Like many environmental laws, the Clean Water Act (CWA) includes a “permit shield.” Under this provision, the holder of a NPDES permit is shielded from both agency enforcement and citizen suits, provided—and this is key—the permittee complies with permit terms. The battle is typically not whether a permit condition has been violated. That part of the analysis is relatively straightforward. The real war is over whether the permit shields a permittee for pollutants actually discharged, which are known by agencies to be present in the discharge, but not specifically incorporated into a permit limit or condition.
In the last two years, the Fourth, Seventh, and Ninth Circuits have announced an unsatisfying array of legal conclusions that unnecessarily complicate the availability and scope of the CWA permit shield. But when the unique state- and permit-specific issues in these recent cases are put into perspective, the permit shield remains a key defense under the CWA. The most recent decision of the Sixth Circuit affirms the full scope of the shield for pollutants disclosed by the applicant but not limited in the permit, and for general permits to the same degree as for individual permits.
The language of the CWA permit shield provision is simple enough: “[c]ompliance with a permit issued pursuant to this section shall be deemed compliance” for purposes of enforcement and citizen suits involving certain effluent limits, performance standards, and ocean discharges, but not toxic pollutants. See 33 U.S.C. § 1342(k). As explained by the U.S. Supreme Court, the permit shield’s purpose is “to insulate permit holders from changes in various regulations during the period of a permit and to relieve them of having to litigate in an enforcement action the question whether their permits are sufficiently strict.” E.I. du Pont de Nemours & Co. v. Train, 430 U.S. 112, 138 n. 28 (1977).
In the seminal case, Piney Run, the Fourth Circuit crafted the legal test defining the availability of the CWA permit shield. Piney Run Pres. Ass’n v. Cnty. Comm’r, 268 F.3d 255 (4th Cir. 2001). In considering whether a permit holder may continue to discharge an unlisted pollutant, Piney Run held that a NPDES permit will shield subsequent enforcement if (1) the permit holder complies with the express terms of the permit and the CWA’s permit application requirements and (2) the permit holder does not make a discharge not within the “reasonable contemplation” of the agency when the permit was issued. Id. at 259.
The Piney Run test was not new. In 1995, U.S. EPA observed that: “[a] permit provides authorization and therefore a shield for . . . pollutants resulting from facility processes, waste streams and operations that have been clearly identified in the permit application process when discharged from specified outfalls.” EPA, Revised Policy Statement on Scope of Discharge Authorization and Shield Associated with NPDES Permits. In 1998, U.S. EPA’s Environmental Appeals Board concluded that when a permittee makes “adequate disclosures” in a NPDES permit application, unlisted pollutants may be shielded even if they are not specific permit conditions. See Ketchikan Pulp Co., 7 E.A.D. 605, 621 (EAB May 15, 1998).
Despite its apparent simplicity, the Piney Run two-part test continues to generate an assortment of interpretations. In several cases filed by the Ohio Valley Environmental Coalition (OVEC) against coal companies, the central issue was whether a NPDES permit shields violations of water quality standards (WQS) where those WQS were incorporated by reference into the permit. See OVEC v. Alex Energy, Inc., 2014 U.S. Dist. LEXIS 44668 (S.D.W. Va. Mar. 31, 2014); OVEC v. Fola Coal Co., 2013 U. S. Dist. LEXIXS 178319 (S.D.W. Va. Dec. 19, 2013); OVEC v. Elk Run Coal Co., U.S. Dist. LEXIS 509 (S.D.W. Va. Jan. 3, 2014). Of these cases, the most important is OVEC v. Marfork Coal Co., which involved claims of unpermitted selenium discharges at coal mines. OVEC v. Marfolk Coal Co., 966 F. Supp. 2d 667 (S.D.W. Va. 2013).
Under the OVEC cases, if a NPDES permit incorporates WQS—itself a topic worthy of analysis—and the WQS are being exceeded, the shield will not preempt an enforcement action or a citizen suit, even if there is no specific permit condition related to the substance of the WQS. For the court, because the NPDES permit incorporated WQS, a WQS violation equaled a permit violation.
Even if the NPDES permit did not incorporate WQS, the permit shield would not protect a permittee, according to Marfolk. In a fundamental misunderstanding of WQS and permit conditions, the court observed that “[n]ow that sampling has revealed persistently high levels of selenium above the [WQS] during the life of this permit, the amount of selenium actually discharged . . . was not within the reasonable contemplation of the agency at the time of the permit, and consequently not within the permit shield.” Id. at 686. The court failed to appreciate the NPDES permit program’s reliance on a time-tested application process that requires applicants to collect and submit at least one grab sample for a range of different pollutants, but rarely more than one sample. Technically, the kind of discharge variability at issue in Marfork simply would not have been known to either applicant or agency. The court also failed to recognize that the NPDES regulatory scheme provides a process for adding more stringent limits when it is later discovered (e.g., with new data) that the existing limits are inadequate to ensure WQS will be achieved.
In another case involving coal facilities, the Fourth Circuit considered a permittee’s disclosure obligations, a predicate to asserting the permit shield defense. In Southern Appalachian Mountain Stewards v. A&G Coal Corp., environmental groups sued A&G, claiming unpermitted discharges of selenium. 758 F.3d 560 (4th Cir. 2014). Although A&G did not disclose the presence of selenium in its application, A&G asserted the permit shield defense because its NPDES application disclosed coal and coal processing, and selenium was known by the agency to be present in discharges involving coal processing. Rejecting this argument, the district court zeroed in on the disclosure requirements in the permit application, which required A&G to identify a number of pollutants, including selenium, as “Believed Present” or “Believed Absent.” Id. at 566.
