Flower diversity and bee reproduction in an arid ecosystem

Background: Diverse flower communities are more stable in floral resource production along the flowering season, but the question about how the diversity and stability of resources affect pollinator reproduction remains open. High plant diversity could favor short foraging trips, which in turn would enhance bee fitness. In addition to plant diversity, greater temporal stability of floral resources in diverse communities could favor pollinator fitness because such communities are likely to occupy the phenological space more broadly, increasing floral availability for pollinators throughout the season. In addition, this potential effect of flower diversity on bee reproduction could be stronger for generalist pollinators because they can use a broader floral spectrum. Based on above arguments we predicted that pollinator reproduction would be positively correlated to flower diversity, and to temporal stability in flower production, and that this relationship would be stronger for the most generalized pollinator species. Materials and Methods: Using structural equation models, we evaluated the effect of these variables and other ecological factors on three estimates of bee reproduction (average number of brood cells per nest per site, total number of brood cells per site, and total number of nests per site), and whether such effects were modulated by bee generalization on floral resources. Results: Contrary to our expectations, flower diversity had no effect on bee reproduction, stability in flower production had a weakly negative effect on one of the bee reproductive variables, and the strength of the fitness-diversity relationship was unrelated to bee generalization. In contrast, elevation had a negative effect on bee reproduction, despite the narrow elevation range encompassed by our sites. Discussion: Flower diversity did not affect the reproduction of the solitary bees studied here. This result could stem from the context dependence of the diversity-stability relationship, given that elevation had a positive effect on flower diversity but a negative effect on bee reproduction. Although high temporal stability in flower production is expected to enhance pollinator reproduction, in our study it had a weakly negative—instead of positive—effect on the average number of brood cells per nest. Other environmental factors that vary with elevation could influence bee reproduction. Our study focused on a small group of closely-related bee species, which cautions against generalization of our findings to other groups of pollinators. More studies are clearly needed to assess the extent to which pollinator demography is influenced by the diversity of floral resources.