The Curious Beekeeper

August 2014


Running the Risk, Part I: An Overview of

Honey Bee Risk Assessment Basics

Pesticide Research Institute


The decline in pollinator populations and the crisis of unsustainable losses in managed honey bee colonies in several regions of the world has pressured regulatory agencies to revamp pesticide risk assessment methods for honey bees. While some countries are beginning to require more rigorous risk assessment protocols for honey bee exposure to pesticides, there is still considerable uncertainty involved in the process and wide gaps in the information needed to fully understand pesticide risks to bees.

In this installment of The Curious Beekeeper, we provide part I of Risk Assessment 101—a general overview of pesticide risk assessment for pollinators, including information on how regulatory agencies determine a reference dose for bees and estimate bee exposures. In Part II next month, we will take a look at the more involved Tier II and Tier III studies and compare EPA’s new pollinator risk assessment guidance published in June of 20141 to the methods currently used in Canada2, 3 and the European Union.4

Understanding Honey Bee Risk Assessments

Whether evaluating the risks to humans or honey bees, the process of risk assessment is focused on four questions:

1)    What kind of adverse effects does the pesticide cause?
2)    What dose is required to cause the adverse effects?
3)    What exposure is predicted based on pesticide use patterns?
4)    How does the predicted dose compare to the dose that is known to cause adverse effects?

Adverse effects: Looking specifically at honey bee risk assessment, the adverse effect currently used by regulatory agencies is death of 50 percent of a test population. The dose that is lethal to 50 percent of the test bees is commonly known as the LD50.
About now, you might be wondering why the death of 50 percent of your bees was determined to be the best regulatory endpoint. The answer is that the LD50 is relatively easy to measure in the lab. But allowing exposures up to the LD50 means that pesticide-related losses are likely, even when the application is legal. In the new US EPA guidance, the dose at which the agency may require mitigation is 40% of the LD50 value (see below), so it’s not quite as bad as it sounds.

One important component missing from the current risk assessment process is the determination of more subtle adverse effects that pesticides may have on bees, including impairment of queen health that might lead to drone-laying, poor brood patterns, and supersedure; poor survival of larval bees such as that caused by insect growth regulators; and in adult bees, impairment of immune function, navigation, hive communications, and longevity.

Determining a dose of concern: The dose at which the adverse effect occurs is determined by treating a group of test bees with a range of different doses and observing the changes in the occurrence of effects with dose. Two exposure routes are relevant for honey bees: contact with treated foliage and oral exposure from eating pesticide-containing nectar and pollen.

For the acute contact LD50 test, a known quantity of the pesticide is dissolved in a solvent (generally water or acetone) and applied to the top of the thorax in test bees.5 For the oral LD50 test, a known concentration of the pesticide in 50% sucrose solution is fed to test bees in a single dose. Bees are observed for 48 hours, and the number of dead bees at each dose is counted. In the U.S., the acute contact LD50 is the only study that is now required for every pesticide. US EPA has developed a High-Medium-Low toxicity scale (see box below) to define pesticide toxicity to bees.

Bees are also exposed through the food they eat, so the acute oral LD50 is an important measure of toxicity, particularly for systemic pesticides that may occur in the pollen and nectar bees gather for food. This test is not currently required by US EPA, but can be required if the regulatory staff determine it is necessary. Table 1 provides sample LD50 values for representative pesticides that are either widely used or commonly found in beehive matrices.

Estimating exposure: Honey bees can be exposed to pesticides in many ways, most commonly through direct sprays to blooming crops, dust drifting from planting pesticide-treated seeds, or through contact with spray or dust residue on plants while the bees are foraging. Bees are also exposed through their food and water, which can be contaminated by direct sprays or through uptake of systemic pesticides through the roots or leaves of the plant.

Once the potential sources of exposure have been identified by examining the use patterns of the pesticide, it is then possible to predict the dose that might be encountered by different castes of bees—nurse bees, foragers, queen bee, and larvae. For the foragers, exposure is estimated by evaluating the application rate of the pesticide (in lbs/acre) and translating this value into an amount of active ingredient deposited on a bee in the field. Exposures through food can be estimated by measuring the concentration of the pesticide in pollen and nectar (in µg/kg of food) after a pesticide application, then multiplying this concentration times the weight of food a typical bee consumes in a day. The result is a dose in micrograms of active ingredient per bee per day (µg/bee/day).

In the current risk assessment paradigm, chronic exposures over longer time periods are not considered. The current approach of assuming only a single exposure does not account for the fact that bees will work a food source until it is exhausted, storing excess pollen and nectar in the hive. The colony may then be exposed whenever the bees tap into it (which may not be until winter). Exposure continues until the food is used up or until the pesticide degrades.

In general, the longer an organism is exposed to a pesticide, the lower the dose required to show a toxic effect. In a recent paper, Sanchez-Bayo et al. explored the concept of a T50—the time for a bee to consume an LD50’s worth of pesticide, assuming normal consumption rates of pollen and nectar.10 This approach is the start of a more realistic risk assessment process, but few longer-term toxicity tests are currently available.

Assessing risk: The final step in the risk assessment process is to compare the  ...