The Fourth Circuit affirmed, finding A&G could not claim “willful ignorance” in the application process and then rely on the permit shield defense. The court’s logic was clear: “There is no question that selenium is a pollutant under the CWA. And there is no question that A&G was required by its NPDES permit application instructions to test for the presence of selenium and by federal and state regulations to, at minimum, report whether it believed selenium to be present or absent.” Id. at 569. For the court, no selenium disclosure meant no defense.
In clarifying the burdens on permittees, the court held that a permittee cannot rely on engineering or process theory—it must, after “appropriate inquiry,” test for all substances to be included behind the shield. For the court, “[s]ilence as to the existence of a referenced pollutant is not adequate.” Id. at 566. A&G’s position was flawed because it focused on an applicant’s knowledge of pollutants rather than the agency’s contemplation of those pollutants.
The court concluded that A&G’s argument would “tear a large hole in the CWA.” Id. at 569. Responding to claims that the “test everything” approach would generate an “endless disclosure of known pollutants,” the court opined that testing was both limited by the requirements of the NPDES application and not too high a price to pay for the significant protections offered by the CWA permit shield.
In a straightforward decision, the Seventh Circuit addressed whether the CWA permit shield protected discharges from mining permits issued in lieu of state NPDES permits. In Wisconsin Resources Protection Council v. Flambeau Mining Co., the Wisconsin Department of Natural Resources (DNR) regulated defendant’s storm water discharges through a mining permit rather than a NPDES permit. 727 F.3d 700, 704 (7th Cir. 2013). Sierra Club argued because copper discharges were being made under a mining permit rather than a NPDES permit, Flambeau’s copper discharges were not covered by the CWA permit shield. Even if Flambeau’s permit was legally invalid, the court found that it could not, consistent with the requirements of due process, impose a penalty on Flambeau for complying with what DNR deemed a valid NPDES permit.
Addressing a different issue, the Ninth Circuit tackled whether the permit shield applied to general stormwater permits under the CWA. In Alaska Community Action on Toxics v. Aurora Energy Services, environmental groups alleged coal fell into Resurrection Bay and nearby waters, either directly or as coal dust, during the over-water transfer of coal from the stockpiles to the ship holds and other activities at the facility—they alleged these non-stormwater discharges were not covered by the Multi-Sector General Permit for Stormwater Discharges Associated with Industrial Activity. 765 F.3d 1169 (9th Cir. 2014). Applying Piney Run, the district court found defendant’s NPDES permit did not expressly allow for coal discharges into the bay, but coal discharges were nonetheless within the scope of the permit’s protection because defendants complied with the general permit and discharges were clearly disclosed to, and reasonably anticipated by, the permitting authority as part of the general permit. The record showed Aurora made full disclosures during the permitting process, followed agency instructions, and adhered to the terms of the permit, which included coal-control measures as part of its stormwater pollution prevention plans.
Focusing on the definition of stormwater discharges and the nature of non-stormwater discharges of coal, the Ninth Circuit reversed, finding the “plain terms” of the General Permit prohibited defendant’s non-stormwater discharge of coal. Because the permit required a permittee to “eliminate non-stormwater discharges not authorized by a NPDES permit,” and because coal was not included in the list of “non-stormwater discharges,” coal discharges were not permitted under the general permit. Id. at 1172. It was that simple for the court.
In the most recent decision involving the permit shield, the Sixth Circuit ruled that a coal company was shielded from liability for discharges of selenium in excess of the state water quality standards under a state NPDES general permit because the state knew at the time it issued the permit that the mines in the area could produce selenium. Sierra Club v. ICG Hazard, LLC, Case No. 13-5086, 2015 U.S. App. LEXIS 1283 (6th Cir. Jan. 27, 2015). The decision stands in stark contrast to the A&G decision, with the primary difference being what was disclosed to, and within the reasonable contemplation of, the state when it issued the permit. In the ICG Hazard case, the company submitted all required disclosure information (including a one-time sample for selenium at some time during the term of the permit, but not necessarily before issuance); in the A&G case, the company did not. Read together, the cases reinforce the importance of full disclosure to the permitting agency. Without such disclosure, the availability of the shield will be in question and susceptible to challenge.
Despite recent dents from three federal circuits, the permit shield remains alive and relatively healthy under the CWA. A&G and ICG Hazard underscore the importance of full disclosure to the permitting agency, one of the fundamental predicates to being able to assert the shield. Judging from recent decisions, it may be better to provide more information with the application—even at the risk of facing more limits and conditions—than to risk enforcement on claims that certain discharges are not shielded from liability. Aurora calls into question whether a general permit is a suitable option for a facility that discharges pollutants not specifically authorized by that general permit. Historically, permittees have been allowed to augment the scope of a general permit’s shield through disclosures and actions made pursuant to the general permit process. But following the logic of Aurora, this may no longer be defensible. It may be safer for a facility to seek coverage under an individual permit, where the scope of the shield may be fashioned entirely out of facility-specific disclosures in the application process.