Predicting the effects of flower density on pollinator species composition

Background/Question/Methods

Empirical studies in a variety of systems have found shifts in flower visitor species composition with increasing floral density.  Because flower visitors vary in their quality as pollinators, changes in the species composition of flower visitors could have significant effects on plant fitness.  However, little is known about what drives changes in flower visitor species composition with changing floral density, particularly amongst non-territorial species, such as those populating many temperate pollinator guilds.  Here I present a foraging model in which I explore how flower visitors with diverse traits should distribute themselves across resource patches differing in floral density.  I allow visitors to vary in search speed, flower handling time, metabolic rates during flight and while handling flowers, and the foraging metric they maximize (net rate of energy intake or energetic efficiency).  I also explore how the resulting distributions of flower visitors could influence relations of plant fitness to floral density, assuming realistic variation in the visitors’ quality as pollinators.  

Results/Conclusions

Variation in any of a variety of traits can cause shifts in flower visitor species composition with changing floral density, suggesting that such shifts should be common in nature.  Specifically, flower visitors with relatively slow search speeds, short flower handling times, low metabolic rates while handling flowers, and/or high flight metabolic rates should be found in dense patches, while species with the opposite traits should be found in sparse patches.  In addition, flower visitors that maximize energetic efficiency are more likely to be found in dense patches whereas species that maximize net rate of energy intake are more likely to be found in sparse patches.  Depending on how pollinator quality is associated with other flower visitor traits, the shifts in flower visitor species composition predicted by this model can generate a variety of effects of floral density on plant fitness, including increases or declines in plant fitness with increasing floral density.