
                                       
                                       
                                       
                         Response to Comment Document
                                       
                               Nitric Acid NSPS
                                       
                                       
                              Part 60 Subpart Ga
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                   May 2012
Commenters
Docket Number
Respondent
Affiliation
EPA-HQ-OAR-2010-0750-0086
John Ambrose
Rentech
EPA-HQ-OAR-2010-0750-0087
Barbara L. Cabot
Dyno Nobel
EPA-HQ-OAR-2010-0750-0088
Chris S. Leason
Gallagher and Kennedy on behalf of The Fertilizer Institute (TFI)
EPA-HQ-OAR-2010-0750-0089
Adam Diamond
Agrium
EPA-HQ-OAR-2010-0750-0090
William E. Flederbach
ClimeCo
EPA-HQ-OAR-2010-0750-0091
Russell S. Frye
SSM Coalition
EPA-HQ-OAR-2010-0750-0092
Larry D. Fitzwater
LSB Industries
EPA-HQ-OAR-2010-0750-0093
Lorraine Krupa Gershman
American Chemistry Council (ACC)
EPA-HQ-OAR-2010-0750-0094
Jim Schellhorn
CF Industries
EPA-HQ-OAR-2010-0750-0095
Desi M. Chari
Dow Chemical
EPA-HQ-OAR-2010-0750-0096
Burl Ackerman
JR Simplot
EPA-HQ-OAR-2010-0750-0097
Richard S. Vickery and John A. Dege
DuPont
EPA-HQ-OAR-2010-0750-0098
William C. Herz
The Fertilizer Institute (TFI)
EPA-HQ-OAR-2010-0750-0099
Aruna Prabhala and Kassie Siegal
Climate Law Institute, Center for Biological Diversity
EPA-HQ-OAR-2010-0750-0100
Melvin E. Keener
Coalition for Responsible Waste Incineration (CRWI)
EPA-HQ-OAR-2010-0750-0101
Kassie Siegal
Center for Biological Diversity- addendum to main comment (EPA-HQ-OAR-2010-0750-0099)
EPA-HQ-OAR-2010-0750-0102
Peter Zalzal
Environmental Defense Fund (late)
EPA-HQ-OAR-2010-0750-0103
Michael A. Livermore and Jason A. Schwartz
Institute for Policy Integrity
EPA-HQ-OAR-2010-0750-0104
Teresa Clemmer
Vermont Law School (on behalf of EIP, Sierra Club, and NRDC)
EPA-HQ-OAR-2010-0750-0105
Stephenie Moyer
Nebraska Department of Environmental Quality, Air Quality Division, NSPS Coordinator
Comments and Responses

   1. NOX Standard
1a. BACT Determinations
Comment 1.1:   Commenters (0092, 0094, 0098) urge EPA to consider its recent BACT determinations that are set forth in various consent decrees with industry. According to the commenter, the proposed standards are based on a BSER which is unsupported by and more stringent than recent EPA BACT determinations. The commenter asserts that the EPA should fully identify and justify the basis for its BSER determination. The commenter states that EPA should look to its BACT/LAER clearinghouse and other recently established permit limits because, the commenter asserts that NSPS is the statutory floor, and cannot, by definition, be lower than BACT. 
Commenter 0094 states that in addition to the determinations referenced in the preamble, EPA recently issued BACT determinations regarding nine of the commenter's nitric acid plants. The commenter states that, effective June 6, 2011, the Consent Decree (see United States, et. al. v. Terra Industries Inc., et. al., Civil Action No. 11-4038 (U.S. Dist. Ct. N. Dist. Iowa)) requires the commenter to reduce NOX emissions from each covered nitric acid plant and comply with the following "BACT-equivalent" levels: a short-term NOX limit of 1.0 lb/ton; and a long-term NOX limit of 0.60 lb/ton. The commenter states that these BACT determinations are inconsistent with and do not support the proposed standards. According to the commenter, the recent BACT determinations (and specifically those made by EPA) do not support a 0.50 lb/ton standard; rather, they universally identify a 0.60 or 0.61 lb/ton standard. 
The commenter further argues that the proposed standards should utilize and be based on the EPA BACT determinations identified by the commenter and in the preamble. The commenter notes that similar information is evaluated for both BACT and BSER purposes. Although BACT determinations may include site-specific considerations which are not applicable to all nitric acid plants and therefore have to be accounted for in the BSER determination, the commenter maintains that the identified EPA BACT determinations do not appear to be based on or influenced by any site-specific factors. The commenter emphasizes that the EPA BACT determinations universally identified a 0.60 or 0.61 lb/ton standard with a number of such determinations being effective as late as June 2011.
Commenter (0094) states the proposed standards are unsupported by and are contrary to recent BACT determinations. Citing three BACT determinations referenced in the preamble, the commenter notes that only one of the referenced BACT determinations indicated compliance included periods of startup, shutdown, and malfunction (i.e. Agrium Kennewick determination). The commenter also cites a June 2011 Consent Decree entered into by the commenter and EPA regarding nine of the commenter's nitric acid plants. In that Consent Decree, the commenter notes requirements to meet short-term limits that do not apply during periods of startup, shutdown, or malfunction; and to meet a long-term limit that is applicable at all times, including periods of startup, shutdown, or malfunction. The commenter reiterates that these determinations show that EPA BACT determinations are inconsistent with and do not support the proposed standards.
Response to Comment 1.1:  Determining the level of NSPS standards during a review of an existing NSPS is an inherently data-driven process. Based on the emissions and control technology data available to the agency, we are promulgating an emission standard of 0.50 lb/ton NOX. The fact that past BACT determinations exist that are less stringent than the limit being promulgated in this action does not undermine the process used to develop this standard. While BACT determinations and NSPS reviews look at some of the same technologies and processes, BACT determinations and NSPS standards are developed under two different regulatory regimes and there is no statutory or definitional bar to the level of an NSPS being more stringent than a past BACT determination. We note that that both recent BACT determinations and this NSPS review have found that emissions well below the standard promulgated in subpart G are achievable for new and modified sources.
1b. General Duty Clause
Comment 1.2:  Commenter 0091 objects to EPA's inclusion of a new "General Duty Clause" in the nitric acid plant NSPS. According to the commenter, EPA proposes to include in the Proposed Standards, without any explanation or even mention in the preamble to the Proposed Rule, a provision stating a "general duty to minimize emissions." The commenter notes that the provision is similar to, but differs in important respects from, the "general duty clause" of the NSPS General Provisions, 40 CFR §60.11(d), which EPA has relied upon for decades and for dozens of other NSPS, including the original NSPS for Nitric Acid Plants at 40 CFR 60 subpart G.
The commenter urges EPA to justify its promulgation of a new regulation and its decision that the existing regulation is insufficient. According to the commenter, the EPA has not proposed to amend the NSPS General Provisions, so there could be unresolved conflicts between the two provisions, both of which apparently would remain applicable to facilities subject to the Subpart Ga Nitric Acid Plant NSPS. The commenter suggests that EPA delete proposed section 60.72a(b) from the final rule. If it fails to do so, according to the commenter, it would need to re-propose the provision with some explanation of the basis and purpose for the provision, to allow the public an opportunity to provide meaningful comments, as required by CAA section 307(d)(3). 
According to the commenter, the NSPS General Provisions general duty clause has a long history of application and interpretation, which EPA would effectively jettison for purposes of the Proposed Standards. The commenter adds that while the General Provisions general duty clause states that owners and operators shall maintain and operate an affected facility in a manner consistent with good air pollution control practices for minimizing emissions "to the extent practicable," the general duty provision in the Proposed Standards contains no such qualification. The commenter asks EPA to compare 40 CFR §60.11(d) with proposed section 60.72a(b). The commenter asks that EPA justify its deletion of that very important phrase. Additionally, if EPA retains a separate general duty provision in the Subpart Ga NSPS, the commenter asks that EPA include a sentence making it clear that the requirement to operate processes and pollution control equipment to minimize emissions does not require reducing emissions below those allowed by the applicable emission limitation, for the reasons described earlier in these comments.
