The distribution of genetic variation in a population is a function of the environment through which genes move. In insect pollinator systems, how a pollinator perceives and interacts with a landscape is critical to the movement of pollen. In this study, we model varying landscape resistances at multiple resolutions and across stratigraphies to examine which combination of environmental factors best explains pollen pool structure in Cornus flordia. We used lidar data to describe forest structure and microsatellite markers to create pollination networks among 17 C. florida maternal trees. Multiple cost distance algorithms and analyses were used to create resistance distances and relate them to genetic distances.
Results/Conclusions
Our results show that accounting for the 3-dimensional structure of the landscape is important for describing pollen movement in C. florida. In this system, the sub-canopy vegetation stratum, which includes conspecific C. florida canopy cover, may be attracting pollinators. Additionally, void spaces in the sub-canopy appear to be facilitating the movement of pollen among C. florida individuals. Finally, the performance of the cost distance algorithms varied across different combinations of environmental factors. LCP appears to be more robust across resistance ratios whereas Circuitscape was sensitive to changes in resistance ratios. This study highlights the importance of describing the vertical dimension in landscape resistance surfaces for C. florida and may be applicable more broadly across taxa. Here we provide a larger framework for accounting for landscape 3-dimensionality in the creation of resistance surfaces for future landscape genetic studies.