Spatio-temporal variation in interacting plant-pollinator populations: the role of phenology, species traits, and geography in sympatric populations of spring-flowering Camassia

Background/Question/Methods

Multispecies complexes of native plants and their associated pollinator faunas vary spatially and temporally, often reciprocally influencing these ecological interactions and the similarly dynamic evolutionary processes that also drive the patterns observed in natural populations. Variation in the abundances and preferences of ‘generalist’ pollinators is still little understood relative to sequentially flowering, sympatric populations of species with similar floral morphologies.  We studied temporal and spatial variation in pollinator abundances and preferences for multiple populations of two Camassia species, C. quamash, and C. leichtlinii. Both provide nectar and pollen for social and solitary bees in spring flowering wet-prairie habitats. To understand how flowering phenologies might affect plant-plant interactions through shared pollinators, we tracked phenological differences in insect and plant populations over time, as well as foraging behavior; per flower and plant visitation rates; and two measures of pollinator effectiveness in sites where Camassiaflowered in unispecific and mixed patches. We further manipulated plant density, species, and floral traits to determine the extent to which differences in insect visitation are influenced by these variables in early and late flowering populations. We created artificial flowers and quantified pollinator visitation to three traits that were varied independently.

Results/Conclusions

We detected significant geographic variation in the frequency distributions of the most prevalent insect pollinators, and spatiotemporal variability in the abundances of native and non-native foragers on Camassia in focal communities. Several lines of evidence suggest a strong preference for C. leichtlinii, including visitor abundances up to 5 x greater for solitary bees and social bees than on C. quamash and high rates of visitation to artificial flowers with C. leichtlinii traits. Both species shared a common ‘generalist” community of foragers, but Bombus was less prevalent on C. quamash, yet still effective in pollen transfer, in contrast to bombyliid flies. Visitation rates were significantly reduced to C. quamash but not C. leichtlinni in mixed versus unispecific patches. Introduced Apis mellifera was most prevalent on both species during flowering overlap. More studies spanning the full growing season are needed for such communities. While specialization can be advantageous for optimally foraging insects and for reducing heterospecific pollen transfer, opportunistic ‘generalist’ foragers often predominate where spatiotemporal variability exists in pollinators or floral resources. With climatic change, resource availability may vary asynchronously with insect life cycles, leading to fitness consequences for populations at multitrophic levels, necessitating a nuanced understanding of interacting complexes of species.