Court: District Court, M.D. North Carolina; March 17, 2014; Federal District Court
Plaintiff's Motion for Summary Judgment concerning the Plant Modernization Program Claims against Duke Energy seeks rulings on remaining claims related to alleged violations of the Clean Air Act (CAA), particularly the Prevention of Significant Deterioration (PSD) provisions. The court outlines the standard for summary judgment, emphasizing that it should be granted when there is no genuine dispute over material facts affecting the case outcome.
The case, ongoing for fourteen years, arises from the United States alleging that Duke Energy violated PSD provisions by modifying coal-fired power plants without necessary permits after placing them in Extended Cold Shutdown (ECS) as part of its Plant Modernization Program (PMP). Thirteen plants are still in contention.
The PSD provisions, added to the CAA in 1977, require that operators of certain pollution sources obtain permits from the EPA before modifying their facilities. The Plaintiff contends that Duke unlawfully modified existing plants without these permits, rather than constructing new ones. The definition of "modification" under the CAA includes any physical or operational change leading to increased emissions. Significant modifications trigger the requirement for a permit only if they result in a significant net emissions increase. The court must determine whether Duke's changes constitute such modifications and whether they meet the criteria for requiring a permit under the CAA. The Plaintiff argues that Duke's actions fall within the definition of modification, warranting summary judgment. The court will evaluate both the physical and operational changes alleged in the PMP.
The term "physical change" lacks a specific regulatory definition, prompting courts to adopt a broad interpretation. In this context, a program requiring physical alterations at power plants qualifies as a "physical change." However, exceptions exist for changes deemed "routine maintenance, repair, and replacement" (RMRR), which do not necessitate PSD permitting. Duke contends that its program modifications fit within the RMRR exception, asserting a genuine issue of material fact, while the plaintiff disputes this claim.
Duke bears the burden of proof to demonstrate that its changes were routine maintenance, with the court responsible for determining the "routine" nature of those changes as a legal question. Expert opinions are relevant but not decisive in this analysis. The court utilizes the multi-factor WEPCo test to assess the modifications based on (1) nature and extent, (2) purpose, (3) frequency, and (4) cost. The court has previously established that these factors should be evaluated in relation to the industry as a whole, while also considering specific practices at individual units. The court adopts this approach, integrating both industry standards and specific unit assessments to determine the applicability of the RMRR exception.
In 1983, Wisconsin Electric Power Company (WEPCo) identified significant performance declines in its aging power plants, built between 1935 and 1950, necessitating extensive renovations to continue operations. The utility's "life extension" program aimed to allow these plants to function until 2010, addressing critical issues such as steam drum cracking and air heater deterioration. The rehabilitation included repairing or replacing turbine-generators, boilers, and other essential equipment, with each of the five units undergoing repairs over nine-month periods.
In the current case, after analyzing the evidence and arguments, the court determined that factual disputes exist regarding the physical changes made under Duke’s Planned Maintenance Program (PMP), warranting a trial rather than granting summary judgment. The Plaintiff contends that the PMP closely resembles WEPCo’s life extension project in terms of scale, required downtime, and costs, asserting that the PMP involved significant alterations that were not routine maintenance. The Plaintiff provided a detailed list of modifications made to each unit, highlighting repairs and upgrades, and argued that the PMP's cost, which ranged from $17 to $30 million per unit, equaled or exceeded the original construction costs.
Conversely, Duke disputes the Plaintiff's assertions, arguing that not all changes fell under the PMP and that the work at each unit comprised separate projects with distinct justifications. Duke claims that the Plaintiff's aggregation of work leads to inflated assessments of the PMP's scope, duration, and expenses.
Duke conducted an "upgrade and reliability study" at Allen Unit 1 from January to May 1985, involving a comprehensive team to identify necessary refurbishments for life extension. Duke communicated to the NCUC that the boiler required modifications, new feedwater heaters, turbine rotor repairs, and generator rotor reinsulation. The initial budget for renovations was $21,670,000, covering extensive installations and repairs, with costs expanding by 1989 to include an additional $3,673,000 for boiler repairs. Significant upgrades included replacing sections of the reheater, burner panels, and control systems, leading to an outage lasting over six years until August 1990, culminating in a final cost of $23,619,349, compared to the original $18 million construction cost in 1957.
Duke’s expert distinguished between boiler repairs and other maintenance activities, citing a report from 1987 that outlined the boiler's issues, indicating that work commenced thereafter. The expert suggested that the consolidation of repairs allowed efficient use of resources over time. Duke utilized the ECS to perform maintenance during off-peak periods, resulting in a prolonged project duration compared to typical turbine outages. Duke argued that the extended outage was due to system demand rather than the volume of work performed and that the ECS period allowed for both corrective and predictive maintenance. The court acknowledged genuine disputes regarding the work's nature, extent, purpose, frequency, and cost, leading to the denial of Plaintiff's motion for summary judgment on the issue of physical change.
