Improving Honey Bee Health and Crop Visitation during Pollination

Progress report for LNE19-392R

Project Type: Research Only
Funds awarded in 2019: $148,827.00
Projected End Date: 06/30/2021
Grant Recipient: University of Maryland, College Park
Region: Northeast
State: Maryland
Project Leader:
Kirsten Traynor
University of Maryland
Co-Leaders:
Dennis vanEngelsdorp
University Maryland
Expand All

Project Information

Summary:

Our objective is to improve honey bee health during pollination. Our novel approach ameliorates pesticide exposure from contaminated pollen. Through the use of multifactorial experimental designs, we can determine if trapping pollen (reducing immediate pesticide exposure) and/or supplemental pollen feeding (improving nutrition) improves colony health. If effective, this novel approach provides beekeepers with inexpensive protection against the deleterious effects on honey bee health resulting from pesticide exposure during crop pollination, ensuring growers access to reliable pollination services. This progress report is intended only to update the work completed on Hypothesis 1 and 2.

Revised Project Objective

This study used standardized colony assessments to test whether pollen trapping and pollen supplements reduces the effects of potential pesticide exposure in Maryland field (corn/soybean) and vegetable crops and improves the health of honey bee colonies in a commercial honey bee operation. Here, we conducted a field trial comparing 3 treatment groups (pollen trapped with no supplemental feed, pollen trapped with supplemental feed, no pollen trapped with supplemental feed) and a control (no pollen trap and no feed) to ascertain which provides the greatest benefit and colony performance. 

Inspection the colony for frames of bees, queen status and brood condition
Taking data on colony inspection
Inspecting and sampling (for varroa and nosema) colonies at the Clarksville yard
Upper Marlboro yard
Colony with trap installed. This one is being fed a pollen supplement but the trap is not engaged.
This colony has the pollen trap engaged and is not being fed a pollen supplement

 

Project Objective:

Our objective is to improve honey bee health during pollination. Our novel approach ameliorates pesticide exposure and potentially increases pollen foraging on target crops, ensuring a steady pollination supply for blueberry and cranberry growers. This novel approach provides beekeepers with inexpensive protection and if “precision pollination” training is effective, significantly improves pollination of target crops, an important innovation as bees notoriously stray from blueberry and cranberry fields to more appealing resources within flight range.

Cooperators

Click linked name(s) to expand
  • Dennis vanEngelsdorp (Researcher)

Research

Hypothesis:

Hypothesis 1: Pollen trapping reduces colony exposure to pesticides during pollination.

Hypothesis 2: Feeding pollen and nectar substitutes during pollination improves colony nutrition and resilience, reducing colony losses.

Hypothesis 3: Precision pollination—training bees to crop odors prior to pollination—can enhance foraging on target crops.

Hypothesis 4: Fungicide exposure increases queen events, while general pesticide exposure induces brood cannibalism, and nutritional stress.

Materials and methods:

Project Revisions

Due to project changes approved by NE SARE March, 2020; 1) the field study was conducted in field crop plots in MD rather than blueberry fields, and 2) nectar substitutes were not included in the study.

A cohort of 48 colonies distributed over 3 agricultural yards in Maryland (16 colonies/yard) were monitored between 30 July and 20 August 2020. In each yard trial colonies were assigned using a stratified random assignment strategy to one of four (4) treatment groups:

  1. Control: no pollen trapping, no feeding (12 colonies)
  2.  Pollen trapping and supplemental feeding (12 colonies)
  3. Pollen trapping and no supplemental feeding (12 colonies)
  4. No pollen trapping, supplemental feeding (12 colonies)

 

Pollen supplementation (Bee-Pro brand) was provided continuously during this experiment. A 2 lb supplement patty was applied each week during the duration of the study. At the end of each week, any unconsumed patties were removed, weighed and a new patty was put in place. All colonies had pollen traps installed but only the colonies assigned trapping had the entrance engaged. Pollen was collected every 3 days and weighed. At the end of the experiment, pollen trapped from the Trap/No feed group from each yard (3) and Trap/Feed group from each yard (3) was aggregated, homogenized and sent to the USDA/AMS lab for pesticide analysis. Colony health inspections (queen status, varroa and nosema sampling, overt disease inspection, etc.) were conducted twice – at the beginning and end of the project (day 21). No colonies received any additional protein or carbohydrate supplementation and no colonies were equalized or divided during the duration of the trial.

Research results and discussion:

Queen status and Colony mortality – There was no effect of treatment on queen status or colony mortality. Two colonies became queenless (one Feed+/Trap+ and one Feed-/Trap- from two separate apiaries) and one colony died (Feed-/Trap+) over the course of the study. The dead colony was removed from the dataset prior to analysis. 

Varroa – Overall, mean Varroa per 100 bees increased from day 0 to day 21 of the study (day 0 mean = 10.79, day 21 mean = 16.22). This result is consistent across treatments (See Figures 1a-c), with the negative control colonies exhibiting the lowest net change in mean Varroa load over time (Figures 2a,b). However, there are no statistically significant differences among treatments in either the net change or the rate of change of mean Varroa per 100 bees (Figures 2a,b; Table 1).

