Study of Pesticide Risk in Wild Bee Species Highlights EPA Risk Assessment Inadequacies

 Study of Pesticide Risk in Wild Bee Species Highlights EPA Risk Assessment Inadequacies

A study in Science of The Total Environment calculates and compares pesticide risk in 594 wild bee species associated with crops in North America. Current pesticide risk assessments that analyze effects on bees primarily focus on a limited subset of species and do not provide comprehensive protection of all wild bees. “Species commonly proposed as models for pesticide risk assessments may not accurately represent risk for those bee species facing the highest potential risk in agricultural contexts,” the authors postulate.

The researchers continue, “This study presents a novel approach to characterize and compare the relative potential pesticide risk among wild bee species of their association with crops in North America using suites of intrinsic bee traits to quantify species’ vulnerability and extrinsic factors based on the toxic load of crops for bees and the strength of each species’ association with those crops.” In considering multiple factors that vary by species and determining potential harm to each from pesticide exposure, this study highlights the inadequacies of the current risk assessment process used by the U.S. Environmental Protection Agency (EPA).   

The system for risk assessment for pesticides that impact bees includes a tiered process, with Tier I as a screening tool within the laboratory and Tiers II and III as field studies. According to EPA, Tier I uses “conservative assumptions regarding exposure (i.e., assumptions that are likely to overestimate exposure) and uses the most sensitive toxicity estimates from laboratory studies of individual bees to calculate risk estimates.” These studies, however, primarily focus on honey bees such as Apis mellifera and do not consider the varying sensitivity in other bee species.

As the authors point out, “the use of a very unusual and non-representative species, A. mellifera, as a model for all bee species in pesticide risk assessments [is] because of the ease with which the species can be maintained by humans, well developed risk assessment protocols for the species, and its cost effectiveness.” The use of this species is not due to its ability to well-represent all other bee species.

EPA’s process does not factor in the growing body of scientific evidence regarding the negative impacts of pesticide exposure on a wide range of bee species, as well as other vital pollinators. This study suggests that efforts should be focused “on the subset of wild bee species likely experiencing the highest potential pesticide risk as a starting point for protection and conservation goals” instead of a reliance on a single species where the majority of data collected is under laboratory settings that do not mirror real-world exposure. See more on EPA’s failure to protect bees here.

Over 20,000 known wild bee species exist worldwide, with about 3,600 native to North America. Of those, 739 species are known to be associated with agricultural crops. For their research, the authors of this study obtained complete information for all life history categories and size for 594 of those identified bee species to analyze. Since exposure to agricultural pesticides is one of the multiple interacting drivers of wild bee declines globally, representative risk assessments are imperative to protect biodiversity and ecosystem services and thus need to be thorough.

For proper risk assessments, “a comprehensive understanding of both the nature of the risk (risk = hazard x exposure) attributable to pesticide use on crops and the nature of the intrinsic vulnerability of bees to pesticides is required,” the researchers state. In chemical-intensive agriculture, each crop requires different pesticide use regimes, each pesticide has varying toxicity, and bee species have different vulnerabilities that need to be considered. The authors continue in saying, “Across North America, the amount, type (active ingredient, systemicity, and toxicity), application method, and environmental persistence of pesticides used in agriculture vary by crop. The combination of these defines the bee toxic load unique to each crop.”

The main objectives of this study are to:

1. characterize and compare the relative potential risk experienced by wild bee species that are associated with agriculture in North America using (a) suites of bee traits to quantify species’ intrinsic vulnerability and (b) extrinsic factors based on the toxic load of crops to bees and the strength of each species’ association with those crops; and
2. describe the relative influence of extrinsic factors and intrinsic traits on calculated relative risk to bees from agricultural pesticides.

This was achieved by combining “both the Bee Vulnerability Score (BVS) and the Crop Association-Weighted Toxic Load for each bee species to calculate the relative potential risk experienced by that species from its association with agriculture” that then allows the researchers to rank the species by their potential risk score. Higher point values correspond to higher risk of exposure and/or susceptibility to agricultural pesticides.

The available data that was utilized in this study focus on crops for alfalfa, almond, apple, blueberry, cherry, corn, cotton, cucumber, eggplant, melon, pear, peppers, plum, potato, pumpkin, raspberry, soybean, strawberry, sunflower, tomato, and watermelon; these crops are known to be associated with 713 total bee species.

