Syrup preparation
Control (0 ppb clothianidin) sucrose solution was mixed at 1:1 w:w (e.g. 500 g sucrose:500 mL distilled water). Sucrose was added to distilled water in a 5-gallon bucket and mixed using an electric drill with a mortar mixing attachment until sugar was completely dissolved. Sucrose solution for solutions with clothianidin (PESTANAL, CAS# 210880-92-5) was mixed in the same manner but 50 mL was withheld (thus “short”) to allow for the added volume of respective clothianidin spikes. 500 g of sugar is dissolved in 450 mL of distilled water to allow for the addition of a 50 mL spike to achieve 1 kg of treatment solution. 950 g of “short” sugar solution was transferred to a Nalgene bottle, then the spike added to each individual bottle. A 10 ppm clothianidin stock solution was made by dissolving 1.0 mg of clothianidin, in 100 mL of distilled water, using a mixing bar but without heat. To avoid problems with static electricity, the clothianidin was weighed into a small, nonreactive plastic receptacles and those receptacles were placed in the solution, the solution stirred, and the receptacles removed after confirming the clothianidin had dissolved. For the 5-ppb solution: 0.5 mL of the stock solution was mixed into 49.5 mL of distilled water to achieve 50 mL of spike solution, which was then added to 950 g of the short sucrose solution to achieve 1 kg of 5-ppb clothianidin syrup. For the 20-ppb solution (only in 2nd experiment) 2.0 mL of stock solution was mixed into 48.0 mL of distilled water, and that solution added to 950 g of the short solution to achieve 1 kg of 20-ppb clothianidin syrup.
AZ 2017 experiment
On 20 April, 2017, 24 bee colonies were established from packages of Italian honey bees (A. mellifera ligustica) (C.F. Koehnen & Sons, Inc., Glenn, CA 95,943) of approximately 1 kg honey bees in painted, 10-frame, wooden Langstroth boxes (43.7 l capacity) (Mann Lake Ltd., Hackensack, MN) with migratory wooden lids. At establishment, each colony was given 4 full or partial frames of capped honey, 2 frames of drawn but empty comb, 2 frames of partially drawn with some capped honey, 3 frames of foundation and a 1-frame feeder. Queens were marked, and during the course of the studies any queen replacements, such as for supersedure queens, was done with queens from the same breeder. Hives were placed on stainless steel electronic scales (Tekfa model B-2418 and Avery Weigh-Tronix model BSAO1824-200) (max. capacity: 100 kg, precision: ± 20 g; operating temperature: − 30 °C to 70 °C) and linked to 16-bit dataloggers (Hobo UX120-006 M External Channel datalogger, Onset Computer Corporation, Bourne, MA) with weight recorded every 5 min. The scales were powered by deep-cycle batteries connected to solar panels. The system had an overall precision of approximately ± 20 g. Hives were arranged in a circular pattern around a central box that contained the batteries and electronic gear. Hives within such a group were 0.5- 1 m apart and groups were > 3 m apart. During the course of the experiments the power systems had occasional malfunctions, resulting in short periods of missing data for some hives.
Colonies were all fed 2 kg sugar syrup (1:1 w:w) and 250 g pollen patty, made at a ratio of 1: 1: 1 corbicular pollen (Great Lakes Bee Co.): granulated sugar: drivert sugar (dry fondant sugar with approximately 8% invert sugar) (Domino Foods, Yonkers, NY). On 10 July a temperature sensor (iButton Thermochron, precision ± 0.06 °C) enclosed in plastic tissue embedding cassettes (Thermo Fisher Scientific, Waltham, MA) was stapled to the center of the top bar on the 5th frame in the bottom box of each hive and set to record every 15 min. The same day, pieces of slick paperboard coated with petroleum jelly and covered with mesh screens were inserted onto the hive floor to monitor Varroa mite fall within the hive58. The boards were removed 2 days later, and the number of mites counted on each board. Infestation levels of Varroa were again monitored during and post-treatment. Colonies were treated with amitraz (Apivar, Arysta LifeScience America Inc., New York, NY) on 19 October.
