Principle

A honeybee risk assessment is generally carried out in a tiered manner, where a Tier 1 analysis is intended to be a conservative screen that separates pesticides that are not expected to present a risk from those pesticides that may. Higher tiers are intended to refine the estimates or measures of potential exposure, effects, and the resulting characterization of risk. This tiered risk assessment approach is shown in the figure below.
 

The first step is an assessment of the likelihood of exposure of honeybees to the pesticide. This may be different for foliar spray application and for soil and seed treatment using systemic pesticides. Routes of exposure to the pesticide and subsequent possible effects on the bees are discussed elsewhere.

In the first tier risk assessments, acute toxicity or chronic of the pesticide is generally compared with its predicted environmental concentration (PEC), by calculating risk quotients. Toxicity data as well as PECs are different for adult bees and for bee brood. The risk assessment can be refined by using additional toxicity data and/or by applying more precise exposure estimates (e.g. measured pesticide residues in pollen and nectar).

If the pesticide risk is unacceptable after the first tier assessment, higher tier studies may be carried out. These can be semi-field trials (e.g. tunnel tests) or field studies with free ranging bees. In higher tier evaluations, no risk quotients are calculated such as in first tier assessments, but risks are characterized for the specific situation studied. Higher tier assessments therefore tend to be more specific to local cropping and pesticide application conditions.

Risk mitigation measures can be considered to reduce pesticide risk at any tier. If effective risk reduction measures can be identified and applied, higher tier evaluations may not be required. Risk mitigation for bees is discussed in the Risk Mitigation Module.

The risk assessment procedure described here refers specifically to honeybees (Apis spp.). At present, no validated tiered risk assessment procedures are available for non-Apis bees, such as bumblebees, stingless bees or solitary bees (but see EFSA guidance below). Approaches for extrapolating the risk of a pesticide from honeybees to other non-Apis bees is discussed in a separate section.

Procedures/models

Various guidelines and models are available for honeybee risk assessment. The listed methods all follow very similar risk assessment procedures, based on the general scheme described above. The main differences are in the estimation of exposure and the trigger values leading to higher tiers.

First tier risk assessment methods described for temperate/Mediterranean countries can in principle be applied to honeybees in (sub-)tropical countries. Higher tier studies tend to be more locally specific.

1. EPPO
   The Environmental Risk Assessment Scheme – Honeybees, of the European Mediterranean Plant Protection Organization (EPPO), is one of the oldest risk assessment schemes for honeybees. It has been developed in collaboration with the International Commission for Plant-Pollinator Relationships (ICPPR) and until recently was the recommended method used in Europe. The scheme is specific to honeybees.
   The latest version of the risk assessment scheme has been published in 2010 in the EPPO Bulletin. 
2. North America
   The USA and Canada have recently published their Guidance for Assessing Pesticide Risks to Bees. This joint guidance document replaces previous risk assessment methods used by the two countries. Exposure estimates are provided for different types of pesticides and different stages of honeybees.
   The guidance is specific to honeybees. However, the guidance considers honeybees as a surrogate species for other Apis and non-Apis bees and other insect pollinators.
   The latest version of the guidance document can be downloaded from the US- EPA web site.
3. EFSA
   The European Food Safety Authority (EFSA) published in 2013 its Guidance Document on the risk assessment of plant protection products on bees (Apis mellifera, Bombus spp. and solitary bees). The risk assessment methodology described in the guidance document is now used for pesticide registration in the European Union.
   Detailed exposure scenarios for different types of pesticides are described. Specific risk assessment for bumblebees and solitary bees is included, although the proposed methods are still relatively new.
   The guidance document can be downloaded from the EFSA web site. In addition, an Excel spreadsheet (“Bee tool”) is available to facilitate the calculations of the assessments. It is strongly recommended only to use the spreadsheet after thorough review of (at least) Chapter 3 of the guidance document, which summarizes the different risk calculations.

Data required

The following data are generally required to be able to conduct a local honeybee risk assessment:

• Acute oral and contact toxicity – adults
• Acute and chronic oral toxicity – bee brood
• Chronic toxicity – adults (if available)
• Residual toxicity – adults (if available)
• Good Agricultural Practice (GAP) table, or detailed direction for use
• Measured pesticide residue levels in pollen and/or nectar (if available)
• Higher tier study results (if available/required)

For more information on data requirements for honey bees, go to the relevant section of the Data Requirements module.

Interpretation of the outcome

When interpreting the outcome of the risk assessment, it is important to evaluate to what extent the exposure scenarios used for the assessment represent the expected use of the pesticide in the local situation. In particular, the registrar should assess the uncertainties associated with the risk assessment and possible under- or overestimation of the expected risk.

The risk of the pesticide is generally considered acceptable if, after the Tier 1 assessment, the “risk quotient” (or “risk ratio”, or “hazard quotient”; terminology depends on the risk assessment method) is less than the relevant “trigger value” (or “level of concern”). Note that trigger values are specific to the risk assessment model and cannot be interchanged!

If the risk is unacceptable (risk quotient is greater than the trigger value) the registrar may decide to require higher tier studies and further evaluate the risk for the specific local situation concerned.

Alternatively, the registrar may can assess whether any risk mitigation measures can be applied. For honeybees, risk mitigation is an often applied measure to reach an acceptable risk, in particular for insecticides which often pose an inherent high risk to insect pollinators. Typical risk mitigation measures include prohibition of spraying when crops are flowering or not registering the pesticide for use on crops that depend on pollination (see the Risk Mitigation Module). If risk mitigation measures are required, one should also evaluate whether they are feasible and realistic under the local conditions of use of the pesticide.