EPA New Source Review Reform: Possible & Probable Oregon Impacts
By Thomas R. Wood
6/1/2002

The US Environmental Protection Agency (EPA) recently announced that it will be making significant changes to the federal New Source Review program. The final rules are currently undergoing review in Office of Management and Budget and could be out as final rules as early as next month. EPA has caused quite a stir by suggesting that the changes may be viewed as part of the mandatory minimum requirements for local program approval. One of the big questions is how these changes will impact Oregon's New Source Review program, administered by the Department of Environmental Quality (DEQ) and the Lane Regional Air Pollution Authority (LRAPA).

Under the New Source Review (NSR) air permitting program, major new sources and major modifications at existing sources are subject to a complex, expensive and time consuming set of legal and procedural requirements. The current federal NSR program dates back to 1980 when, after having a previous try tossed out by the courts, EPA adopted the fundamental elements of today's program. Under NSR, a new major source or major modification must determine and employ state-of-the-art controls. These sources must also model to demonstrate compliance with the state and national ambient air quality standards or-in areas that are not deemed in compliance with ambient air quality standards-obtain offsets that ensure a net air quality benefit. Sources must also perform modeling to ensure the protection of air quality related values in certain designated parks and wilderness areas. The process has extensive public involvement.

While it is relatively easy to tell if a brand new facility wants emissions large enough to trigger NSR, determining whether a change at an existing plant has increased emissions enough to qualify as a major modification is much more difficult. Make a change at a plant subject to the federal program and you have to compare the plant's actual emissions before the change to the plant's 'potential to emit' after the change. It is complicated enough to make this determination in relation to the actual device being changed. However, you also have to take into account all of the processes upstream and downstream of the process being modified to determine what impacts this change could have on their potential to emit. This latter process is referred to as a "debottlenecking" analysis.

As it is in so many ways, Oregon is unique when it comes to NSR. While based in large part on the federal requirements, Oregon chose early on to take its NSR program in a different direction. Perhaps the most strikingly original element was the implementation of the Plant Site Emission Limit (PSEL).

Oregon decided early on that it wanted a more practical and transparent means of tracking whether NSR was triggered. By establishing PSELs for the pollutants that could trigger NSR, Oregon established federally enforceable caps. These caps allowed for easy recognition of a facility's maximum potential to emit. No matter what, the potential to emit couldn't exceed the PSEL. In this regard, sources exchanged increased regulation for greater permitting certainty. In determining whether NSR was triggered at an existing source, the facility just needed to compare the PSELs to the facility's baseline emission rate (or "netting basis"). Since both numbers were typically established in a source's air permit, everyone could tell at a glance what size increase would trigger NSR.

The federal NSR program differs from Oregon's program in two primary ways (and many more subtle ways). First, under the federal program emission increases and decreases are only aggregated for the five year period prior to when construction is commenced. This means that under the federal program a large decrease that occurred five years and one day before construction is anticipated to commence cannot be used to 'net out' of NSR. Under the Oregon program, all decreases and increases must be considered since the last time the source went through NSR for that pollutant. Second, the federal program lacks a plantwide emission cap akin to Oregon's PSEL-so what may appear to be minor changes with no impact on actual emissions can have drastic and unforeseen impacts on a source's potential to emit. Many sources subject to the federal program have paid dearly as the result of not paying attention to this issue.

While the current federal NSR system works relatively predictably for brand new sources, existing sources are often loath to make simple changes for fear of how that could alter their potential to emit. They are also nervous about implementing emission reduction projects due to fear of losing their ability to net against the decrease in the future. Such concerns have frequently resulted in sources operating higher polluting processes longer than they would have in the absence of the federal NSR netting limitation.

In 1996, EPA proposed sweeping changes to the NSR program. 61 Fed. Reg. 38249 (July 23, 1996). These changes were designed to address the oddities identified above as well as other persistent problems with the program. Not until last month, however, were steps taken to actually implement any of these changes. Now it appears that in a matter of weeks the country will finally see some long awaited improvements to the federal NSR program.

