Marc-Olivier Beausoleil

McGill University
Temporally varying local selection revealed by fitness landscapes in Darwin’s finches

Marc-Olivier Beausoleil1,* Andrew P. Hendry, Luke O. Frishkoff, Leithen K. M’Gonigle, Joost A.M. Raeymaekers, Sarah A. Knutie, Luis Fernando De León, Sarah K. Huber, Jaime A. Chaves, Dale H. Clayton, Jennifer A. H. Koop, Jeffrey Podos, Rowan D. H.

Ecological speciation, the formation of new species following evolution of reproductive isolation due to divergent selection, is considered to be a major driver of the production of biodiversity (Rundle and Nosil 2005; Nosil 2012; Svensson and Calsbeek 2012). Fitness landscapes, which describe the relationship between trait or genotypic variation and a fitness measure, can help understand the role of divergent selection during population differentiation (Wright 1932; Simpson 1953; Schluter and Nychka 1994; Benkman 2003; Gavrilets 2004; Martin 2016). However, fitness landscapes typically show a static representation of a particular selection regime, and alone cannot provide insights into agents the agents of selection that shape them. Thus, they may not capture the longer-term dynamics that occur within complexes of introgressing species, especially when temporal environmental heterogeneity causes shifts in the relationships between trait variance and fitness (Calsbeek et al. 2012). Here, we build fitness landscapes by quantifying the relationships between survival probability and phenotype using a 15-year mark-recapture dataset of over 653 medium ground finches (Geospiza fortis), and relate changes in the topology of these landscapes to climate variables. Our results help to shed light on the role of climate in driving temporally varying nonlinear selection, which can act to both facilitate and impede ecological speciation.