895 F.3d 1
D.C. Cir.2018Background
- EPA promulgated the "Brick/Clay Rule" (2015) regulating hazardous air pollutants from brick, structural clay, and clay ceramic kilns, using MACT standards for metals and health-threshold standards for acid gases.
- Environmental petitioners (Sierra Club, NRDC) challenged EPA’s use of health thresholds for acid gases, EPA’s MACT floor methodology (upper prediction limit, UPL), and allowance of alternative emissions floors for brick kilns.
- Industry petitioners (Brick Industry Assoc., Tile Council, Kohler) challenged EPA’s MACT calculations for specific subcategories, data choices, standing, and use of certain data sources (including a reactivated scrubber at Kohler’s Kiln 10).
- Court applied Chevron and arbitrary-and-capricious review under the CAA; consolidated multiple petitions for review of the final rule and EPA’s partial denial of reconsideration.
- Court denied Industry petitions in full, granted Environmental petition in part, and remanded to EPA for further proceedings on specified issues.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Use of health thresholds for acid gases (cancer risk) | EPA wrongly concluded acid gases are noncarcinogenic without sufficient proof | EPA relied on existing toxicity assessments and absence of carcinogen classifications | EPA acted unreasonably; remand granted (EPA failed to show cancer risk absence) |
| Noncarcinogenic thresholds (HCl, HF reference levels) | EPA relied on low-confidence HCl RfC and declined CA EPA benchmark; HF acute exposures not adequately prevented | EPA said low-confidence benchmarks can be used and modeling assumptions are conservative | EPA acted unreasonably as to HCl and HF thresholds; remand granted |
| Ample margin of safety for health-based limits | CAA requires an "ample margin of safety" added to thresholds | EPA built margin via conservative modeling assumptions and worst-case scaling | EPA failed to show an ample margin of safety; remand granted |
| Use of Upper Prediction Limit (UPL) to set MACT floors | UPL application to limited datasets and ad hoc adjustments may be arbitrary | UPL is a permissible statistical tool; EPA explained its use for limited datasets | UPL use generally upheld for limited datasets, but EPA must better explain five specific ad hoc adjustments; remand limited to those five floors |
| Provision of alternative emissions floors for brick kilns | Allowing multiple alternative "best" metrics lets sources avoid a MACT floor based on the category's best performers, violating CAA | EPA has discretion to choose the metric and to offer compliance flexibility | Court held alternative floors unlawful; EPA must revise (grant) |
| Use of synthetic minor sources in MACT data (brick PM/nonmercury) | Including synthetic minor sources improperly mixes area sources with majors | EPA reasonably interpreted "potential to emit considering controls" to include synthetic minors | Inclusion was permissible; petition denied |
| Reliance on anomalous PM data / exclusion of fabric-filter tests | EPA acted arbitrarily by keeping unexpectedly low-emitting non-fabric-filter data without resolving anomalies | EPA investigated, rejected unreliable fabric-filter tests, requested more data, and reasonably used best-12% across all controls | Court upheld EPA’s approach; petition denied |
| Kohler Kiln 10 data (representativeness) | EPA erred by using emissions from Kiln 10 when scrubber was not in ordinary operation | EPA reasoned emissions with scrubber were "achieved in practice" during prior operating period and could be tested later | Court upheld EPA’s use of Kiln 10 data; petition denied |
| Tile Council standing to challenge Tile NESHAP | Tile Council challenged listing and beyond-the-floor analysis | EPA noted no member was a major source or imminently so | Court found Tile Council lacked standing; petition dismissed |
Key Cases Cited
- Sierra Club v. EPA, 479 F.3d 875 (D.C. Cir. 2007) (vacating prior 2003 Rule and framing floor-setting requirements)
- NRDC v. EPA, 529 F.3d 1077 (D.C. Cir. 2008) (discussing problems with health-based standards)
- NACWA v. EPA, 734 F.3d 1115 (D.C. Cir. 2013) (critiquing EPA’s explanation of UPL predicting best performers)
- U.S. Sugar Corp. v. EPA, 830 F.3d 579 (D.C. Cir. 2016) (upholding EPA’s UPL explanation with caveats for small datasets)
- White Stallion Energy Ctr., LLC v. EPA, 748 F.3d 1222 (D.C. Cir. 2014) (agency not required to resolve every scientific uncertainty before regulation)
- Cement Kiln Recycling Coal. v. EPA, 255 F.3d 855 (D.C. Cir. 2001) (MACT floors must reflect what best performers actually achieve)
- Nat’l Lime Ass’n v. EPA, 233 F.3d 625 (D.C. Cir. 2000) ("in practice" requirement for emissions data)
- Med. Waste Inst. & Energy Recovery Council v. EPA, 645 F.3d 420 (D.C. Cir. 2011) (use of data from sources operating with controls to set floors)
- Lujan v. Defenders of Wildlife, 504 U.S. 555 (U.S. 1992) (standing requires actual or imminent injury)
- Hunt v. Wash. State Apple Advert. Comm’n, 432 U.S. 333 (U.S. 1977) (associational standing requirements)