The commenter notes that the general duty provision applies to "any affected source," while the General Provisions general duty clause applies to "any affected facility." The commenter states that the former is not defined in the Proposed Standards nor in the General Provisions; the latter is a term defined in the General Provisions and used throughout the NSPS regulations. If, in fact, EPA intends for the general duty provision of the Proposed Standards to apply to the entire stationary source, and not just the "affected facility," the commenter asserts that the proposed provision goes beyond EPA's authority under CAA section 111.
Comment 1.3:  Commenter 0100 suggests that the general duty to minimize emissions language of §60.72a(b) is redundant and is not needed in the final rule. According to the commenter, the language appears to be similar to the language in §60.11(d). The commenter notes that while there are differences between the two paragraphs, both are intended to require the affected source to minimize emissions at all times. Since all affected sources regulated under 40 CFR Part 60 Subpart Ga are also required to meet the requirements in 40 CFR Part 60 Subpart A, the commenter feels that the inclusion of this paragraph appears to be redundant. The commenter suggests that if the Agency agrees with this suggestion, there are two references to §60.72a(b) that EPA needs to be changed to §60.11(d). These are in §60.76a(f)(2) and §60.77a(f)(2).
Response to Comments 1.2-1.3: We agree with the commenters concerning the use of "affected facility" versus "affected source." As a result, the EPA has changed "affected source" to "affected facility" within the regulatory text in Subpart Ga.
We have removed the proposed paragraph §60.72a(b) from the final rule. The two cross references in §60.76a(f)(2) and §60.77a(f)(2) have been changed to reference §60.11(d).
1c. Standard based on 30-day rolling emission rate
Comment 1.4:   Commenters (0093, 0095) support EPA's rationale in choosing a long-term rolling emission rate as the appropriate basis for setting the compliance standard so as to better reflect actual nitric acid facility operations.
Comment 1.5:  Commenters (0092, 0094, 0096, 0098) recommend a 365-day rolling average or a 12-month rolling average as the appropriate period to demonstrate compliance.
Comment 1.6:   Commenters (0094, 0096, 0098) state that the proposed standards are unsupported by and are contrary to recent BACT determinations. Commenter 0094 cites three BACT determinations referenced in the preamble, stating that none of the referenced BACT determinations included a compliance demonstration period based on a 30-day emission rate (i. e., Agrium North Bend was based on a 365-day rolling emission rate and Agrium Kennewick was based on a 12-month rolling emission rate). The commenter also references BACT determinations issued in June 2011 resulting in a Consent Decree requiring: a short-term NOX limit of 1.0 lb/ton with compliance based on a rolling 3-hour average (rolled hourly); and a long-term NOX limit of 0.60 lb/ton with compliance determined based on a rolling 365-day emission rate (rolled daily). Commenters (0094, 0096, and 0098) state that as opposed to a 30-day compliance period, such determinations specify a 365-day rolling emission rate or a 12-month rolling emission rate as the appropriate period to demonstrate compliance with the 0.60/0.61 lb/ton standards. Lastly, although several of the determinations indicate the 0.60/0.61 lb/ton standards are applicable at all times, the commenter  indicates that they are based on a one year compliance period, not a 30-day period. Commenter 0096 supports a 365 day period rather than a 30 day period if the 0.50 lb/ton limit remains in place for periods of startup and shutdown.
Comment 1.7:  Commenter 0092 suggests that for only new, modified or reconstructed "dual-pressure" technology nitric acid plants, the NSPS rolling emission rate should be increased to 365 days with a limit of 0.60 pounds NOX per ton of nitric acid (100% basis).
Comment 1.8:  Commenter 0097 disagrees with the proposed 30-day rolling emission rate. According to the commenter, at least one of the units used by EPA to establish the standard apparently could not comply with the new standard since compliance is required 100% of the time and the EPA analysis shows that they could only achieve compliance 99% of the time. To address this, the commenter suggests EPA increase the time period for averaging (e.g. 365 day rolling average), increase the numerical standard, or both. 
Response to Comments 1.4-1.8:  The EPA disagrees with the commenters who recommend a change to the compliance period used for the NOX standard. The standard established under Subpart G of Part 60 provided for a 3-hour period during which the NOX emissions could not exceed the standard of 3.0 lb NOX/ton acid but did not include periods of SSM. EPA believes that the change to a 30 day period is reasonable considering that Subpart Ga of Part 60 also includes periods of SSM. An emission standard based on the data from the test plants calculated as a 365 day rolling emission rate would be lower than the 0.50 lb/ton value calculated as a 30 day rolling emission rate. The value for the standard included in this NSPS is based on the EPA's technical evaluation of the emissions data available to the Agency. As discussed in Section V of the final rule preamble, that technical evaluation leads the Agency to a standard of 0.50 lb NOX /ton acid calculated as a 30 day rolling emission rate. Were the Agency to recalculate the standard based on a 365 day emission rate, the value obtained would not be the 0.60 lb NOX /ton acid standard recommended by the commenters. The standard would likely be substantially lower than 0.50 lb NOX/ton acid if a 365 day emission rate were used. The 30 day emission rate is a sufficiently long time period to account for short term operational variability that may create compliance problems with a 0.50 lb NOX/ton acid if the standard was based on a short term value, such as the three hour rolling emission rate.
 The EPA has communicated periodically with industry during the last 3 years, has been receptive to any facility-specific emissions data that could be provided, and asked for data related to BACT determinations based on reconstructions and modifications on more than one occasion. However, EPA did not receive any data from industry during that period that could be useful in evaluating the proposed standards in relation to modified and reconstructed nitric acid plants other than the data for Agrium North Bend. Furthermore, data received during and after the close of the public comment period for the proposed rule did not change the results of our analysis. See Statistical Evaluation of CEMS Data to Determine the NOX Emission Standard (Updated Memo for Final Standard), and Analyses of Data Received Between Proposal and Promulgation of Part 60, Subpart Ga. See Section V of the final rule preamble for discussion of achievability of the standard for the test plants. 
1d. Standard should be changed to a site-specific standard
Comment 1.9:  Commenter 0097 suggests that as an alternative to the proposed limit, EPA should adopt a technology-based standard, e.g. install SCR, and base the emission limit on stack testing (case by case). According to the commenter, the sources could then conduct performance tests at typical operating conditions to establish reasonable limits and averaging times that would also include startup and shutdown events typical for the type of operation at that source.
Response to Comment 1.9:  The EPA disagrees that the emission limit should be changed to a technology-based, site-specific standard. The standard must be achievable by all sources that become subject to this NSPS using best systems of continuous emission reduction. Based on the data available to the Agency, we believe the standard included in the final rule is achievable by facilities using such systems. See Section V of the final rule preamble for further discussion of the level of the standard. CAA 111(b)(5) provides that, except for standards set under CAA 111(h),  CAA 111 does not require or authorize EPA to require that a source use any particular technological system to comply with an NSPS. Additionally, the application of design, equipment, work practice, or operational standards can be applied when the criteria in section 111(h)(2) are met, which is not the case for nitric acid production units. 
 2. Test Methods and Monitoring
2a. Accuracy Ranges of Monitoring Equipment
Comment 2.1:  Commenter 0097 states that all monitoring systems have inherent ranges of accuracies of at least +/- 20%. Thus the commenter asks that the standard be established at least 20% above the numerical data provided in the proposal. According to the commenter, good operation and maintenance requirements included in General Provisions provide adequate certainty that operators are minimizing emissions within the design capability of the equipment.
Response to Comment 2.1:  We agree with the commenter that all measurement data have some inherent uncertainty but we believe that the emissions limit and the averaging time adequately account for that uncertainty as well as normal process variability and that no additional adjustment is necessary. As with other EPA rulemaking projects, we collected emissions data from multiple sources to use in determining an achievable emissions limitation that represents the best systems for emissions reduction. In the case of the nitric acid plant NSPS data collection, we received data collected by CEMS installed on nitric acid plants and certified to the applicable performance specification. Those data inherently represent the measurement uncertainty associated with CEMS measurements. We also applied statistical procedures at proposal to determine the 99[th] percentile that provided additional assurance that we have considered measurement uncertainty along with process variability in establishing the limit. This analysis can be found in the memorandum Statistical Evaluation of CEMS Data to Determine the NOX Emission Standard (Updated Memo for Final Standard). We note that the final rule used the maximum 30-day rolling emission rate rather than the 99[th] percentile. See Response to Comments 5.6-5.7.