Figure 1. Mean Varroa per 100 bees at day 0 and day 21 of the field study plotted by colony treatment (Fig. 1a = Feed+/Trap+, Feed+/Trap-, Feed-/Trap+,Feed-/Trap) as well as plotted showing each factor individually (Fig. 1b Treatment = Feed+/Feed -; Fig. 1c Treatment = Trap+/Trap-).

 

Figure 2. Box and whisker plots representing the Net (Fig. 2a, upper panel)  and Rate (Fig. 2b, lower panel) of change in Varroa per 100 bees over the 21-day study period. Black horizontal lines with boxes represents the median, the upper tip and lower tip of vertical whiskers represent the first and third quartile and the filled black circles located above or below the colored bars show treatment outliers. Dashed horizontal red bar at the 0 y-axis represents no change. For rate of change values, an increase or decrease by 1 along  the y-axis represents a 100% change in median Varroa per 100 from day 0 to day 21). 

Colony Size (Frames of Bees) – Overall, mean colony size measured as frames of bees increased from day 0 to day 21 of the study (day 0 mean = 10.79, day 21 mean = 16.22). As with Varroa and brood pattern, no statistical differences were observed among treatments in net and rate of change of brood pattern (Figure 3a,b; Table 1).

Figure 3a,b. Box and whisker plots representing the Net (Fig. 3a, upper panel)  and Rate (Fig. 3b, lower panel) of change in colony size (measured as frames of bees) over the 21-day study period. Black horizontal lines with boxes represents the median, the upper tip and lower tip of vertical whiskers represent the first and third quartile and the filled black circles located above or below the colored bars show treatment outliers. Dashed horizontal red bar at the 0 y-axis represents no change. For rate of change values, an increase or decrease by 1 along  the y-axis represents a 100% change in median Varroa per 100 from day 0 to day 21).

 

Brood Pattern –  Mean brood pattern quality declined over time (mean 0 = 3.53, day 21 mean = 2.80); again no significant treatment differences were detected (table 1).

Figure 4a,b. Box and whisker plots representing the Net (Fig. 4a, upper panel)  and Rate (Fig. 4b, lower panel) of change in brood pattern (evaluated as a score between 0-5s) over the 21-day study period. Black horizontal lines with boxes represents the median, the upper tip and lower tip of vertical whiskers represent the first and third quartile and the filled black circles located above or below the colored bars show treatment outliers. Dashed horizontal red bar at the 0 y-axis represents no change. For rate of change values, an increase or decrease by 1 along  the y-axis represents a 100% change in median Varroa per 100 from day 0 to day 21).

Nosema – Mean Nosema load (reported as millions of spores per bee;) increased over the course of the study  (mean day 0 = 0.11, day 21 mean = 0.15). Significant differences were not detected among treatments; however, model results indicated there may be some significant differences among treatment groups at day 0 with regard to Nosema load and needs to be further evaluated. 

Figure 5,b. Box and whisker plots representing the Net (Fig. 5a, upper panel)  and Rate (Fig. 5b, lower panel) of change in Nosema (evaluated as million spores per bee) over the 21-day study period. Black horizontal lines with boxes represent the median, the upper tip and lower tip of vertical whiskers represent the first and third quartile and the filled black circles located above or below the colored bars show treatment outliers. Dashed horizontal red bar at the 0 y-axis represents no change. For rate of change values, an increase or decrease by 1 along  the y-axis represents a 100% change in median Varroa per 100 from day 0 to day 21).

Pesticides – We expect the results from the pesticide analysis within 1-2 months and will provide the analysis of that with our final report. 

Research conclusions:

Treatment effects – No statistical significant differences were detected in the Net change of Varroa, Nosema, FOBs, or brood pattern. However, model results did detect significant differences among treatment groups at day 0 for FOBs and Nosema (Table 1). Therefore, additional analyses are underway to account for these variations in initial conditions, and initial results will be updated when data analysis is completed.

 

Table 1. Model results showing statistical significance of Net Change in individual treatment factors (Feed+/-, Trap+/-) as well as their interactive effect (Feed+/- * Trap+/-) over the course of the study. 

 

Colony mortality – Colonies will be reevaluated in Spring 2021 to determine overwintering mortality rates. 

Overt disease symptoms – Data analysis is underway to evaluate disease symptoms during the field trial. 

Pollen Pesticide Analysis – Pollen collected from traps during the study have been sent to the USDA AMS laboratory pending pesticide analysis. Once data are available, data will be analyzed to evaluate whether there is any association between pollen pesticide profiles and treatment effects on colony health metrics.

Participation Summary

Project Outcomes

2 Grants applied for that built upon this project
1 Grant received that built upon this project
$50,000.00 Dollar amount of grant received that built upon this project
Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.