Our failure to understand or predict evolutionary dynamics under climatic change precludes much conservation planning. Evolution may reduce extinction under global warming, but few studies have explored how genetic covariation, the norm for most quantitative traits, will affect the course of evolution under rapid climatic change. To draw attention to and begin to fill this gap, we draw from the population genetics literature and explore climate-driven evolution using a multi-trait model under two qualitative scenarios of climate change. Under a monotonic change in the mean environment and a change in the amplitude and frequency of a periodic environment, we show that the angle between the direction of the largest genetic covariation and the selection gradient is important in determining a population's fitness decline, or lag load. When the environment changes monotonically in the direction of the greatest covariation, the population is able to more closely track the changing environment resulting in a lower lag load. In contrast, when the environment changes in a direction of low covariation, the ability of the population to track the changing environment is lower, and the population experiences a higher lag load. In a periodic environment, populations suffer a higher lag load under increased environmental amplitude than under increased frequency. These observations suggest that populations where the angle between the largest genetic covariation and the selection gradient is large, as well as populations experiencing an increased magnitude of environmental extremes, may be vulnerable to extinctions and genetic bottlenecks and may benefit from conservation efforts that enhance the preservation of genetic diversity. To make specific predictions of evolutionary trajectories and obtain estimates of lag loads for natural populations, climatic changes have to be quantified in terms of fitness landscapes and genetic covariation among climate-related traits must be measured. We performed an extensive review of the literature and found only 24 studies that quantify covariation in traits involving climate.
Bibliographical noteFunding Information:
Thank you to the 2002 Biodiversity Working Group at the University of British Columbia, including A. Peters, S. Shirley, R. Sargent, and D. Ally, and to participants in the BIRS Retreat in Mathematical Ecology and Evolution for discussions regarding this paper. R. Gomulkiewicz, S. Otto, D. Promislow, and D. Schluter provided suggestions on the manuscript. J. Kerrins and J. Sunday contributed to literature searching. We thank two anonymous reviewers for comments that substantially improved the manuscript. This research was supported, in part, by the Center for Biodiversity Research at UBC and by the Pacific Institute for the Mathematical Sciences (Canada).
Copyright 2019 Elsevier B.V., All rights reserved.
- Correlated evolution
- Environmental change
- Environmental variance
- Fitness landscape
- G matrix