Drivers of adaptive capacity in wild populations: implications for genetic interventions

Abstract

The unprecedented rate of environmental change in the Anthropocene poses evolutionary challenges for wild populations globally. Active human interventions are being increasingly considered to accelerate natural adaptive processes. Evolutionary models can evaluate how species may fare under future climate, elucidate which evolutionary processes are critical to rapid adaptation, and how active interventions may influence fitness trajectories of organisms. Here we use polygenic metapopulation adaptation models to quantify the relative importance (effect sizes) of different eco-evolutionary parameters on the rates of adaptation in wild populations i) without active interventions, and ii) under a subset of active interventions. We demonstrate that genetic diversity (heterozygosity, He), population connectivity and the effect size of additive genetic variance are the primary drivers of natural adaptation rates. We quantify the effect sizes of these parameters on population fitness across three proposed assisted evolution scenarios and identify critical thresholds for intervention effectiveness and implementation. Specifically, the interventions tested here were most effective at low levels of genetic diversity in target populations (He < 0.2) and when timed during a cold-to-warm phase of an ENSO-like oscillation. Beneficial levels of connectivity were highly dependent on desired outcomes for the meta-population. We also present a global meta-analysis of genetic diversity in tropical reef-building corals as a case study of how thresholds derived from evolutionary models can be used to guide decision making by managers. We find genetic diversity to be highly variable by coral taxon and region, highlighting how thresholds from evolutionary models can be used in conjunction with empirical data to assess intervention needs and priorities. Finally, we highlight the critical knowledge and data gaps to produce the next suite of applied models for conservation management decision-support.