Modeling the pollination of highbush blueberry (Vaccinium corymbosum) to explore varying pollination strategies

Background/Question/Methods

Over seventy percent of plant species grown for food benefit from animal and insect-mediated pollination and production of bee-pollinated crops is increasing across the world. It has recently been estimated that global food crop pollination is worth over $200 billion annually and as such, there is significant global concern over the long-term sustainability of pollination. Production of most pollination-dependent crops, including highbush blueberry (Vaccinium corymbosum), relies on managed hives of the European honey bee (Apis mellifera) although native bees may provide supplemental pollination services. In highbush blueberry, bumble bees (Bombus spp.) have been shown to be more efficient pollinators than honey bees, with lower numbers necessary to achieve full pollination of the crop. Currently, bumble bees are the only commercially available alternative pollinator to honey bees for blueberry. The goal of this research was to develop a model of blueberry pollination to test hypotheses about how to utilize these two pollinators for sustainable crop pollination. Optimizing the use of commercial pollinators will help to ensure a reliable food supply for future generations.

Results/Conclusions

A deterministic model of blueberry pollination was developed to predict annual yield for one acre of highbush blueberry under typical spring weather conditions. This model relies upon inputs of highbush blueberry cultivar, weather, and managed pollinator inputs to calculate pollination during the bloom period and subsequent blueberry yield, based on experimentally-derived parameter estimates. Highbush blueberry base temperature was found to range from 7.14 to 7.96°C and was then used to characterize blueberry bloom phenology based on accumulating growing degree-days. Average percent fruit set was found to differ significantly among blueberry cultivars. The model was used to examine effects of different pollinator stocking levels and combinations on yield and to determine profit-maximizing pollinator input strategies. For the ‘Bluecrop’ cultivar, the profit-maximizing stocking level of honey bees was determined to be 1.5 hives per acre compared to 3.25 bumble bee colonies per acre for fields stocked only with bumble bees. When examining combinations of the two pollinators, a constant rate of input substitution was found, indicating no benefit of combining managed pollinators under normal weather conditions. An online interface of this model is being developed to communicate the results of this research to a broad range of stakeholders.