Exploring the synergistic potential of fungicides and parasites as stressors of bumble bee health and pollination services in greenhouse tomatoes

Study Location

All data collection was completed at a commercial tomato greenhouse facility[1] in northern Ohio. This facility consists of three greenhouses sized 265m x 245m. Each greenhouse contains 33 rows of plants on both the north and south sides, 2m wide, for a total of 66 rows of tomatoes. The tomato variety grown in each greenhouse was tomato on the vine.

Study Species

The Common Eastern Bumble Bee (Bombus impatiens)

The common eastern bumble bee is a stable species traditionally found across the eastern half of the united states [1]. B. impatiens is a generalist species, typically easy to rear, and is an important crop pollinator of tomato greenhouses [2,3]. The majority of commercial bumble bees in the US come from two sources, Koppert Biological Systems Inc. or BioBest Ltd.[4]. Each colony arrives with one queen and anywhere between 20-50 female workers.

Study Design

Colony Placement

Each greenhouse had a total of 70 hives at a time, as decided by our grower collaborator, and we monitored 10 randomly selected hives per greenhouse. We placed these 10 colonies of commercial B. impatiens colonies in each of the three tomato greenhouses between July 20 – October 30 of 2022. Each colony was present in the greenhouse for a total of 11 weeks. We weighed each colony prior to placement inside the greenhouse, and upon removal from the greenhouse. Before being placed inside the greenhouse we also removed ten workers from each colony to be sampled for the presence of Crithidia bombi. Each experimental colony was placed along the walkway spanning the length of the greenhouse, roughly equal distance apart (10m).

Pollination

During the 11-week period that the bumble bee hives were located in the greenhouse, a trained pollination scorer recorded the level of anther bruising on 6 plants in each of the 66 sections within the greenhouse, for a total of 396 pollination-score samples each week. Bumble bees pollinate through a method called “buzz pollination”, where they grasp the anther of a flower and vibrate to dislodge the pollen [5]. The resulting necrotic tissues can be measured on a scale of bruising intensity to quantify pollination efficiency [6]. Higher bruising levels are correlated with more desirable pollination and higher fruit yields [6]. The scorer recorded anther bruising on a scale of 0-3, with 0 = none, 1 = low, 2 = intermediate, 3 = high pollination, respectively [5].

Colony health

At the conclusion of the 11-week period we collected each of the 10 colonies in each greenhouse. Each colony was placed in a freezer until all bees were dead. We then collected all of the workers, queens, drones, and new gynes from each colony. We also collected a sample of brood cells, including new-queen cells when possible, and honey pots from each colony. We counted the number of each caste, and will measure the radial cell of a sample of workers and drones from each colony to make inferences on body quality [7].

Crithidia bombi infection

We gut-dissected 10 workers collected from each colony prior to placement in the greenhouse, and 10 workers collected from each colony after removal from the greenhouse. We will use PCR methods to compare the presence of Crithidia bombi infection in the bumble bees before and after being placed in the greenhouse.

Chemical application and testing

Our collaborator provided us with complete records of pesticide applications that took place during the duration of our experimental hives presence in the greenhouse. We will send samples of the collected brood cells, new queen cells, and honey pot contents to the USDA-Agricultural Marketing Service National Science Laboratory (Gastonia, NC, USA) for testing of the presence of each of these chemicals. This will allow us to investigate if the chemicals sprayed on the plants are carried back to the colony and fed to brood in detectable quantities by workers.

Statistical analysis

We will correlate the pollination score data, as measured by anther bruising record, with timing of each of the pesticides sprayed throughout the monitoring period. We will analyze these results using a generalized linear model with a negative binomial distribution. We will hold chemical as the explanatory variable, with average greenhouse temperature, and average greenhouse hours of light as covariate, and greenhouse number as a random effect. We will use this same model structure to correlate counts of each caste with the presence or absence of Crithidia in the hive, either before or after placement in the greenhouse.

[1] Exact location kept anonymous at the request of collaborator.

References

[1]       R. Hatfield, S. Colla, S. Jepsen, L. Richardson, R. Thorp, S. Foltz Jordan, IUCN Assessments for North American Bombus Spp., The Xerces Society for Invertebrate Conservation, Portland, OR, 2015.

[2]       D.R. Artz, B.A. Nault, Jnl. Econ. Entom. 104 (2011) 1153–1161.

[3]       J.P. Strange, Journal of Economic Entomology 108 (2015) 873–879.

[4]       H.H.W. Velthuis, A. van Doorn, Apidologie 37 (2006) 421–451.

[5]       L.A. Morandin, T.M. Laverty, P.G. Kevan, Ec 94 (2001) 462–467.

[6]       L.A. Morandin, T.M. Laverty, P.G. Kevan, Ec 94 (2001) 172–179.

[7]       C.B. Muller, T.M. Blackburn, P. Schmid-Hempel, Ins. Soc 43 (1996) 227–233.