2b. Additional Monitoring Equipment to Measure Larger Span 
Comment 2.2:   Commenters (0096, 0098) assert that the EPA failed to recognize limitations with monitoring equipment during periods of startup and shutdown that significantly underestimate emissions during these times.
In order to comply with the current NSPS in Subpart G, the commenters note that existing NAPs (including the Test Plants) utilize CEMS to monitor their NOX emissions. According to the commenters, these analyzers typically are equipped with only a single span range that is calibrated to monitor emissions in the 0-500 ppm NOX range, in compliance with Subpart G. The commenter's members typically experience NOX emissions in the range of 2,000 to 5,000 ppm during startup and shutdown events. According to the commenters, the Subpart G rules' span value requirements prevent the NAPs' CEMS from accurately measuring the high emissions during startup and shutdown. According to the commenters, the analyzers max out at 500 ppm during these periods. It appears to the commenters that the EPA did not consider this issue when evaluating the Test Plants' CEMS data, and simply assumed the CEMS data it was analyzing accurately reflected emissions, even during the high emissions typical of startup and shutdown. The commenters assert that this assumption has the effect of grossly underestimating emissions during startup and shutdown at 500 ppm, when such emissions are actually four to ten times higher. The commenters note that this effect materially undercuts the EPA's proposed NSPS because under Subpart Ga NAPs are required to measure emissions at all times (including startup and shutdown), and such emissions are factored into compliance. Because the EPA's assumption capped startup and shutdown emissions at 500 ppm (due to the limitations of CEMS as calibrated under Subpart G), the commenters believe that the EPA's proposed NSPS does not accurately account for actual startup and shutdown emissions. (The commenters note that the Test Plants were atypical and the vast majority of NAPs experience very high emissions during startup and shutdown that are not accurately measured by a CEMS calibrated in accordance with Subpart G.) In order to support any proposed NSPS that requires compliance at all times that is achievable (as it must be under CAA § 111), the commenters state that the EPA must account, accurately, for all emissions.
In addition to the Subpart G limitations, the commenter's members and their consultants have found that when CEMS are calibrated to a 0-2,000 ppm span value to attempt to accurately capture emissions during startup and shutdown (as would be required by the proposed NSPS), the NOX monitor does not function accurately at the range that the NAPs would be required to maintain at least 99 percent of the time in order to comply with the proposed NSPS.
Response to Comment 2.2: We understand the concern about the possibility that the average CEMS concentrations for some facilities in the database may be biased low if the CEMS output were detector range limited. The commenters provided no additional data to help evaluate to what degree such bias may have occurred. On the other hand, we are aware that at least some of the emissions data in the database that the sources provided represent startup and shutdown periods with significant spikes in emissions (e.g., a factor of 10) over emissions reported for normal production periods. We believe that the database accounts for emissions spikes that occur during startup and shutdown periods and that the final NOX emissions limit established is representative and inclusive of emissions for all periods of operation. 
We do agree with the commenter that a CEMS intended to measure emissions across a wide range of concentrations must be calibrated for all measurement ranges including spikes that may occur during SS periods. In some cases meeting this need will require a dual range analyzer with separate sets of calibration gases (and associated span values). We have revised the rule to address this concern. Specifically, we have included a second CEMS span value applicable to dual range systems based on the highest NOX concentration expected during startup and shutdown periods.
In response to the commenter's assertion that the lack of dual span monitors impacts the data used in the analysis, the EPA notes that dual span monitors are in use at the Agrium North Bend plant. Therefore, the EPA believes that the emissions during startup and shutdown were accurately captured for this facility.
We note that periods of startup and shutdown do not constitute a large fraction of operating time and are relatively short term events. We also note that the method for calculating the 30 operating day rolling emission rate has been clarified to ensure that days with only a few hours of operation, that may include a startup or shutdown, do not have an artificially exaggerated impact on the 30 operating day emission rate. For the final rule, the 30 operating day emission rate is calculated by summing the lb NOX/ton of acid values calculated each operating hour and dividing by the total number of operating hours in the 30 operating day period. See Section V of the final rule preamble and the memo entitled Statistical Evaluation of CEMS Data to Determine the NOX Emission Standard (Updated Memo for Final Standard).
2c. CERMS
Comment 2.3:  Commenters (0093, 0095) object to EPA's proposed requirement for facilities subject to this NSPS to install a continuous emissions rate monitoring system (CERMS) comprised of a NOX CEMS and a continuous exhaust gas flow rate monitoring system. According to the commenters, the EPA has not provided data comparing flow rate data to support its requirement for installing CERMS. According to the commenters, having to install a flow rate monitoring system would put new or modified facilities at a competitive disadvantage, given that the EPA estimates the capital cost of such units to be approximately $39,000, plus the additional costs of maintenance and testing. In addition, the commenters believe that the installation of CERMS would likely result in additional facility startups and shutdowns due to additional equipment maintenance and testing requirements, which could result in facilities not being able to comply with the proposed emission standard.
Furthermore, the commenter asserts that existing nitric acid facilities that undergo a modification may not be able to install a flow rate monitoring system without completely overhauling existing metering and control systems. The commenters note that there may not be space in existing control rooms for the installation of the additional meter, and a new data collection system may need to be installed in order to accurately and correctly capture the additional data.
Commenter 0093 states that if the EPA finalizes the CERMS requirement, commenter 0093 supports EPA's finalization of the proposed performance criteria for such a system, which would allow facilities the ability to choose from a range of possible metering systems.
Comment 2.4:  Commenter 0092 asks that if the NSPS requires installation of a continuous emissions rate monitoring system (CERMS) comprised of a NOX continuous emissions monitoring system (CEMS) and a continuous exhaust gas flow rate monitoring system, the commenter proposes that the selection of the type of flow rate monitoring device (e.g., pitot tube vs thermal resistance flow meters) be at the option of the entity in whose name the Title V Air Permit is issued, subject to approval of the regulatory authority issuing the permit. 
Comment 2.5:  Commenter 0095 suggests as an alternative to the requirement to install a flow rate monitoring device, the existing Subpart G allows use of flow measurement reading along with CEMS data to establish a conversion factor for the purpose of converting monitoring data into units of the applicable standard (lb/ton). The commenter suggests that EPA add an option to the proposed Subpart Ga that allows the owner/operator to reestablish the conversion factor annually to calculate NOX emissions in the proposed Subpart Ga instead of requiring the continuous flow rate monitoring system.
Comment 2.6:  Commenter 0097 asks that the EPA provide data to support the assumption that a CERMS is more accurate and that the benefits are worth the additional cost, especially for small plants. The CERMS approach is a more complex and potentially a more unreliable system compared to systems required by the current standard. According to the commenter, the CERMS could result in more startups and shutdowns associated with maintaining the new equipment. The commenter recommends that the EPA retain the current factor method and add the new CERMS method as a second monitoring option. 
Response to Comments 2.3-2.6: The costs for NOX continuous emissions monitoring systems were included in the analysis of the economic impact of the initial new source performance standard as the commenter noted. The costs for a NOX concentration monitor apply to a new source whether constructed under the existing subpart G or under subpart Ga. New sources would apply CEMS in much the same manner under subpart Ga as under subpart G, so the economic analysis included only the incremental costs for CEMS operation over those already projected for new sources under subpart G, i.e., the costs for flow meters. We agree with the commenter that the economic analysis for the proposal failed to include all of the incremental additional costs associated with implementing the CEMS requirements of the revised rule (e.g., complying with 40 CFR part 60, appendix F procedure 1). We have updated the economic analysis for the final rule by including the cost of dual span CEMS and annual RATA tests required to certify the CEMS. For more information see Impacts of Nitric Acid NSPS Review-NOX (Updated Memo for Final NSPS). Also see Section V of the final rule preamble.
The EPA believes that the use of CERMS is warranted. Using a factor established during the steady state conditions of a RATA (as allowed by Subpart G) is not necessarily representative of real-time operations that fluctuate to meet increased demand, or due to ambient temperature changes. CERMS offers instantaneous compliance determinations. With the CERMS, the facility will know the flow rate of the emissions stream at any given time. When combined with the concentration of NOX and the rate of nitric acid production, real-time compliance can be determined at any time.
