Pollinator Stewardship Council v. U.S. Environmental Protection Agency, 806 F.3d 520 (9th Cir. 2015)
In this case, the Pollinator Stewardship Council (Pollinator) petitioned the Ninth Circuit for review of a pesticide registration promulgated by the United States Environmental Protection Agency (EPA). Pollinator challenged EPA’s unconditional registration of sulfoxaflor under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA). The Ninth Circuit concluded that EPA had based its unconditional registration of sulfoxaflor on insufficient data, thus failing to meet the substantial evidence standard required by FIFRA. In addition, because a failure to vacate EPA’s unconditional registration of sulfoxaflor presented a threat of environmental harm and because EPA could arrive at a different decision regarding the registration on remand, the court vacated the registration and remanded to EPA.
Under FIFRA, a pesticide may not be sold or used without EPA’s approval and registration. The manufacturer of the pesticide must submit an application to EPA describing the manner in which the pesticide will be used, its benefits, its ingredients, and a description of studies conducted regarding, among other things, the pesticide’s environmental effects. EPA may then take one of three actions: 1) deny the application if denial is “necessary to prevent unreasonable adverse effects on the environment,” 2) conditionally register the pesticide when there is “insufficient data to evaluate the environmental effects of [the] pesticide” for “a period reasonably sufficient for the generation and submission of required data,” or 3) unconditionally register the pesticide if there is sufficient data to evaluate the environmental risks.
In making its decision, EPA follows a multi-tier framework called the Pollinator Risk Assessment Guidance for assessing pesticides affecting pollinators, which was established in response to concerns about the rapid decline in bee populations. Under Tier 1, EPA determines the level of risk to bees by comparing the dose at which half the tested bees die (the acute median lethal dose) with the likely concentration of the pesticide in the environment should EPA approve the pesticide. If the resulting figure (the risk quotient) is above the level of concern (LOC), then EPA moves on to Tier 2. If the risk quotient is below the LOC, EPA will approve the pesticide. For bees, EPA set the LOC at 0.4. Under Tier 2, EPA measures the effect of the proposed pesticide on bees by analyzing data from “semi-field” studies. Although there is a Tier 3, EPA did not reach Tier 3 in this case.
In 2010, Respondent-Intervenor Dow Agrosciences LLC (Dow) submitted an application to EPA to register three different products containing sulfoxaflor at a maximum application rate of 0.133 lb a.i./A. EPA analyzed sulfoxaflor within the Pollinator Risk Assessment Framework. In its Tier 1 analysis, EPA found that the risk quotient for oral exposure fell between 0.8 and 5.7 depending upon the type of bee, a range greater than the 0.4 LOC.
Because the risk quotient of sulfoxaflor at the suggested application rate was greater than the LOC, EPA moved onto Tier 2 and analyzed the semi-field studies Dow had submitted with its application. These studies confine bees to a tunnel enclosure containing one type of crop sprayed with sulfoxaflor. At the outset, the court noted that these studies generally have several shortcomings: 1) during these studies, bees die at an above-average rate due to study-induced stress; 2) the measured effect is overstated because a bee’s natural diet consists of more than just one type of crop; and 3) the studies are of limited duration and therefore fail to capture long-term adverse effects. In this case, the set of studies submitted by Dow had a further shortcoming: all but one of the six tests failed to use the proposed maximum application rate of 0.133 lb a.i./A. In addition, the one test that used the proposed maximum application rate, the Ythier 2012 study, failed to provide conclusive information because the crop chosen, cotton, was a suboptimal source of pollen, and the study was designed to measure pesticide residue on the cotton rather than the biological effects of sulfoxaflor on the bees. As a result, EPA concluded that the effect of sulfoxaflor on bee mortality, flight activity, and behavioral abnormalities was unknown, and the effect of sulfoxaflor on brood development and long-term colony health was inconclusive.
Due to the gaps in data at Tier 2, EPA’s environmental risk assessment concluded that one or more additional Tier 2 semi-field studies were required for Dow’s application to be approved. Based on this finding, in January 2013, EPA conditionally registered sulfoxaflor at a lower maximum application rate of 0.09 lb a.i./A. In order for EPA to register sulfoxaflor at the higher application rate of 0.133 lb a.i./A, the additional studies would need to assess the impact of sulfoxaflor on brood development and long-term colony strength, as well as use a pollinator-attractive crop.
Despite the record being void of additional studies, in May 2013 EPA unconditionally registered sulfoxaflor at the 0.09 lb a.i./A application rate. This unconditional registration included various mitigation measures, including: 1) longer minimum intervals between applications, 2) certain crop-specific restrictions on blooming crops, and 3) a required warning label for continuously blooming crops. In explaining its final decision, EPA concluded that while there is a potential hazard to bees from sulfoxaflor, the hazard would be mitigated by the adopted mitigation measures. Furthermore, EPA found that application of sulfoxaflor according to the label would not cause unreasonable adverse effects to bees, and that the benefits of application would outweigh the costs.
After EPA filed its final decision to unconditionally register sulfoxaflor, Pollinator filed for review in the Ninth Circuit. Pollinator claimed that EPA’s decision violated FIFRA because the decision was not supported by substantial evidence in the record. EPA and Dow responded that the limited studies provided enough data for EPA to make its registration decision and that EPA has the flexibility to determine what type of data is needed to support its registration decisions. The Ninth Circuit reviewed EPA’s decision under the substantial evidence standard as required by FIFRA.