On June 13, 2002, the EPA Administrator forwarded to the President a report discussing changes that EPA intends to make to the NSR program. After decades of defending the application of NSR program to existing sources, EPA made a shocking admission: this aspect of NSR has hurt the country. Specifically, EPA stated that in relation to existing sources, the NSR program "does discourage projects that improve capacity or efficiency but do not result in increases in actual air emissions." In a June 28, 2002 briefing, Jeff Holmstead, EPA's Assistant Administrator for Air and Radiation, stated that NSR has done more harm than good through its application to existing sources. He cited the perverse incentives that the program creates by penalizing sources that voluntarily reduced air emissions. Mr. Holmstead went on to say that while EPA recognized the benefits of applying NSR to new sources, no significant environmental benefit has been achieved by applying NSR to existing sources.

EPA is intending to revise NSR in seven respects: Plantwide Applicability Limits (PALs); Clean Unit Exemption; Pollution Control Project Exemption; Actual to Potential Emissions Comparison; Plantwide Applicability Limits (PALs); Routine Maintenance and Repair; and Project Aggregation.

Plantwide Applicability Limits (PALs)
One of EPA's premier concepts that it plans to move ahead on immediately is the idea of the Plantwide Applicability Limit (PAL). This concept, proposed in 1996, draws heavily from the Oregon PSEL. In fact, Mr. Holmstead even referred to Oregon as a federal pilot program, a disclosure that must certainly be news to those Oregonians who created the PSEL 20 years ago. As with the PSEL, the PAL will allow sources to set a plantwide cap on emissions. So long as they maintain actual emissions below that ceiling, then nothing that they do will trigger NSR. Increases would be allowed in the PAL, but only if the general NSR requirements (modeling, control technology review, public participation) were complied with.

It is unclear how or if the PAL will impact Oregon's program. Since Oregon was the model for this concept, it already has many of the key features. Implementing PAL should clear up any concerns that EPA's enforcement group-which has expressed concerns about PSEL in the past-has about this aspect of Oregon's air quality program.

Clean Unit Exemption
The second of EPA's principal fixes to NSR is the creation of the Clean Unit Exclusion. This idea, like the PAL, was part of the 1996 proposal. It is intended to create an incentive for sources to install the best emissions control technology in exchange for a period of time where NSR does not apply. If a source previously installed controls or implemented processes that are inherently clean or lower emitting and which are deemed to be equivalent to the level of control that NSR would have required at the time of installation, that source is exempt from NSR unless or until permitted allowable emissions increase. The 1996 proposal indicated that allowable emissions would be measured based on the source's maximum hourly potential to emit. Under this exemption, a source need not have actually gone through NSR so long as the controls installed would have satisfied the NSR requirements at the time installed. The exemption is expected top last 10 to 15 years; the time will be determined in the final rule.

It remains to be seen if this rule has any real impact in Oregon because, again, of the role of the PSEL. In Oregon, a source would never trigger NSR without an increase in the permitted allowable emission rate. A change in the hourly potential to emit will not, alone, trigger NSR in Oregon. Therefore, the primary goal of the rule is not likely to do much for Oregon sources. However, the rule is expected to include a provision that sources that have already installed current hazardous air pollutant requirements (e.g., Maximum Achievable Control Technology or "MACT") would not have to undergo a similar review under NSR where the same pollutants are involved. Similar provisions in the Oregon program could be useful to some sources.

Pollution Control Project Exemption
The third key element of the NSR improvement plan is to take a long-standing and inconsistently applied policy and put it in the rules.

One of the ironies of NSR is that it is possible to trigger the program by trying to install pollution controls or implement a project that will overall decrease emissions. For example, a company might want to install a thermal destruction technology that would dramatically reduce volatile organic compounds, including a number of hazardous air pollutants, but the technology will result in a small increase in the emission of oxides of nitrogen (NOx). Everyone may agree that the control installation is the best thing for the environment, but-absent any exemption for pollution control projects-the installation could be delayed by a year as the permitting authority complies with NSR's procedural requirements. In the meantime, the source is prohibited from installing the controls. EPA intends to finally exclude projects that result in a net air quality benefit from having to go through NSR. This obvious improvement was already incorporated into Oregon's rules, with EPA's blessing, a year ago. While technically EPA's move to catch up to Oregon should have little impact on sources in this state, the reality is that some within DEQ have refused to follow their own rule due to concerns that EPA would object to the streamlining of environmentally beneficial projects. This action by EPA should alleviate these concerns and ensure that Oregon sources can make these changes without time and expense penalties.