 3. GHG Regulation
Comment 3.1:  Commenter 0099 states that the high degree and certainty of the threat to public health and welfare from greenhouse gases such as N2O has been well-established; in 2007, the Supreme Court found that GHGs are air pollutants under the Clean Air Act (CAA) in Massachusetts v. EPA. The commenter asserts that the EPA echoed that finding in 2009 when the agency concluded that N2O and other GHGs "may reasonably be anticipated both to endanger public health and to endanger public welfare." According to the commenter, the agency has also acknowledged that N2O emissions are 4.6% of total GHG emissions, that nitric acid production is the third largest source of N2O emission in the United States, and that nitric acid production contributes 6% to total N2O emissions by the U.S.
Comment 3.2:   Commenter 0104 states that according to EPA's 2009 Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air Act (74 FR 66496) six well mixed greenhouse gases, including nitrous oxide, endanger public health and welfare within the meaning of the Act. According to the commenter, the EPA has stated that global warming endangers public health and welfare in numerous ways.
According to the commenter, EPA has further explained that children, the elderly, and the poor are the most vulnerable to these impacts; that greenhouse gas pollution endangers both current and future generations; and that the risk and severity of these impacts is expected to increase over time.
Comment 3.3:  Commenters (0103, 0104) state that N2O is a potent GHG -- 310 times more potent than carbon dioxide. Therefore, even when emitted in small quantities, the commenters state that this gas can pose a significant threat to public health and welfare. The commenters assert that EPA must utilize the opportunity of NSPS review to regulate N2O emissions from nitric acid plants. Although EPA has some discretion to set standards of performance for particular air pollutants under NSPS, the commenters state that its discretion is limited.
Comment 3.4:  Commenter (0102) asks the EPA to adopt 111(b) standards covering N2O emissions. The commenter notes that section 111(b)(1)(A) requires the Administrator to list a category of sources if "it causes, or contributes significantly to, air pollution which may reasonably be anticipated to endanger public health or welfare." According to the commenter, the Administrator then must publish "standards of performance for new sources within such category." The commenter notes that nothing in the CAA language requires the Administrator to make an individualized endangerment finding for each pollutant from the listed category. To the contrary, the commenter argues that the definition of "standard of performance" is expansive, and does not limit performance standard applicability to the pollutant(s) that triggered a sector's listing under Section 111. The commenter further argues that greenhouse gas emissions are air pollutants under the CAA and are therefore subject to performance standards issued under Section 111 for listed sectors. As such, according to the commenter, EPA must set standards under Section 111(b) for emissions of N2O from nitric acid plants.
Comment 3.5:   Commenter (0099) notes that as the EPA is aware, the Climate Action Reserve has recently finalized its Nitric Acid Production Project Protocol, under which nitric acid producers may obtain credits for nitrous oxide reductions that are not otherwise required by law. The commenter states that to the extent that a desire to promote this voluntary emissions reduction system as a substitute for existing legal requirements underpins EPA's proposal to omit nitrous oxide from the revised NSPS, such a rationale would be both illegal and inappropriate. The commenter argues that immediate and assured GHG reductions are urgently needed in order to avert truly catastrophic climate change damages. According to the commenter, the NSPS program provides a time tested and successful mechanism to achieve such reductions, and the EPA may not choose to disregard the law in order to facilitate payments to polluters for actions that they would otherwise be legally obligated to perform.
Comment 3.6:  Commenter (0099) urges the EPA to fully consider emission standards that include N2O and to promulgate an emissions limit for both NOX and N2O emissions. The commenter notes that merely recommending the use of technology that would control N2O emissions rather than requiring it represents both a statutory and policy failure by the agency. 
As quoted by the commenter, under §111, the EPA "shall...review, and if appropriate, revise" its new source performance standards for each regulated source. According to the commenter, EPA has interpreted the term "appropriate" to turn on two factors, "(1) the amount of emissions from that source category, and (2) the availability of demonstrated control measures." The commenter states that the EPA acknowledges in its proposed NSPS revision that "these nitric acid production units also emit another nitrogen compound known as nitrous oxide (N2O), which is considered a greenhouse gas (GHG)." According to the commenter, the agency then goes on to say, "technologies such as nonselective catalytic reduction (NSCR) are effective in controlling both NOX and N2O." In 2010, the EPA prepared a white paper titled "Available and Emerging Technologies for Reducing Greenhouse Gas Emissions from the Nitric Acid Production Industry." According to the commenter, the white paper concluded that four different types of control technology (including primary, secondary, other tertiary controls and NSCR) all effectively demonstrated in practice that they could reduce N2O emissions. The commenter notes that the EPA concluded that secondary and tertiary controls were capable of achieving greater than 80% reduction of N2O emissions from nitric acid production units. 
The commenter asserts that the EPA must regulate N2O emissions alongside NOX emissions when revising its NSPS for nitric acid plants. In addition to being a plain violation of the statutory text and the agency's mandatory duties, the commenter argues that the EPA's failure to do so is also arbitrary, capricious, and unreasonable. The commenter further notes that the EPA's failure to propose an emissions standard that includes N2O is particularly egregious given the ready availability and low cost of nitrous oxide emissions reductions.
The commenter asserts that the EPA cannot pick and choose with which statutory obligations it will comply and which it will ignore. The commenter states that the agency must take into account N2O emissions when establishing new NSPS for nitric acid plants and also correct the proposal's additional shortcomings as discussed above.
Comment 3.7:  Commenter (0103) states that EPA is required to "establish Federal standards of performance for new sources within [the nitric acid source category]." The commenter notes that while EPA will be acting to set emissions standards for N2O during a review and not upon a new listing of a category, once it commences a review, EPA must "follow the procedure required by this subsection for promulgation . . . ." According to the commenter, because cost‐benefit analysis is not precluded as a standard for determining which pollutants to regulate at promulgation, it is likewise not precluded during review.
Comment 3.8:  Commenter (0103) claims that EPA's justifications for not including all pollutants in other rules (Standards of Performance for Coal Preparation and Processing Plants; Standards of Performance for New Stationary Sources: Onshore Natural Gas Processing) do not apply for N2O. According to the commenter, although some of the same technology that limits NOX emissions can control N2O, without a specific standard such technology will not be widely implemented. According to the commenter, this is indicated by EPA reports that show only seventeen percent of plants are using NSCR technology, the only technology shown to control the emissions of both NOX and N2O. Likewise, the commenter notes a 1987 rulemaking for various manufacturing processes, where the agency decided to regulate only the pollutant that was emitted in the greatest quantity. [Standards of Performance for New Stationary Sources; Polypropylene, Polyethylene, Polystyrene, and Poly(ethylene terephthalate) Manufacturing Industry, ("These pollutants, however, are emitted at much lower quantities . . . and, as a result, standards development for this industry is focusing initially on limiting emissions of VOC."). ] According to the commenter, this approach is improper because quantity of emissions does not necessarily correlate to the cost of emissions reductions or the relative net benefits of pollutant reduction. Basing the decision on the quantity emitted could be considered unreasonable. Lastly, the commenter notes that EPA has looked to Nat'l Lime Ass'n v. EPA to support its argument in favor of discretion; however, in that case, the agency cites cost and the unavailability of adequate technology as reasons for not regulating certain pollutants.
Comment 3.9:   Commenter (0104) feels that EPA has a duty under CAA §111 to gather the information needed to develop performance standards for all pollutants emitted in significant quantities by a listed industry, including greenhouse gas pollutants.
The commenter notes that if EPA felt it was lacking information concerning nitrous oxide control technologies for the nitric acid industry, it should have requested information from the industry, particularly from the 14 nitric acid process units in the United States which are already using NSCR, a technology which controls both nitrous oxide and NOX. Similarly, the commenter asserts that EPA should have used its information gathering authority under Section 114 to develop the information needed to fully analyze the extent to which nitric acid plants emit carbon dioxide, methane, and other greenhouse gases and whether there are any technically feasible and economically reasonable methods for controlling such emissions. According to the commenter, EPA's failure to gather the necessary information concerning greenhouse gas emissions and controls, failure to conduct the required analysis of greenhouse gas emissions and controls, and failure to include any greenhouse gas standards in its proposed rule render this rulemaking unlawful.