The authors find, “The highest BVS was shared by bee species that were small, ground nesting, solitary, and with crop specialization,” which includes the species Andrena melanochroa, Panurginus atramontensis, and Pseudopanurginus albitarsis. The lowest scores were found in Bombus species. Within the 90th percentile for the vulnerability scores, the researchers note that, “Five families (Andrenidae, Apidae, Colletidae, Halictidae, Megachilidae), 20 genera, and 60 bee species were represented.”

From the results, the authors point out that, “Importantly, species that are commonly used as models to assess the effects of pesticides on wild bees, like Bombus impatiens, Megachile rotundata, and Osmia species, all exhibited BVS below the median.”

The data also reveals that corn, peppers, potato, raspberry, and cherry have the highest toxic loads per crop for bees. As the authors say, “The high toxic loads of corn and peppers can be explained by the intensive use of bee toxic insecticides, including pyrethroids (e.g., cyfluthrin, zeta-cypermethrin, cypermethrin, bifenthrin) and organophosphates (chlorpyriphos) in corn, and neonicotinoids (e.g., imidacloprid, clothianidin, thiamethoxam, dinotefuran), pyrethroids (e.g., zeta-cypermethrin, cyfluthrin, permethrin, bifenthrin, lambda-cyhalothrin), and organophosphates (naled) in peppers… The intensive use of neonicotinoid (e.g., imidacloprid, thiamethoxam) and pyrethroid (e.g., zeta-cypermethrin, bifenthrin) insecticides in potato, raspberry, and cherry also contributes significantly to the high toxic loads of these crops.”

The researchers establish that extrinsic factors with environmental exposures are more strongly associated with risk to bees than intrinsic traits within the species. Extrinsic factors, such as bees not only foraging from crops but also nesting or overwintering in soils, can increase their pesticide exposure. Life history traits can vary greatly between species, “showing differences in phylogeny, nesting behaviour, sociality, size, reproductive strategies, phenology, larval provisioning strategies, diet breadth, and ability to detoxify pesticides.” All factors must be considered for risk assessments as studying a single species with a certain subset of traits is not representative of all bees.

Species that can detoxify pesticides more efficiently than others are less vulnerable to the effects of pesticides they are exposed to. For species that do not have this ability, such as “some species of the Megachile genus [that] lack important detoxification genes found commonly in other bee groups,” they have substantially higher sensitivity. This includes Megachile rotundata, which the authors identify in the top 10% for potential risk.

The authors highlight that, “Pesticide exposure for bees visiting treated perennial crops such as orchards may be higher than in annual crops because there is no crop rotation and no soil tillage, meaning that persistent pesticides are likely to accumulate in soil. Conversely, annual crops planted with pesticide-treated seeds may also bear higher risk to wild bees because some highly bee toxic neonicotinoid insecticides (e.g., imidacloprid, thiamethoxam) are often applied in this manner and because potential exposure of ground-dwelling bees to pesticide residues in the soil is high. Systemic pesticides can travel into nectar and pollen from their points of application, and persistent chemicals can accumulate in soil, causing an increase in toxic load over time that is especially relevant to bees that nest or overwinter in the ground.”

Pesticides used as seed coatings were not included in the data for this study, and as such the “reported toxic loads for alfalfa, corn, cotton, cucurbits, eggplant, pepper, potato, soybean and sunflower (i.e., crops that may be grown using treated seeds) may be well below their true toxic load and should be understood in that light,” the authors say.

This study helps to highlight relative potential risk, not absolute risk, for the varying bee species currently not considered by EPA when performing risk assessments even though hundreds of species are exposed to pesticides through foraging and nesting behaviors within North America. For a more representative risk assessment process, the authors “suggest an approach that combines information about intrinsic suites of bee traits that define a species’ vulnerability to pesticides with extrinsic factors such as the toxic load born by crops and the strength and breadth of a species’ association with those crops that define potential environmental risk for that species in agroecosystems.”

The researchers hope that this study “can empower stakeholders to (1) prioritize research efforts towards studying species or groups identified as being at highest risk, (2) address environmental factors contributing to risk generally, (3) tailor management practices in specific crops to mitigate risks effectively, (4) design conservation plans for agriculture, and (5) inform future risk assessment protocols, particularly by highlighting bee species or groups that exhibit the highest vulnerability based on their unique traits.”

While the risk assessment process for toxic pesticides is lacking and needs improvement, a better solution exists with organic land management. The holistic approach with organic practices provides a healthy alternative to the detrimental effects of chemicals that pollute the environment and all organisms within it. Protecting all bee species, as well as other pollinators, from pesticides is crucial to agricultural and economic productivity, as well as food security. Take action to advance organic, sustainable, and regenerative practices and policies and be part of the organic solution by becoming a member of Beyond Pesticides today.