Posted on August 9, 2015 · Posted in Bees

Why Analyzing Bees for Pesticide Contamination Isn’t Always Useful

I’ve been seeing posts lately about pesticide-related bee-kills that people are experiencing this summer and wanted to share some insights we have gained from our Hive Tracking Project, in which we are evaluating potential contributors to colony losses, including pesticides, pathogens, and parasites. In the process of doing this study, we have done quite a bit of analysis of bees, pollen, wax, honey and brood and also evaluated work others have done. Here is what we’ve learned.


You don’t often find pesticide residues in bees that have consumed contaminated nectar

The published data show that detections of pesticide residues in bees rarely exceed the detection limits the lab is capable of. There are several reasons for this.

1) Metabolism by bees: Live bees (and bees that have died recently) metabolize pesticide chemicals, transforming them into degradates that the lab isn’t looking for. You can’t detect what you don’t look for, and in many cases, we don’t know the identity of the degradates in honey bees. Researchers who analyzed the rate of metabolism of imidacloprid found that 20 minutes after they dosed the bees with imidacloprid, 30% of the chemical had already been degraded (Suchail, et al. 2004). So the concentration of pesticide remaining in the bees may not be high enough for the lab to see. To maximize your chances of finding pesticide residues in bees, sample only bees that are still alive and twitching, get the sample into a freezer as soon as possible, and ship it to the lab in a cold shipping box.


2) Degradation by environmental factors: Pesticides in bees that have been lying dead in front of the hive in the hot sun will be rapidly degraded by microbes, oxygen, sunlight and high temperatures, to the point that insufficient pesticide remains in the bees for the lab to detect. Do not waste your money on analysis of long-dead bees!

3) Concentration limitations: If a bee has ingested contaminated nectar, it has to be above a certain concentration to be found in a lab test. Let’s do a simple calculation to see why.  We’ll assume the following:

a) A typical amount of nectar carried by a forager bee is 30 microliters (uL).
b) A typical forager bee weighs about 100 milligrams (mg)
c) The Limit of Detection (LOD) for pesticides analyzed by the USDA lab in Gastonia, NC ranges between 1 and 100 ug/kg (the same as parts per billion, ppb), with an average detection limit of 10 ug/kg. A very sensitive laboratory detection limit is 1 ug of pesticide/kg of bees analyzed.
d) There are 1,000,000 microliters (uL) in a liter (L)
e) There are 1,000,000 milligrams (mg) in a kilogram (kg)


We’ll then ask the question “What does the minimum concentration of pesticide in the nectar have to be so the lab can detect it when it’s inside the bee?” and do a little math to find out the answer:

The detectable amount of pesticide per bee for a pesticide with a detection limit of 10 ug/kg is:
10 ug of pesticide/kg of bees (average detection limit) *
100 mg/bee (weight of a bee) *
1 kg/1,000,000 mg (conversion factor) = 0.001 ug of pesticide/bee

The minimum concentration of pesticide in nectar inside a forager bee that can be detected is then:
(0.001 ug of pesticide/bee) ÷
(30 uL of nectar/bee * 1 L/1,000,000 uL) =
33 ug of pesticide/L of nectar (ppb)

We must also account for the fact that the process by which pesticides are extracted from bee tissue is not 100% efficient. Typical recoveries are 70-90%, so the actual minimum detectable concentration of pesticides in nectar inside a bee is more like 40-50 ug/L.

Sensitivity will be greater for pesticides with lower detection limits and less for pesticides with higher detection limits. Similarly, if recovery efficiencies are higher or lower, the detection limits will be better or worse.

4) Analysis provides an average value: Finally, it is important to consider that a random sample of bees taken will likely contain varying amounts of pesticide residue per bee. Because 3 grams of bees (approximately 30 bees) are used for a typical laboratory analysis, the result will be an average residue level over all of the bees analyzed. In the 30 bees, there will be a mix of bees that consumed quite a bit of the contaminated nectar and those that consumed none or just a small amount of it. There will also be  variation in concentrations depending on the extent of degradation that has occurred in the bees analyzed. This averaging effect can result in the amount of pesticide extracted in the analysis being below the detection limit.

Where residues in bees may be high

There are several scenarios in which it is likely to find measurable residues in bees:

1)  Bees exposed to pesticides at the concentrations they are applied: Bees sprayed directly during a pesticide application, foraging on plants still wet with a pesticide application, or encountering pesticide-contaminated dust in their foraging activities (as from dust from the planting of treated seeds) may have detectable residues. The bees may collect some pesticide residue or dust on the outside of their bodies. Because the concentration of the pesticide in these situations is quite high, it won’t take much exposure to have high enough levels of pesticide for the lab to detect. In this case, the pollen the bees have collected will likely have pesticide residues as well, often at much higher concentrations than in the bees.

2) Bees exposed to miticides in the hive: Bees in hives that have been treated with miticides for Varroa typically have moderate to high levels of miticides or their degradation products in their bodies, depending on the chemical used and when the bees were treated. Most commercial beekeepers and many hobby beekeepers treat their hives with miticides to keep the Varroa population down. If the miticides are applied correctly, they do not cause massive bee-kills; nevertheless, the residues will be detectable. This may be problematic for the beekeeper, because regulatory staff investigating a bee-kill incident will often make a determination that beekeeper-applied miticides killed the bees because the only chemical detected in the laboratory analysis was a miticide. However, analysis of many samples of beehive matrices show that even healthy beehives can contain moderate to even relatively high levels of miticides, so the detection of these chemicals in bees is unlikely to be the primary cause of a pesticide-related bee-kill. More likely is that the pesticide causing the kill is below the detection limit in the bees for the reasons explained above.

Find out more about sampling bees and beehive matrices for pesticides in the articles we wrote for the American Bee Journal:

About the Author

Susan Kegley is Principal and CEO of Pesticide Research Institute. She is a PhD Organic chemist with expertise in pesticide chemistry, fate and transport and toxicology of organic chemicals, and U.S. pesticide regulation.