Anthropogenic effects, not climatic, shaped Holocene population expansion of an insular bee fauna
AimRemote island systems with small landmasses and reliable estimates of human occupancy are ideal model systems to disentangle the roles of global climatic changes and local human occupation on biota. Here, we tested this with demographic reconstructions of native, ground-nesting bees with non-specialised floral visitation habits to infer changes in effective population size (Ne). LocationViti Levu, Fiji, southwest Pacific. TaxonLasioglossum (Homalictus) bees (Hymenoptera: Halictidae) MethodsWe applied a combination of both mitochondrial and genomic datasets from five species distributed across the elevational range for our analyses. Both kinds of data were used for validation of population assignments and testing for structure, and to investigate any demographic shifts. ResultsAll lowland species and populations showed strong signals of increasing Ne that correspond to the timing of human occupation of Fiji, but not Holocene climatic changes. Highland populations, with greater isolation and less affected by anthropogenic impacts, did not show evidence of recent rapid increases in Ne. Population expansion rates across the elevational gradient differed between taxa, with earlier and larger increases in predominantly lowland species. This is consistent with the movement of people inland and into montane elevations of the island, and corresponding landscape changes that benefit the ecology of these bees. Main conclusionsSpecific life history traits of these bees, combined with substantive clearing of forest cover and floristic changes at lower elevations, has likely increased nesting opportunities and abundance of invasive floral resources. Our findings contrast with recent evidence that human occupation of Fiji has resulted in decreased ant biodiversity and raise the paradoxical possibility that human-mediated environmental changes may benefit some native montane tropical insect faunas.