The commenter states that EPA should have used its information-gathering authority under Section 114 to develop information concerning whether the technologies under consideration for controlling target pollutants, such as NOX, may have collateral environmental effects by increasing emissions of greenhouse gas pollutants. According to the commenter, EPA's technology of choice for controlling NOX emissions -- SCR --  is known to generate additional emissions of nitrous oxide depending on the temperature of the exhaust gas. The commenter asks that EPA consider and evaluate this issue before issuing a final rule.
Comment 3.10:  Commenter (0103) states that the CAA does not expressly lay out a decision making framework by which EPA must determine which "air pollutants" get standards of performance and which do not. According to the commenter, the Act does specify that standards of performance only apply to "air pollutants." However, according to the commenter, N2O satisfies this requirement, as both the Supreme Court and the EPA have concluded greenhouse gases, of which N2O is a constituent, are "air pollutants" under the CAA. The commenter states that the primary decision making framework laid out in Section 111 applies only to listing categories of sources: any category which, in EPA's judgment "significantly contributes to air pollution that endangers public health or welfare" should be included. According to the commenter, determining which of the pollutants emitted by already‐listed source categories should be regulated, however, need not be based on a finding of "significant contribution" and "endangerment." According to the commenter, nitric acid plants have been listed as a source category since March 1971, so EPA need not determine that N2O from nitric acid plants significantly contributes to air pollution that endangers public health or welfare in order to set a standard of performance for that pollutant. The commenter notes that although the definition of "standard of performance" specifies a particular emissions standard (i.e., "best," "cost," "demonstrated"), it does not specify when a standard must apply to a particular pollutant. According to the commenter, EPA has some discretion to make this determination.
Comment 3.11:  Commenter (0103) remarks that to the extent EPA does have discretion to select which pollutants to regulate, it must exercise that discretion in a non-arbitrary way. The commenter notes that a cost-benefit framework provides a clear and rational basis for choosing the standards of performance for any given pollutant  -  it should use this discretion to target pollutants where analysis indicates benefits will justify costs. Consistent with the requirements of executive orders, the commenter asks that EPA assess the costs and benefits of setting standards of performance for each potential pollutant emitted from a source, and then target those pollutants where regulation would maximize benefits. According to the commenter, cost‐benefit analysis counsels toward setting standards for N2O emissions from nitric acid plants. The commenter notes strong evidence that indicates that the benefits of regulating N2O will justify the costs. According to the commenter, the benefit of regulation of N2O is predominantly the reduction of the negative impacts of global climate change. The commenter notes that EPA has previously determined that, GHGs, of which N2O is a constituent, pose a danger to public health and welfare. The commenter adds that the social cost of carbon can also be used to approximate the benefits of reducing N2O, once N2O pollution has been translated into carbon‐dioxide equivalent units. According to the commenter, the EPA has not prescribed a single monetized number for the social cost of carbon; instead it uses four different quantifications, ranging from $4.70 to $64.90 per ton. The commenter recommends using the higher numbers. Even using all but the lowest possible values, the commenter notes that NSPS for N2O will likely produce benefits that exceed costs. According to the commenter, using the central value of $21 would produce benefits that far exceed the costs of technological upgrades and process changes. According to the commenter, the EPA's recent review of current and emerging technologies available for nitric acid plants to reduce N2O emissions indicates that the cost of reduction are relatively low. The commenter states that by utilizing technology "demonstrated in practice," nitric acid plants can reduce over eighty percent of their N2O emissions at a cost of $2.32 ‐ $6.49 for every ton of carbon‐dioxide equivalent reduced.
Comment 3.12:  Commenter (0103) states that the Proposed Rule itself provides no justification for choosing not to regulate N2O. According to the commenter, the Proposed Rule acknowledges that some NOX control technologies could also control for N2O and concludes that "any controls applied to control NOX emissions would not preclude installing cost effective N2O control technologies in the future." The commenter states that EPA must provide a rational basis for its failure to include N2O in the Nitric Acid NSPS within the final rule. The commenter notes that no such rational basis may exist, given EPA's efforts to begin to regulate GHG emissions from other sectors, the current availability of control technology, and the fact that the benefits of regulation of N2O emissions from nitric acid plants outweigh the costs. According to the commenter, failure to include nitrous oxide standards of performance for nitric acid plants would be arbitrary and capricious. 
Comment 3.13:   Commenter (0104) states that in determining whether to revise an existing standard as well as whether to add standards for pollutants not previously covered by a performance standard, the commenter notes that EPA has long interpreted the CAA's "appropriateness" determination to turn on two, and only two, factors: (1) the quantity of emissions from the source category, and (2) the availability of demonstrated control measures.
The commenter asserts that there is a large body of scientific data and evidence that would provide a sound technical basis for EPA to establish nitrous oxide controls for nitric acid plants. According to the commenter, if EPA had applied its traditional two-factor approach, there is only one conclusion it could have reached -- that regulation of nitrous oxide is appropriate and required. With respect to the first factor, the commenter notes that the nitric acid industry emits substantial quantities of nitrous oxide as well as other greenhouse gases. The commenter also states that nitric acid plants are the largest industrial source of nitrous oxide emissions in the United States. 
With respect to the second factor, the commenter asserts that there are numerous well-established and highly effective control technologies available for regulating nitrous oxide emissions from the nitric acid industry at a reasonable cost, and there may well be feasible strategies for controlling other greenhouse gases as well, as discussed further below.
Comment 3.14:   Commenter (0099) states that the EPA's choice of SCR rather than NSCR violates both the letter and intent of section 111. According to the commenter, courts have held that the NSPS program "looks toward what may fairly be projected for the regulated future, rather than the state of the art at present" since it applies to new plants. Similarly, the commenter notes that Congress' intent in establishing the NSPS process was "to induce, to stimulate, and to augment the innovative character of industry in reaching for more effective, less costly systems to control air pollution." Moreover, in determining what technologies are "adequately demonstrated" to form the basis for an "achievable" performance standard for a particular source category, the commenter asserts that courts have held that EPA's analysis can  -  indeed should, as appropriate  -  look beyond facilities within the United States. The commenter notes that EPA may base its standards on the application of systems of control that are in use in other countries, as well as looking at technology transfers across industries as the basis for an "achievable" performance standard. According to the commenter, the EPA's failure to even discuss and analyze N2O reductions, let alone require them, is an egregious violation of the agency's statutory duty.
The commenter asks that EPA analyze the potential for both NOX and N2O emission controls, and must consider all emissions reductions systems, including those discussed in the EPA's 2010 white paper. The commenter asserts that EPA may not simply ignore an important pollutant when revising the standard.
Comment 3.15:  Commenter (0104) suggests that nitrous oxide controls for nitric acid plants are technically feasible and can be implemented at a reasonable cost. The commenter notes that under the Clean Air Act, a standard of performance must be "achievable," and it must "take into account the cost of achieving such reduction and any nonair quality health and environmental impact and energy requirements" According to the commenter, this legal standard is meant to be technology-forcing, and courts have "recognized that section 111 `looks toward what may fairly be projected for the regulated future, rather than the state of the art at present.'" In the case of nitrous oxide emissions from nitric acid plants, however, the commenter notes that it is not necessary to look toward the future. According to the commenter, EPA has already collected extensive data establishing beyond doubt that technically feasible and economically reasonable controls for nitrous oxide emissions are available now.
Comment 3.16:  Commenter (0104) states that EPA has produced several reports listing seven control technologies that are highly effective in reducing N2O emissions during the nitric acid process. According to EPA, the commenter notes that secondary and tertiary controls are the most commonly applied, and "[t]hese technologies are capable of and typically achieve greater than 80 percent reduction of N2O emissions."