Hives were assessed on 12 July, and approximately every 5–6 weeks thereafter until November, the again 13 February 2020 and finally on 29 March using a published protocol21,28. Briefly, the hive was opened after the application of smoke, and each frame was lifted out sequentially, gently shaken to dislodge adult bees, photographed using a 16.3 megapixel digital camera (Canon Rebel SL1, Canon USA, Inc., Melville, NY), weighed on a portable scale (model EC15, OHaus Corp., Parsippany, NJ), and replaced in the hive. Frame photographs were analyzed later in the laboratory (see below). During this first assessment (but not subsequent assessments), all hive components (i.e. lid, inner cover, box, bottom board, frames, entrance reducer, internal feeder) were also shaken free of bees and weighed to yield an initial mass of all hive components. At the initial inspection, 3–5 g of wax and honey were collected from each hive into 50 ml centrifuge tubes and stored at − 80 °C; samples collected in September, prior to treatment, were pooled and subjected to a full panel analysis for residues of pesticides and fungicides, from all major classes, by the Laboratory Approval and Testing Division, Agricultural Marketing Service, USDA. Wax samples were collected only at the initial assessment in order to establish a baseline exposure—the lack of agriculture or landscaping within foraging distance excluded the possibility of further exposure. Honey samples from later assessments were pooled within treatment group and subjected only to neonicotinoid residue analysis.
Newly-emerged bees (NEBs) were sampled by pressing an 8 cm × 8 cm × 2 cm cage of wire mesh into a section of capped brood, then returning the following day to collect NEBs that had emerged within the cage over the previous 24 h. The NEBs were then placed in a 50 mL centrifuge tube, frozen on dry ice, and stored at − 80 °C. At the laboratory, 5 bees per hive per assessment date were placed in Eppendorf tubes, weighed, dried for 72 h at 60 °C, then re-weighed to determine average wet and dry weight per bee. NEBs were collected on 12 July and 24 August 2017 (brood levels were too low in October 2017 for sampling).
After the first assessment, hives were ranked with respect to adult bee mass and then randomly assigned to treatment group, ensuring that the average bee masses per group were approximately equal and after eliminating assignments that resulted in excessive spatial clumping of the colonies. Just prior to treatment all broodless frames containing honey and/or pollen were replaced with frames of empty drawn comb collected earlier from the same apiary. Colonies were then fed 2–3 kg syrup twice per week from 14 July to 21 August, with clothianidin concentrations depending on their treatment group. Syrup consumption per colony was recorded. Hives were assessed approximately every 5–6 weeks thereafter until November, and again in February and March. At each of those subsequent assessments, the same protocol was followed but only the frames, hive lid and inner cover were weighed. The hive lid and inner cover weights were compared to previous values and used to correct for moisture content changes in the hive components and improve estimates of adult bee mass. Food resources in the colonies were very low by mid-November so all colonies were provided with an additional 2 kg sugar syrup at that time.
AZ 2018 experiment
The 2017–2018 experiment was conducted in the same manner, with the same or similar equipment and using the same bee suppliers. Varroa mite fall onto adhesive boards was monitored 6–9 July, and hives were assessed and sampled on 5 July in the same manner as before. NEBS were sampled on 6 July, 23 August and 4 October, 2018. CO2 probes (Vaisala Inc., Helsinki, Finland), calibrated for 0–20% concentrations, were installed in five hives in each treatment group and set to record CO2 concentration every 5 min. Colonies were fed 3 kg sugar syrup twice per week from 12 July to 20 August with the same pesticide concentrations as the previous year and assessed as before, with the experiment ending in February. Varroa infestation levels were monitored at the end of August and again at the beginning of November. Colonies were treated with amitraz (Apivar) on 19 October. Unlike the previous year, colonies were found to have sufficient resources to last to spring and so they were not fed any additional syrup after the treatment period.
