Evolutionary rescue in novel environments: towards improving predictability

Authored by Michael Barfield, Robert D Holt

Date Published: 2016

Sponsors: French National Research Agency (ANR) United States National Science Foundation (NSF)

Platforms: R

Model Documentation: Other Narrative Mathematical description

Model Code URLs: Model code not found

Abstract

Background: Populations are often subject to changes in their environments (either locally or due to movement of a population), which, if large enough, require them to adapt in order to persist. This is `evolutionary rescue'. Questions: What factors affect the ability of a population to recover after a sudden change in its environment? What can be measured about an initial population, prior to the environmental change, that can improve the predictability of evolutionary rescue? Methods: A deterministic model and simulations of an individual-based model (IBM). Results: Heritability that decreases with decreasing population size could prevent evolutionary rescue in the deterministic model. For the IBM, the probability of rescue decreased with increasing magnitude of the environmental change and with decreasing initial population size. At times, heritability of a trait can increase as selection occurs. Most extinctions occurred shortly after the change. Rescue depended significantly on the genetics of the population at the time of the environmental change, and predictive power about which populations go extinct, or persist, is improved by knowing the mean genotypic value and genetic variance in the initial population. However, there remains considerable uncertainty in such predictions. Conclusions: Persistence after a sudden environmental change was greater in populations with more individuals and more genetic variance at the time of the change, and depended on rapid adaptation soon after the change, without which extinction was likely. Understanding the amount and dynamics of genetic variation can improve predictability of persistence, but there is inescapable randomness in evolution and ecology that will always, we believe, preclude tight predictions.
Tags
Prevent extinction