The commenter states that it is indisputable that nitrous oxide control technologies are technically feasible. The commenter notes that a broad array of controls have been widely used in the nitric acid industry for a long time, both in the U.S. and abroad. If EPA had analyzed these technologies during its review, the commenter believes that EPA would have had many different options to choose from in determining which technology constitutes BDT for purposes of a nitrous oxide performance standard. Indeed, the commenter notes that EPA has made no attempt to argue that nitrous oxide control technologies are not technically feasible. According to the commenter, EPA's passing comment that it has "limited data" concerning nitrous oxide controls is simply not accurate in light of its own December 2010 white paper and its reliance on 24 recent technical papers and other reference materials in developing that document. In the absence of any explanation, the commenter does not understand why EPA failed to include this information in the administrative record and failed to propose an emission standard for nitrous oxide. Nevertheless, the commenter asserts that EPA's failure to do so is clearly unlawful.
Comment 3.17:  Commenter (0099) states that it is not justifiable for the EPA to set an NSPS emission standard for NOX alone. By establishing an emission standard based on SCR as the BSER, the commenter believes that the EPA has failed to account for the control technology currently being used by existing nitric acid plants. The commenter notes that nonselective catalytic reduction and SCR combined with secondary or tertiary catalysts control both for NOX and N2O.
The agency asserts several reasons for its determination that SCR is BSER for NSPS for nitric acid plants, the chief reason being that "NOX emissions from nitric acid trains using SCR are lower than nitric acid trains using NSCR." However, the commenter believes that this does not negate the fact that SCR simply does not control for N2O and disregards the significant emission reduction NSCR is capable of. For example, while EPA asserts some NSCR facilities have shown difficulty in meeting the 0.50 lb NOX/ton limit, the commenter notes that the EPA has data from 2 other facilities (Dyno Nobel-Deer Island and JR Simplot-Helm) using NSCR that achieved monthly block averages between 0.15-0.36 lb NOX/ton acid. According to the commenter, this suggests that NSCR technology is more than capable of meeting the agency's new revised NSPS emissions limit and controlling harmful emissions of N2O. 
         While we understand the difference between a block average and rolling average, the block averages from the Dyno Nobel-Deer Island and JR Simplot-Helm plants support the conclusion that the EPA has set an insufficiently protective standard for NOX. EPA has not explained why the distinction between a block average and a rolling average would be determinative or even relevant on this point.] It is important to note that the agency itself noted that NSCR is one of several air pollution control technologies [that] are effective in controlling NOX emissions. 
The commenter urges the EPA to analyze the potential for both NOX and N2O emission controls, and consider all emissions reductions systems, including those discussed in the EPA's 2010 white paper. The commenter notes that the EPA may not simply ignore an important pollutant when revising the standard. By not requiring new nitric acid production units to control for N2O, the commenter states that the EPA is failing to meet its statutory obligations, squandering an opportunity to take a substantial step in reducing the amount of GHGs emitted into our environment, and leaving the public vulnerable to harmful air pollutant for years to come.
Comment 3.18:  Commenter 0103 states that the EPA should consider the interactive effects that controlling N2O will have on the reduction of NOX. The commenter states that the costs and benefits of additional NOX reductions should then be included in any cost‐benefit analysis that EPA conducts with regard to N2O. According to the commenter, following this approach will decrease costs and increase benefits of N2O regulation, resulting in net public health and welfare improvement. Because of these interactive effects, the commenter asks that EPA set a N2O standard at the same time EPA revises the NOX standard. By setting concurrent standards, the commenter believes that EPA can maximize net benefits across both pollutants. In most cases, the commenter notes that it is more cost effective for plants to design emissions reductions as part of the initial construction rather than adding them to existing plants. Therefore, as new or modified plants are faced with added capital expenditures to meet a more stringent NOX standard, the commenter believes that it would be cost‐effective for them to simultaneously meet a N2O standard, rather than requiring expensive capital investment to comply with future N2O regulation.
Comment 3.19:   Commenter 0104 points out that the NSCR technologies reviewed by EPA were installed twenty years ago or longer, and the agency has not considered the potential for NCSR options available today to meet the NOX standard it has proposed at a reasonable cost. The commenter feels that it is important for EPA to complete this evaluation before promulgating a final rule. If NSCR can meet or exceed the NOX standard EPA has proposed at a reasonable cost, the commenter believes that the agency can and should simultaneously establish nitrous oxide limits based on this technology.
Comment 3.20:   Commenter 0104 states that nitrous oxide controls for nitric acid plants are technically feasible and can be implemented at a reasonable cost. The commenter states that it is clear that control technologies for N2O are economically reasonable. According to the commenter, the EPA sets forth detailed data for three types of N2O emission control technology which shows that they are very cost-effective in a 2006 report, and reiterated its conclusions in its 2010 white paper. The commenter also notes that the European Commission's BAT analysis for various catalytic N2O reduction strategies showed that these technologies are very cost-effective.
According to the commenter, even looking only at N2O reductions (rather than CO2 equivalent), the technologies would be cost-effective. The commenter notes that the low end of the range presented by EPA would involve only $620 per ton in capital costs and $43 per ton in operating costs. The commenter notes that cost-effectiveness of nitrous oxide controls is further demonstrated by the fact that they have been in use, not only in Europe and the United States, but also at nitric acid plants in developing countries, such as China, Pakistan, Chile, South Africa, South Korea, and Egypt.
According to the commenter, these costs are significantly lower than costs that EPA has found reasonable for reducing other pollutants. Accordingly, based on EPA's own data and analysis, the commenter feels that it is beyond dispute that the costs of nitrous oxide control technologies are reasonable. Indeed, the commenter states that nitrous oxide controls are among the most cost-effective for any type of pollutant. According to the commenter, the EPA has made no attempt to argue that nitrous oxide control technologies are not economically reasonable, and EPA could not justify such an argument if it tried to do so. Accordingly, EPA's failure to even consider the costs of nitrous oxide controls and its accompanying failure to establish any standards for this pollutant, as assessed by the commenter, are clearly unlawful.
Response to Comments 3.1-3.20: For the reasons discussed in Section V of the final rule preamble, we are not taking any final action with respect to a GHG standard for nitric acid plants at this time. The EPA is in the process of gathering and analyzing additional data on GHG emissions from NAPUs that will allow the Agency to continue working towards a proposal for GHG standards for nitric acid plants. These comments will be considered as we develop a proposed rule on a GHG standard for nitric acid plants.
   4. Extension Requests
Comment 4:  Commenters (0086, 0087, 0088, 0089, 0090) asked the EPA to extend the comment period until January 12, 2012.
Response to Comment 4:  Due to the short time period between proposal and the consent decree deadline for signature of the final rule, an extension of the comment period was not granted.
   5. Data Issues
Comment 5.1:   Commenters (0093, 0095, 0098) state that the NAPs EPA relied upon to set the proposed NOX NSPS are not representative of the industry. The commenters assert that EPA must establish that an NSPS is achievable under the range of relevant conditions that affect the emissions to be regulated. Thus, when using existing sources to produce data upon which EPA seeks to rely to support setting an NSPS at a certain level, the commenter states that it is incumbent on EPA "to consider the representativeness" of the test sources for the industry as a whole, particularly in a case such as this one where EPA's "central argument is that the standard is achievable because it has been achieved (at the tested plants)."
Commenter 0098 states that EPA has not given due consideration to identifying and evaluating the variable conditions that impact emissions at NAPs. Instead, according to the commenter, EPA simply has relied on test results of NAPs that are not representative of the industry, and thus, that do not demonstrate that the proposed NSPS is achievable. The commenter notes that EPA's proposed NOX NSPS for NAPs violates the rules set out in National Lime. Therefore, the commenter requests that EPA revisit and expand its analysis to include consideration of all of the variable conditions in the industry (as well as all of the emissions data available to EPA that was not considered), and to create NSPS limits that reflect these conditions.