In determining whether EPA had violated FIFRA, the court analyzed the Tier 2 studies originally submitted by Dow to determine if there was substantial evidence that a maximum application rate of 0.09 lb a.i./A would not have an unreasonable adverse effect on bees. The court resorted to these studies because EPA and Dow did not submit additional studies to the record. In its review of studies relating to the biological effect of sulfoxaflor on bees, the court noted that only two of the six Tier 2 studies applied sulfoxaflor at the 0.09 lb a.i./A rate: the Ythier 2012 study and the Hecht-Rost 2009 study. The Ythier 2012 study failed to give sufficient data because the study was designed to quantify plant residues rather than measure the biological effect of sulfoxaflor on bees. The Hecht-Ross 2009 study also failed to give sufficient data due to a pest infestation, a long preexposure period in the tunnels resulting in higher bee stress, a short observation period, and a lack of bee larvae in the tunnels.
The court next evaluated the studies that measured the effects of sulfoxaflor on brood development and long-term colony health. The court found that only two of the Tier 2 studies measured brood termination rates and both studies used less than half the EPA-approved 0.09 lb a.i./A application rate. In addition, these studies also had high brood termination rates in the control tunnels when compared to studies done according to OECD guidelines, suggesting the controls were not appropriate measuring sticks. The court also found that the three studies measuring colony strength failed to use more than half the EPA-approved 0.09 lb a.i./A application rate. Further limiting their reliability, these studies measured colony strength only 7 to 17 days after sulfoxaflor was applied, which likely failed to account for negative effects that could only appear over a longer time period.
These above deficiencies formed the basis for EPA’s original conclusion in January 2013 that only conditional registration of sulfoxaflor was appropriate. Because the court found that these deficiencies still existed under the 0.09 lb a.i./A application rate, EPA’s decision to unconditionally register sulfoxaflor was unsupported by substantial evidence that sulfoxaflor would not have unreasonable adverse effects on bees. EPA argued that these deficiencies and the inconclusiveness of the Tier 2 studies favored unconditional registration because the inconclusiveness itself demonstrated that sulfoxaflor does not cause unreasonable adverse effects on bees. The court rejected this argument, noting that an agency cannot rely on ambiguous studies as evidence of a conclusion not supported by those studies. EPA also argued that applying sulfoxaflor at the 0.09 lb a.i./A application rate did not create a LOC. Citing to the studies the court rejected this argument. The court noted that while only two of the sixty-six nectar measurements and one of the sixty-six pollen measurements in the Ythier 2012 study were above the LOC; EPA is required by regulation to mandate pollinator field-testing whenever the LOC is exceeded. Furthermore, the court found that EPA must follow its own regulations, even when those measurements were close to being at a level of no concern.
Having decided that EPA failed to meet the substantial evidence standard of review, the court considered whether to vacate the unconditional registration of sulfoxaflor in remanding to EPA. The court noted that vacating EPA’s final decision requires the court to decide whether doing so will result in possible environmental harm, as well as the likelihood that EPA could promulgate the same decision on remand. Because the court concluded that failure to vacate posed a potential risk of causing environmental harm, and that EPA might reach a different decision based on additional Tier 2 studies, the court vacated the unconditional registration of sulfoxaflor.
In sum, the court held that (1) EPA had based its decision to unconditionally register sulfoxaflor on insufficient data and that the decision was therefore unsupported by substantial evidence, and (2) leaving the unconditional registration of sulfoxaflor in effect on remand would pose a risk of environmental harm. Accordingly, the court vacated the registration and remanded to EPA for additional studies.
- Petitioners included American Honey Producers Association, National Honey Bee Advisory Board, American Beekeeping Federation, Thomas R. Smith, Bret L. Adee, and Jeffery S. Anderson.↵
- Dow Agrosciences LLC was a respondent-intervenor in opposition to Pollinator’s petition.↵
- Federal Insecticide, Fungicide, and Rodenticide Act, 7 U.S.C. §§ 136–136y (2012).↵
- Id. § 136n(b).↵
- Id. § 136a(a).↵
- Id. § 136a(c).↵
- Id. § 136a(a).↵
- Id. § 136a(c)(7)(C).↵
- Id. § 136a(c)(5).↵
- U.S. Envtl. Prot. Agency, Guidance for Assessing Pesticide Risks to Bees (2014), available at https://www.epa.gov/sites/production/files/2014-06/documents/pollinator_risk_assessment_guidance_06_19_14.pdf.↵
- U.S. Envtl. Prot. Agency, How We Assess Risks to Pollinators, https://www.epa.gov/pollinator-protection/how-we-assess-risks-pollinators#data (last visited July 16, 2016).↵
- U.S. Envtl. Prot. Agency, supra note 95, at 7.↵
- Id. at 24.↵
- “Lb a.i./A” stands for pounds of active ingredient per acre.↵
- FIFRA, 7 U.S.C. § 136n(b) (2012).↵
- 40 C.F.R. § 158.630(d), (e) n.25 (2015).↵
- See Nat. Res. Def. Council v. U.S. Envtl. Prot. Agency, 735 F.3d 873, 883–84 (9th Cir. 2013).↵
- Idaho Farm Bureau Fed’n v. Babbitt, 58 F.3d 1392, 1405–06 (9th Cir. 1995).↵
- North Carolina v. U.S. Envtl. Prot. Agency, 531 F.3d 896, 929 (D.C. Cir 2008); Allied-Signal, Inc. v. U.S. Nuclear Regulatory Comm’n, 988 F.2d 146, 151 (D.C. Cir. 1993).↵