic envelope models of species' responses to climate change are used to predict how species will respond to increasing temperatures. These models are frequently based on the assumption that the northern and southern boundaries of a species' range define its thermal niche. However, this assumption may be violated if populations are adapted to local temperature regimes and have evolved population-specific thermal optima. Considering the prevalence of local adaptation, the assumption of a species-wide thermal optimum may be violated for many species. We used spatially and temporally extensive demographic data for American ginseng (Panax quinquefolius L.) to examine range-wide variation in response of population growth rate (λ) to climatic factors. Our results suggest adaptation to local temperature, but not precipitation. For each population, λ was maximized when annual temperatures were similar to site-specific, long-term mean temperatures. Populations from disparate climatic zones responded differently to temperature variation, and there was a linear relation between population-level thermal optima and the 30-year mean temperature at each site. For species that are locally adapted to temperature, bioclimatic envelope models may underestimate the extent to which increasing temperatures will decrease population growth rate. Because any directional change from long-term mean temperatures will decrease population growth rates, all populations throughout a species' range will be adversely affected by temperature increase, not just populations at southern and low-elevation boundaries. Additionally, when a species' local thermal niche is narrower than its range-wide thermal niche, a smaller temperature increase than would be predicted by bioclimatic envelope approaches may be sufficient to decrease population growth.