The commenters note that the Test Plants are nonrepresentative of the industry as a whole. According to the commenters, two of these NAPs (PCS Geismer 5 and El Dorado Nitrogen) are "sister plants" and represent the state of the art, as they were the most recent newly constructed NAPs in the industry. According to the commenters, these two NAPs are large, modern, dual-pressure process facilities that can achieve low NOX emissions during all modes of operation, including during startup, shutdown, and malfunction (SSM) events, unlike most NAPs. Further, unlike the rest of the industry, the commenters note that each of these Test Plants employ not only SCR, but also includes at least one abatement technology in addition to SCR. For example, in addition to utilizing extended absorption and SCR, the commenter notes that the Agrium North Bend plant uses HPI as an additional control, which is typically not used in the industry. According to the commenters, many producers have safety and operational concerns regarding the use of HPI. The commenter states that given these concerns, HPI should not be considered BSER, and the data from Agrium North Bend is not representative of emission reductions achievable by the industry as a whole. The commenters note that all three of the plants utilize the largest forms of extended absorption to increase efficiency and, thus, lower NOX emissions, which is not typical in the industry. Unlike the industry, the commenters note that all of these facilities can control NOX emission during startup and shutdown and remain in compliance with their current permit limits.
According to the commenters, the atypical nature of the Test Plants is most distinct when compared to NAPs (typically older plants) that employ NSCR. The commenters note that not only do none of the Test Plants use NSCR, but, as EPA long ago recognized, NSCR is not an end of the pipe control technology, as some would describe SCR.
According to the commenters, a NAP that employs NSCR is designed differently because the NSCR is built into, and is an integral part of, the operation of the NAP, particularly in terms of heat balance and energy generation for other parts of the NAP. The commenters note that NAPs with an NSCR must keep temperatures elevated in order to create the reductive environment within the unit that eliminates both NOX and N2O. In order to ensure that temperatures and pressures are adequate for this reductive environment, the commenters note that NSCR is placed upstream of the tail gas expander. According to the commenters, these expanders will malfunction above the temperatures they are designed to handle, and because NSCR reactions are extremely exothermic, any increased NOX destruction and associated temperature increases related thereto could cripple the expander, and thus, the entire NAP. When a NAP with NSCR increases its ability to reduce NOX emissions, the commenters note that it also results in the increased release of other gases, such as CO and NH3. According to the commenters, the Test Plants do not face these conditions, limitation, or the ability to modify their operations, or their attendant impacts on emissions.
Commenters 0093, 0095, 0098 state that the ICR data provided by the two NSCR facilities evaluated by EPA could not achieve the proposed standard. Commenter 0098 states that EPA evaluated two separate NSCR facilities: PCS Geismar Train 4 and Agrium Sacramento, which demonstrated emissions of 2.41 lb NOX/tHNO3, and 1.29 lb NOX/tHNO3, respectively, on a 30-day rolling emission rate. According to the commenters, these results led EPA to state that "whether NSCR can meet the levels achievable by SCR over a long term, is uncertain." Yet, the commenters state that EPA nevertheless established the same limit for all NAPs  -  0.50 lb NOX/tHNO3  -  newly constructed, modified, or reconstructed, regardless of control technology employed. According to the commenters, there are major differences in NAP design, which in turn create a wide range of operating temperatures, operating pressures, process configurations, and controls. According to the commenters, all of these variables impact plant efficiency and NOX emissions and controls. The commenters believe that the diversity in NAP design and operation, and the diversity in the use of control and performance of control technologies, does not lend itself to a single, uniform standard. Yet, the commenters believe that all of these facts were ignored and not given due consideration.
According to the commenters, the EPA's simple adoption of the Test Plants' results suffers from a potentially fatal assumption that the Test Plants are representative of the industry and that newly constructed plants will be able to reach efficiencies similar to those of the Test Plants. The commenters state that NAPs are not constructed on a single design, but rather are tailored to meet specific production and acid strength requirements. The commenters also state that smaller NAPs cannot attain the same levels of efficiency as large plants without extremely high incremental costs (relatively speaking). Importantly, the commenters note that these costs cannot be offset by increased production. Thus, the commenters assert that it is likely that a smaller new plant, designed in a state of the art manner, may not be able to meet the proposed NSPS.
Comment 5.2:  Commenter (0091) states that if it is possible to gather sufficient representative data reflecting emissions during malfunctions, then EPA is obligated to consider these data in its calculations of NSPS for steady-state operating conditions (since EPA asserts malfunctions are not a separate operating mode). To the extent EPA had access to continuous monitoring data for emission units using technology on which the Proposed Standards are based that included malfunction periods, the commenter states that EPA could have conducted analyses of emissions levels during malfunction events. The commenter suggests that even if the continuous monitoring data are for parameters not regulated by the Proposed Standards, analysis of monitoring data for those other parameters during malfunction events might form a reasonable basis for EPA's assessment of what standards are achieved or achievable during malfunctions. Also, according to the commenter, many types of sources are required by many state agencies to submit deviation reports or malfunction reports when they experience a malfunction that causes an exceedance of an applicable limitation. According to the commenter, the EPA does not appear to have made any attempt to obtain and analyze such reports, in order to assess what type of requirement might reasonably apply to the subject emission units during malfunctions.
Comment 5.3:   Commenter (0098) states that EPA failed to consider other data in its possession, selectively sought out CEMS data, and erred in analyzing the CEMS data it did evaluate. According to the commenter, the EPA received CEMS data from four NAPs with SCR, but claimed to only have received data from three NAPs (the Test Plants it used to support the proposed NSPS). Specifically, the commenter's review of the docket indicates that in addition to receiving CEMS data from PCS Geismar 4 and 5, which are referenced in the preamble, EPA also received CEMS data for PCS Geismar 3. Despite this, the commenter asserts that EPA failed to include an analysis of PCS Geismar 3 (which has an SCR) in its calculations to set the new NSPS. The commenter questions why EPA neglected to include PCS Geismar 3 in its analysis, and are concerned that the data were omitted simply because they did not support the 0.50 lb NOX/tHNO3 conclusion.
The commenter analyzed the CEMS data for PCS Geismar 3 using comparable methods used by EPA on the three SCR units presented in the preamble. According to the commenter, the data were processed from the PCS Geismar 3 raw data, just as the proposed Subpart Ga requires. The data, however, did not have any indicators for when the plant was operational, so the commenter states that data were excluded when the emission factor was less than 0.01 lb NOX/tHNO3, which is a reasonable indicator of periods of nonoperation. According to the commenter, the 99th percentile of this data represented emissions of 2.55 lb NOX/tHNO3. The commenter provided a figure that showed analysis results for PCS Geismar 3 would only meet the proposed 0.50 lb NOX/tHNO3 limit about 30 percent of the time, on a 30-day rolling basis. 
It appears to the commenter that EPA selectively chose to obtain and analyze emissions data from sources that had emissions results that supported a preconceived NSPS limit, and ignored (in the case of PCS Geismar 3), or refused to obtain and analyze data from other NAPs. According to the commenter, this course of action seriously calls into question EPA's conclusion that its proposed NSPS is achievable.
Comment 5.4:   Commenter 0098 notes that upon an independent review, it appears that EPA's analyses of PCS Geismar 5 inaccurately calculated (and subsequently reported) the plant's 30-day rolling emission rate. The commenter notes that the same methodology was used for the Agrium North Bend and El Dorado Nitrogen data. 
According to the commenter, there are two errors present within the EPA's calculations. First, although EPA was provided with CEMS data from PCS Geismar 5 for years 2007, 2008 and 2009, the commenter notes that the EPA only analyzed data from 2009. An independent calculation by the commenter of all of the data from 2007, 2008 and 2009 reveals that PCS Geismar 5 emits more than the EPA's suggested 0.38 lb NOX/tHNO3. The commenter provided the results of this calculation (The table provides the percentile calculations of PCS Geismar 5's CEMS data from 2007  -  2009). The figure provided by the commenter shows a statistical comparison of PCS Geismar 5's emissions by averaging period for 2007  -  2009. Because EPA incorrectly determined the 99th percentile 30-day rolling emission rate factor for PCS Geismar 5, the commenter believes that this further calls EPA's proposed NSPS into question. 
Second, while appropriately attempting to exclude data from non-operating days (i.e., removing the zeros in its calculations), the commenter notes that the rolling emission rates the EPA relied upon for PCS Geismar 5 contained non-operating days. According to the commenter, this error resulted in many 30-day rolling emission rates consisting of far less than 30 operating days and resulted in skewed data that allowed each day in these calculations to carry more weight than they otherwise should in a 30-consecutive operating day limit. Thus, according to the commenter, the EPA's calculations do not support the Agency's proposed NSPS in Subpart Ga. The commenter provided an example of this error in a table in the text. The table presents a screenshot of the EPA analysis of the CEMS data for PCS Geismar 5. The commenter points out that the 30-day rolling emission rate calculation includes only 22 operating days.