Actual to Potential Emissions Comparison
The fourth EPA action item is to change the way in which existing sources calculate whether they trigger NSR. NSR is triggered if the difference between the source's baseline emission level and its post-change emission level is greater than the applicable threshold. Under the current federal program, all but electric utilities compute this difference by comparing pre-change actual emissions to post-change potential emissions. This change to the federal program will allow sources to compare their actual emissions before a change to what they expect their actual emissions will be after the change.

This change is seen as dramatic in the context of the federal program. However, this change is unlikely to have any effect for Oregon sources. This is due to the fact that the Oregon program defines the baseline emission level so differently from the federal program. However, the federal proposal may encourage DEQ to allow sources that were built after 1978 ( the year used for defining the baseline emission rate in Oregon) to establish a baseline consistent with the 10 year look-back provision EPA has indicate it will adopt. This would create a more rationale incentive for newer sources to voluntarily install controls.

Routine Maintenance and Repair
One of the most hotly contested NSR issues over the past several years is whether and when work on an existing emission unit constitutes routine maintenance, repair or replacement. The NSR program exempts routine maintenance, repair and replacement. In recent years, EPA has mounted an aggressive enforcement initiative against electric utilities challenging whether work they had undertaken in the past was truly exempt maintenance and repair (see Wood, Insider #289). EPA has proposed to implement a "safe harbor" test where work below a certain cost threshold (determined by a percentage of the facility's capital replacement cost) will presumptively qualify as routine maintenance or repair. Unlike the prior elements, EPA recognizes that further discussion is merited on this point and the agency has indicated that it plans to propose this idea in draft form before proceeding to final rulemaking. Intriguingly, EPA has said that it will take interim steps to provide additional certainty as to what constitutes routine maintenance and repair during the pendency of the rulemaking.

This concept could have tremendous impact in Oregon. DEQ's rules exempt from the definition of "major modification" the "routine maintenance, repair, and replacement of components." Actions that are not major modifications do not trigger NSR-so knowing what constitutes an exempt activity can be critical to sources. However, DEQ specifically refrained from defining what qualifies as routine maintenance, repair or replacement because of EPA's activities in developing greater clarity at the federal level. Guidance or rules issued by EPA on the subject should assist DEQ and Oregon sources identifying the scope of this exemption.

Debottlenecking
The issue of debottlenecking is another which EPA has stated that it intends to examine further. This occurs under the federal program if a change to one part of the manufacturing process ends up speeding up another part of the process. Due to the PSEL concept, debottlenecking has never been an Oregon problem as sources cannot inadvertently increase their potential to emit (because it is capped by the PSEL). Therefore, changes EPA makes to this part of the NSR program are more of academic interest in Oregon. For companies with facilities outside of Oregon, however, clarification of how debottlenecking is treated is of tremendous importance.

Project Aggregation
The final issue EPA has identified as a priority action item is to further examine when projects are aggregated. Currently, there is much confusion under the federal program as to when two unrelated projects at a source must be permitted as a single action. Often times this decision has tremendous importance as to whether and how NSR applies. EPA has stated that it intends to float draft rules that would clarify that projects need not be aggregated unless either (1) the projects are dependent upon each other to be economically or technically viable or (2) the projects were intentionally split to avoid NSR. Because NSR applies to each modification separately, this rule change could have substantial impact on facilities with substantial, consecutive projects. For most Oregon sources, however, this rule is unlikely to have much impact as DEQ's rules already address many of the typically encountered aggregation questions left unanswered by the federal program.

As stated above, EPA plans to move ahead with the first four measures essentially immediately and without further public comment. The remainder will go through full notice and comment rulemaking. Once the rules are in place, however, it remains to be seen how they will be implemented by states such as Oregon that have existing federally approved programs. For the pollution control project exemption, the presence of federal rules should simply make Oregon's existing program work as it should. What will be most interesting to follow will be how the other changes, some of which are not consistent with how the Oregon program works, will be implemented. EPA has suggested at times that the state and regional permitting authorities will have to adopt the changes. At other times, EPA has suggested that states simply have the option. At a time when the preservation of state's rights is a rallying cry of many, it will be interesting to see whether EPA forces states to adopt its reforms.

For Additional Information, Contact: Tom Wood, Stoel Rives LLP at 503-294-9396.

RE: US EPA New Source Review Reform Information: http://www.epa.gov/air/nsr-review/index.htm