Novel data support model linking floral resources and honey bee competition with bumble bee abundances in coastal scrub

Abstract

Growing evidence supports that many bee populations are declining, with potentially serious consequences for pollination services. Mechanistic models that predict bee abundances from drivers like floral resource availability can be a powerful way to understand and address declines, but remain rare and largely unvalidated. I used temporally and spatially novel data to validate previous analyses linking bumble bee (Bombus spp.) declines in California coastal scrub with loss of floral resources, mediated by drought and competition with non-native honey bees (Apis mellifera). New observations from 2015-2018 were combined with data from 1999-2014 to further test these mechanistic hypotheses and evaluate predictions of a statistical model for Bombus abundances. As predicted, positive associations between spring rainfall and floral abundances and between Bombus and key forage plants were consistent between time periods. Increased A. mellifera abundance corresponded with reduced Bombus use of the most abundant forage plant and lower Bombus abundances in the following year. Quantitative predictions from the Bombus statistical model previously developed for 1999-2014 were relatively unbiased and strongly rank correlated with either spatially or temporally novel data. However, the model consistently underpredicted Bombus abundances when both flower patch and time period were novel. Overall, four new years of data further strengthen evidence that loss of floral resources due to drought and competition with feral Apis mellifera is an important cause of Bombus decline in this habitat. These findings reinforce the value of even simple models that are mechanistically framed, both in understanding past patterns of change and for qualitative prediction.

Divergent responses of generalist and specialist pollinators to experimental drought: Outcomes for plant reproduction

“…Although examining interspecific competition was not the goal of this study, it seems plausible that honey bees may be outcompeting squash bees for nectar in female flowers. Thomson (2021) also found evidence of honey bees outcompeting bumble bees, and this competition was actually heightened by drought conditions due to reduced floral resource production. Future studies could investigate the direct role that drought plays in affecting competition between pollinators collecting floral rewards.…”

“…Water stress, for example, variably affects the performance of generalist versus specialist herbivores (Gutbrodt et al, 2011; Nguyen et al, 2018). With respect to pollination systems, whether or not water stress alters competition between pollinators (Thomson, 2021), drought‐induced reductions in resource quality may prove detrimental to specialists (Rankin et al, 2020) and, subsequently, to their floral hosts (Al‐Ghzawi et al, 2009; Rering et al, 2020; Waser & Price, 2016). Following drought events, host plants that exhibit drought sensitivity may likewise impact the populations of specialist pollinators.…”

Intra‐seasonal and daily variations in nectar availability affect bee assemblage in a monodominant afforested Brazilian Cerrado

“…son 2016;Hachuy-Filho et al 2020). This is especially important in restoration sites located in Cerrado or other habitats where solitary and less competitive native bee species, such as bumblebees and carpenter bees, are abundant (Thomson 2021). Although we did not evaluate whether honeybees excluded the native bee species in our study site, the overall bee diversity remained relatively constant through time regardless of the increase in bee species richness following the increase in flower availability.…”