Modeled climate change effects on distributions of Canadian butterfly species

Climate change effects on biodiversity are being documented now frequently in the form of changes in phenology and distributional shifts. However, the form that these effects will take over a longer timespan is unclear; for this understanding, a quantitative, validated, predictive approach is key. Here, we use ecological niche modeling and general circulation model outputs to estimate future potential geographic distributions of 111 Canadian butterfly species. We develop future estimates under two emission scenarios from each of two climate change modeling centers; future projections for biodiversity are not only scenario dependent (more severe emission scenarios produce more severe effects on species' distributions) but also model dependent (the Canadian Centre for Climate Modelling and Analysis results were more severe than the Hadley Centre results). One interesting feature is the appearance of disjunctions in species' distributions, hence creating "vicariant events" over very short time periods. In general, however, a cost of 1%-3% additional loss of species' distributions is associated with more severe scenarios of emissions and climate change, suggesting that subtle biodiversity consequences are associated with the different climate futures debated in political circles.