Community-level pollination: Biotic and abiotic determinants and temporal variation

Background/Question/Methods

Community-level pollination studies have increasingly used network analysis to describe the visitation of pollinator species to plant species.  Many focus on network topology rather than the basic biotic and abiotic factors shaping plant-pollinator interactions.  Our work is motivated by sources of variation in plant-pollinator interactions: time, temperature, and availability of different flowering species.  This pursuit holds great potential to enhance our understanding of factors determining pollination services.  Moreover, while most plant-pollinator interaction studies have been conducted in warm or temperate climates, our study takes place in the cold-climate steppe mountain in northern Mongolia.  The data were collected within a valley from permanent observations plots set up at two elevations, 1670 m and 1800 m a.s.l.  We counted the total number of available flowers for each plant species and observed pollinators during three diurnal periods, which was repeated 11 times across the summer.  Additionally, we measured air temperature at the two elevations.  At each elevation we determined how pollinator visitation varies diurnally and throughout the summer months.  We also constructed a hypothetical path model to determine how much of the variation in pollinator visitation can be explained by air temperature and floral abundance.

Results/Conclusions

Pollinator visits varied throughout the day and throughout the season.  Visits were highest in frequency during the mid-afternoon time period (1200-1530) at both elevations.  There was a large amount of variation in pollinator visitation throughout the summer at both elevations; at the lower elevation, pollinator visitation was highest during the last two weeks of July while at the higher elevation, visitation was highest slightly earlier, during the first two weeks of July.  Path analysis showed that, in the higher elevation, temperature and floral abundance both significantly explained pollinator visitation, (β = 0.42 and β = 0.37, respectively), but at the lower elevation, only temperature was significant (β = 0.66).  Visitation frequency and the biotic and abiotic factors contribution to pollinator visitation vary within the landscape.  Studies should be aware of landscape-wide variation and account for this when studying plant-pollinator interactions.