Plaintiff contends that each PMP unit experienced an operational change upon renovation and resumption from an extended non-operational period to full operation. Duke counters this argument by asserting that Plaintiff waived it due to a lack of prior notice and that the argument lacks merit regardless of its timing. The crux of Duke's notice argument hinges on whether the initial Notice of Violation (NOV) referenced "operational change" as a basis for the lawsuit. Under the Clean Air Act (CAA), the EPA must notify an entity of violations before enforcement actions can be initiated, and while the EPA's lawsuit must align with the specific violations cited in the NOV, courts generally interpret the sufficiency of the NOV liberally. The CAA does not mandate a specific format for notice; actual notice suffices. In this case, the EPA's NOV indicated that Duke was undertaking modifications to extend the operational life of its coal-fired power plants without the necessary permits. Although the NOV focuses on physical modifications, the CAA recognizes both physical and operational changes as modifications. Therefore, the term "modification" in the NOV provides adequate notice to Duke that the alleged violations could arise from either type of change.
On the merits, to succeed in summary judgment, Plaintiff must demonstrate that Duke's shutdown and subsequent restart of its plants constitutes an operational change under PSD regulations. Plaintiff argues that both the language of the regulations and EPA guidance support the notion that restarting a long-idled power plant qualifies as an operational change necessitating PSD permitting. However, the court remains unconvinced by the broad conclusions drawn from past EPA determinations and finds the existing factual record insufficient to resolve this issue at the summary judgment stage.
The regulations define "operational change" negatively, outlining exceptions that do not necessitate a permit, including the "mere increase in hours" exception. This exception asserts that simply increasing operational hours does not qualify as a "physical change" or "change in the method of operation." The rationale behind this is to allow utilities to adjust production in response to market demands without undue disruption. However, if an increase in operating hours is linked to a physical change or method change, the exception may not apply. The interpretation suggests that merely maximizing plant capacity does not constitute a major modification, but a physical change enabling increased output would nullify the exclusion for longer operating hours.
A trial is deemed necessary to ascertain the extent of physical improvements and maintenance at each unit in question, leading to the denial of Plaintiff's summary judgment motion regarding whether the shutdown and restart of units signified an operational change. The court finds the agency decisions cited by the Plaintiff, namely Cyprus Casa Grande and In re Monroe Electric, insufficient to alter its ruling. In Cyprus Casa Grande, the EPA noted that a power unit in non-operational status for ten years could not invoke the "increase in hours" exclusion, which, while potentially applicable to Duke's units, differed in critical respects, such as maintaining operating permits and inclusion in the state emissions inventory. The Monroe Electric case, where a plant restart after an extended shutdown was viewed as an operational change, is also not compelling at this stage.
The EPA determined that the "increase in hours" exception was inapplicable as the plant transitioned from a "non-operational" and "unmanned" state to "fully operational." The agency concluded that the decision to restart the plant, despite being influenced by market changes, did not reflect the rapid decision-making that the exception was designed to address. Additionally, the EPA noted that while Duke maintained environmental permits for the Monroe plant, it did not purchase or update the necessary permits for the restart. The State of Louisiana treated the plant as having no environmental impact, which affected its emissions inventory and reporting to the Ozone Transport Assessment Group (OTAG). If the Monroe plants had been included, Louisiana would have had to report higher ozone emissions, potentially altering OTAG’s conclusions. The court confirmed the finding of operational change, indicating that a restart would disturb previous environmental impact assessments. It acknowledged the need for a legal determination on operational change before trial, requiring a detailed analysis of relevant case law. Consequently, the court denied the Plaintiff's motion for summary judgment on operational change, deferring the issue to trial. Regarding "significant net emissions increase," the regulations define this as exceeding zero in actual emissions from physical or operational changes. A net emissions increase is considered "significant" if it meets or surpasses regulatory emissions rates for the relevant pollutant.
A significant net emissions increase is determined using an "actual-to-projected-actual" test, which compares pre-PMP actual emissions to projected post-PMP emissions. "Actual emissions" are defined as the average annual emissions over a two-year period preceding a specific date that reflects normal operations, although a different period may be allowed if it better represents normal operations. The Plaintiff contends that for long-shutdown sources, the emissions baseline should be zero, effectively interpreting "normal source operation" as "non-operation." Duke counters this by asserting that the relevant two-year period should be the two years before any unit shutdown. The court has previously noted that the North Carolina Department of Environmental Quality (NC DENR) did not formally endorse Duke’s interpretation, and even if it had, the Plaintiff's conflicting interpretation would take precedence, thereby deserving deference. The current issue for the court is whether a zero emissions baseline applies legally for each plant in the "actual-to-projected-actual" test. The court concludes that it does.
