Sex‐biased dispersal and spatial heterogeneity affect landscape resistance to gene flow in fisher

Abstract

Abstract Genetic connectivity results from the dispersal and reproduction of individuals across landscapes. Mammalian populations frequently exhibit sex‐biased dispersal, but this factor has rarely been addressed in individual‐based landscape genetics research. In this study, we evaluate the effects of sex‐biased dispersal and landscape heterogeneity on genetic connectivity in a small and isolated population of fisher ( Pekania pennanti ). We genotyped 247 fisher samples collected across the southern Sierra Nevada Mountains of California. We tested for genetic evidence of sex‐biased dispersal using sex‐specific population structure and spatial autocorrelation analyses, and sex‐biased dispersal tests of the assignment index, F ST , and F IS . We developed resistance surfaces for eight landscape features hypothesized to affect gene flow and optimized each resistance surface independently by sex. Using multiple regression of distance matrices and an information‐theoretic model selection approach, we fit models of genetic distance to landscape resistance distance separately by sex and geographic region. We found genetic evidence of sex‐biased dispersal with significant differences in F ST , F IS , and spatial autocorrelation between sexes. Optimal resistance values differed by sex, and model variables, fit, and parameter estimates varied substantially both between sexes and between geographic regions. We found a stronger relationship between landscape features and genetic distance for females, the philopatric sex, than the more widely dispersing males. Our results show that landscape features influencing gene flow differed by both sex and regional heterogeneity. Conducting analyses by sex and by region allowed for the identification of landscape genetics relationships not discernible when analyzed together. Our results show that failing to account for these factors can confound results and obscure relationships between landscape features and gene flow.