MS 2018 experiment
Full bee colonies, each comprised of two “deep” boxes as described above, were obtained from a local bee supplier (Gunter Apiaries, Lumberton MS) as nucleus colonies the previous year. Queens were bred locally and subspecies was unspecified. Colonies were placed on hive scales (Tekfa model B-2418) on 16 May 2018. Colonies were assessed, using the methods described above, on 11 July 2018 and temperature sensors (iButtons) were installed on 12 July 2018. Frames of honey were removed on 18 July and colonies were randomly placed in treatment groups. Treatment feeding commenced 24 July, lasting 31 August, using the same concentrations and amounts as described above. Colonies were not fed pollen patty because sufficient pollen was available. Colonies were assessed again 18 September 2018 and finally on 27 March 2019. Samples of 300 bees were collected on 7 May 2018, washed in 70% ethanol and the Varroa mites counted. Colonies were treated for Varroa (Checkmite, Mann Lake Ltd) on 28 June 2018. The apiary site was assessed using the National Agricultural Statistical Service (NASS) Cropscape web site (https://nassgeodata.gmu.edu/ CropScape) to obtain acreage estimates for all land use categories within an approximately 1.8 km radius of the apiary. Bees can forage beyond that distance; the radius was chosen to provide a sufficient area (> 1000 ha) to be representative of the forage available to the bees.
Data analysis
The total weight of the adult bee population was calculated by subtracting the combined weights of hive components (i.e. lid, inner cover, box, bottom board, frames, entrance reducer, internal feeder) obtained at the start of the experiment (model EC15, OHaus) from the total hive weight recorded the midnight prior to the inspection. The area of sealed brood per frame was estimated from the photographs using ImageJ version 1.47 software (W. Rasband, National Institutes of Health, USA) or CombCount59; this method has been described elsewhere20,28. Food resources in the colonies were calculated as the total frame weight, less (1) the mass of the brood (calculated at 0.77 g/cm2) and (2) the mass of an empty frame of drawn comb (555 g)28.
Honey bee colony survivorship was analyzed using Proc LifeReg (SAS Inc. 2002). Survivorship curves were generated for each treatment group within each experiment. Treatments compared using ANOVA (α = 0.05) (Proc Glimmix, SAS Inc. 2002), with experiment as a random factor, with respect to three parameters: (1) the 30th percentile; (2) the 50th percentile; and 3) a shape variable calculated by subtracting the 40th from the 30th percentile.
Daily hive weight change was calculated as the difference between the weight at midnight of a given day to the weight 23 h 55 min later. Continuous temperature data were divided into daily average data and within-day detrended data. Detrended data were obtained as the difference between the 24 h running average and the raw data. Sine curves were fit to 3-day subsamples of the detrended data, taken sequentially by day28. Curve amplitudes, representing estimates of daily variability, were reduced to a data point every 3 days, to ensure no overlap between data subsamples, for repeated measures MANOVA analyses. CO2 concentration data were treated in the same fashion.
Adult bee masses , brood surface area, and total food resources for the 1st sampling occasion after treatment were analyzed across all three experiments using ANOVA (Proc Glimmix, SAS Inc. 2002). Further analyses were conducted among the Arizona experiments across multiple sampling occasions using repeated-measure MANOVA (Proc Glimmix, SAS Inc. 2002). A similar approach was taken with newly-emerged bee weights, i.e. initial analysis across AZ 2017 and AZ 2018 using ANOVA for a single sampling occasion, then for AZ 2018 using MANOVA across two sampling occasions. Daily hive weight change, internal hive temperature average and variability (i.e. amplitudes of fit sine curves) and CO2 concentration average and variability were used as response variables in repeated-measures MANOVA with treatment, sampling date, experiment and day, and all 2-way interactions, as fixed effects and with pre-treatment values as a covariate to control for pre-existing differences. Proc Univariate was used with all response variables to inspect the data for normality. Log transformations were conducted where necessary to improve normality. Analyses of hive weight and temperature were limited to approximately 3 months after the end of treatment to focus on the active season, and consisted of omnibus tests that initially included all three field experiments followed by analyses within each experiment. The reason for this is that effects that are significant in one trial might not be so in another, or might be significant but in a contrary fashion. CO2 concentration data were only collected in AZ 2018.
NEB data were analyzed with Treatment, Sampling date and their interaction, with the July values as a covariate. Varroa fall were analyzed within each Arizona experiment, with the pre-treatment values used as covariates where applicable. Varroa alcohol wash data for MS 2018 were analyzed separately.
Rainfall, and ambient temperature and CO2 data, were obtained for the Arizona site: AmeriFlux US-SRM Santa Rita Mesquite, https://doi.org/10.17190/AMF/1246104; and temperature and rainfall data for Mississippi: National Environmental Satellite, Data, and Information Service, National Oceanic and Atmospheric Administration, Poplarville Experimental Station, MS US USC00227128.
Source: Ecology - nature.com