Baseline emissions for each unit of Duke Energy were determined to be zero. The court referenced a previous memorandum indicating that deviations from using the two years prior to restart were limited to exceptional circumstances, such as strikes or major accidents. Duke's argument for an alternative two-year period was not supported by any regulatory or judicial announcement. For units out of service for around ten years, including Buck Units 3 and 4, and Cliffside Units 1 and 2, the court ruled that a zero emissions baseline was appropriate based on the precedent set in Casa Grande, which also had zero emissions during its shutdown. In contrast, units out of service for approximately three years, like Riverbend Unit 4 and Dan Unit 3, and those shut down for five to seven years, including Allen Units 1 and 2, were still subject to the presumption of using the two years prior to restart. The court distinguished this case from WEPCo, where an alternate baseline was used for units that had been out of service for short periods due to mechanical issues. Duke's units were placed in an idle status for lack of demand, not mechanical failures. Consequently, the court upheld that a zero-emissions baseline applies to all Duke units, adhering to the default rule of using the two years immediately preceding their restart.
A zero-emissions baseline has been established as a legal standard for Duke’s plants, leading to the conclusion that any emissions post-restart constitute a significant net emissions increase. A "net emissions increase" is defined as any rise in actual emissions resulting from changes in operation at a stationary source, as per 40 C.F.R. § 51.166(b)(3). The subsequent analysis focuses on whether these emissions are considered "significant" under the regulations, which define "significant" as any increase in emissions that meets or exceeds specified thresholds for pollutants. For sulfur dioxide and nitrogen oxides, the threshold is set at 40 tons per year. Evidence shows that post-change emissions for each PMP unit exceeded this limit. Duke does not dispute the existence of a significant net emissions increase and acknowledges through its own expert that projected emissions for these pollutants far surpass the 40 tons per year threshold. Duke’s arguments primarily challenge the zero-emissions baseline rather than disputing the emissions data presented by the plaintiff. Consequently, the court concludes that there is no disagreement regarding the emissions exceeding PSD thresholds, warranting summary judgment in favor of the plaintiff.
Plaintiff has failed to establish a causal link between the operational changes and the significant net emissions increase. The court must assess whether a restart constitutes a physical or operational change and if this change leads to a significant increase in regulated pollutant emissions, as outlined in 40 C.F.R. § 51.166(b)(2)(i). Previous court proceedings regarding a motion in limine highlighted a factual dispute regarding the operability of Duke's plants before the changes. This dispute is crucial for determining causation; if the plants were operable, demonstrating that emissions increases were due to the operational change would be more challenging for the Government. Conversely, if the plants were inoperable, Plaintiff could potentially infer causation more easily. The court acknowledges that Duke has produced sufficient evidence to create a legitimate dispute about the plants' operability. The selected plants for ECS were operationally viable but economically unfavorable compared to newer models. Evidence suggests that these plants could generate substantial power and had operational capacity prior to ECS. Given the ongoing dispute over the plants' operability and its implications for emissions causation, the court denies Plaintiff's motion for summary judgment on this issue. It grants the motion in part, confirming that the restart led to a significant net emissions increase but denies it concerning the physical or operational changes and their causative effects. The court also incorporates factual background from its earlier memorandum regarding motions in limine and summary judgment. The terminology regarding power-generating units and plants is used interchangeably for clarity.
Duke argues that a determination from the North Carolina Department of Energy and Natural Resources (NC DENR) supports its claim that a temporary shutdown and restart of its units would not require a Prevention of Significant Deterioration (PSD) permit. This argument references correspondence between Duke and NC DENR from 1983. However, the court has previously ruled that Duke's descriptions of the planned work were inadequate and that failing to notify NC DENR of the planned maintenance project (PMP) undermines Duke's position. Even if NC DENR provided some interpretation, the court asserts that the Environmental Protection Agency (EPA) interpretation of State Implementation Plan (SIP) regulations prevails, especially when the state SIP adopts federal regulations without modification.
PSD regulations necessitate that utilities obtain a pre-construction permit when emissions are expected to rise above the actual average from the previous two years. The critical issue is whether Duke should have anticipated increased emissions post-restart, which would have necessitated a permit prior to executing the PMP. The court references a precedent where a different two-year period for measuring emissions was used due to specific maintenance circumstances, but finds that Duke's attempt to classify the PMP under a routine maintenance exception does not support this approach. Additionally, while a case involving a shorter operational history of the units might suggest a different analysis, Duke's much longer operational history does not provide sufficient grounds for a different two-year baseline in this scenario.
