Dispersal ability and biogeographic gradients influence gene flow of 3 aquatic insects in Laurentian Great Lakes interdunal wetlands

Population genetic connectivity is influenced by multiple abiotic and biotic attributes, including geography, dispersal ability, and life history, which may lead to different patterns of population structure of organisms occupying similar habitats. We investigated how differences in dispersal ability and biogeographic gradients correspond with population structuring of 3 aquatic insect species found within naturally fragmented interdunal wetlands along the eastern shoreline of Lake Michigan in midwestern USA. Interdunal wetlands are small, highly fragmented, and patchily distributed along the eastern coast of Lake Michigan, USA. Our focal species, Anax junius (Drury, 1773), Notonecta undulata Say, 1832, and Caenis amica Hagen, 1861 were chosen as high, intermediate, and low dispersers, respectively. We hypothesized that all insect populations experience isolation by distance with relatively low gene flow among sites, but that the strength of isolation by distance varies with dispersal ability. We used cytochrome c oxidase subunit I sequence data to confirm species identification and restriction enzyme-association DNA sequencing for population genomic analyses. Our cytochrome c oxidase subunit I data revealed that Caenis populations consisted of multiple species split along a latitudinal gradient. Restriction site-association DNA sequencing data showed that A. junius displayed strong isolation by distance, where N. undulata did not. Additionally, both A. junius and N. undulata populations displayed 2 genetic clusters along the coastline, and genetic diversity increased along with latitude. These results indicate that biogeographical variables, such as latitude and covarying abiotic factors, may be stronger predictors of population structure than dispersal ability and that inference of population structure within aquatic macroinvertebrates should be on a species-specific basis.