   Comment 5.5:   Commenter 0098 adds that it appears that EPA failed to apply the proposed calculation methodologies in proposed Subpart Ga Section 60.75a to its analysis of the CEMS data from Agrium North Bend. Instead, it appears to the commenter that the data were calculated using the current Subpart G Section 63.73(b) conversion factor. Based on independent tests of member CEMS data, the commenter stated that the current Subpart G conversion factor and the proposed calculation equations will yield significantly different results. According to the commenter, this further undercuts EPA's proposed NSPS; EPA cannot require compliance with a proposed NSPS based on an untested compliance calculation.
   Response to Comments 5.1-5.5: The EPA agrees with the commenter that there were calculation errors in the analysis of data from PCS Geismar Train 5. These errors have been corrected with little to no impact on the analyses (from 0.378 lb NOX/ton acid to 0.375 lb NOX/ton acid). The corrected spreadsheet is located in docket EPA-HQ-OAR-2010-0750.
   
   In response to the commenter's reference to Train 3 at PCS Geismar, the reasoning for excluding the train from the analysis was inadvertently left out of the memo. The memo (Statistical Evaluation of CEMS Data to Determine the NOX Emission Standard (Updated Memo for Final NSPS)) has been updated to include the reasoning, which is twofold. First, the train only operated for 15% of the year. The EPA does not consider a year with such a low yield to be a representative plant for purposes of the analysis.

Second, this nitric acid plant is not using a best performing SCR. According to a telephone conversation between EPA's Bill Neuffer and Jeremy Richard of PCS Geismar, the SCR is located after the gas expander and is thus an atmospheric SCR. According to Mr. Richard, this prevents optimal mixing of stack gas and ammonia. He felt that the unit could be improved to meet a lower NOX emissions limit on a consistent basis.
Only data from the most recent year was considered in the analyses. Some facilities only submitted one year of data so we only considered one year of data from all submissions so as not to bias the analysis toward a plant with more data.
In response to the commenter's reference to the use of hydrogen peroxide injection (HPI) at Agrium North Bend, the EPA conducted further analyses which show that the limit is achievable when the summer months (the period during which Agrium North Bend utilizes hydrogen peroxide injection) are excluded from the analyses. The analyses are contained in the memo entitled Agrium North Bend Analyses, located in docket EPA-HQ-OAR-2010-0750. Assuming that ozone season (May through September) represents summer months, the analyses shows the maximum 30 operating day emission rate for the year as 0.41 lb NOX/ton acid, for the summer months as 0.40 lb NOX/ton acid and for the non-summer months as 0.39 lb NOX/ton acid. The results of additional analyses are shown in Section V of the final rule preamble.
Comment 5.6:   Commenter 0098 states that using a 99th percentile calculation in an attempt to remove statistically insignificant emission events at the NAPs effectively removed these events from their limit evaluation. Although EPA's calculations are said to include startup and shutdown events, according to the commenter, the exclusion of the top one percent of these averages effectively removed these events from their limit evaluation. Further, the commenter notes that by excluding the top one percent of the data, EPA recognizes that one percent of these rolling averages will be in violation of the proposed NSPS, yet the NAPs are required to be in compliance at all times. According to the commenter, removing the highest readings is the opposite of what is needed to make an achievable standard. The commenter requests that EPA correct the statistical analysis by using an approach that excludes the lowest one percent of the data instead of the highest one percent.
Comment 5.7:  Commenter 0096 states that according to the proposed calculation of the daily mass emissions of NOX, a Nitric Acid Plant must calculate emissions based on the number of operating hours in the day. The commenter notes that emissions are calculated by summing the daily mass emissions and dividing that by a sum of the daily production values. Typically, according to the commenter, startups and shutdowns occur during a portion of a day, and startups and shutdowns can occur multiple times per month. According to the commenter, utilizing the number of "operating hours per day" vs the number of "hours per day" (i.e. 24) skews the daily average (and the 30 day average) by giving more weight to higher emissions during the event(s). According to the commenter, the method of calculation proposed does not account for hours when there are no emissions so a method of calculation that does not include such a bias should be developed.
    Response to Comments 5.6-5.7: The EPA has reanalyzed the data using a more representative method of calculating the 30 day rolling emission rate that sums all of the operating hours rather than the operating days. This method addresses the commenters concerns related to weighted emission rates. See Section V of the final rule preamble for further discussion of the level of the standard.
    Additionally, the EPA agrees with the commenter's assertion that our analysis excluded the top one percent of the data so that one percent of these rolling emission rates could be in violation of the proposed NSPS for the Agrium North Bend facility using the proposed calculation for determining emission rates. However, as shown in the memo Statistical Evaluation of CEMS Data to Determine the NOX Emission Standard (Updated Memo for Final Standard), all three facilities that use SCR as a NOX control have rolling emission rates lower than or equal to 0.50 lb NOX/ton acid at all times using the updated calculation method. Also see Section V of the final rule preamble for further discussion of the level of the final standard.
    Comment 5.8:   Commenter 0098 asks why the EPA did not request CEMS data from all facilities that have installed CEMS equipment. The commenter had CEMS installed at nine NAPs based on a Consent Decree with the EPA, so the EPA should have been aware of these NAPs. Yet the commenter did not receive a CAA §114 request for CEMS data from these NAPs.
    Moreover, the commenter notes that an EPA document from 2009 sets out the NAPs with CEMS (which constitute 58 percent of the NAPs in the industry). Despite being well aware of what NAPs had CEMS (and what controls were employed at each NAP), and being able to obtain relevant data from these NAPs, the commenter notes that the EPA appears to have selectively chosen the NAPs from which to request CEMS data during its CAA §114 request process for this rulemaking.
    Response to Comment 5.8: Requesting data from every NAP with CEMS would have been an expensive and time-consuming exercise for the nitric acid industry and the agency. For that reason, there are limitations on the amount of data that was gathered through the CAA §114 request supporting the proposed rule. Sending a Section 114 request to every nitric acid plant with NOX CEMS was not a viable option for gathering data in a timely manner. Therefore, the EPA gathered data from the best performing nitric acid plants in the United States. The EPA also solicited industry's input on the selection of these plants.
    Also, industry had ample opportunities to voluntarily submit CEMS data, as late as the end of the comment period (November 28, 2011) for this NSPS. No additional CEMS data were received during the public comment period. Additional data were received shortly after the close of the comment period and these data did not change the results of our analysis. See the memo entitled Analyses of Data Received Between Proposal and Promulgation of Part 60, Subpart Ga, for more information on the data received after the public comment period. The EPA used the best available data when doing our analysis.
   
   6. Modifications and Reconstructions 
Comment 6:   Multiple commenters disagree with EPA's statements concerning the lack of modifications and reconstructions of existing nitric acid plants that have occurred since 1971.
Commenter 0098 notes that despite new Subpart Ga applying to construction, modification, or reconstruction of a NAP, the EPA summarily dismisses the possibility that existing sources would undergo a modification or reconstruction triggering Subpart Ga applicability and, as a result, the Agency does not evaluate the achievability of the proposed 0.50 lb NOX/ton acid standard for modified or reconstructed NAPs. Instead, as the commenter quotes the EPA, "it is our understanding that [nitric acid producers] would build new production units rather than modify or reconstruct existing units." The commenter is unsure how EPA reached this conclusion and requests that EPA explain its reasoning.
Commenter 0094 notes that the EPA incorrectly stated in the preamble that no existing nitric acid plant has been modified since 1971. The commenter references a Consent Decree that the commenter entered into with the EPA in June 2011 and notes that the Complaint filed concurrently with the Consent Decree specifically alleged the modification of an existing nitric acid plant. The commenter states:
         see United States, et. al. v. Terra Industries Inc., et. al., Civil Action No. 11-4038 (U.S. Dist. Ct. N. Dist. Iowa) in the Complaint at page 25: "111. After publication of the proposed standard of performance for nitric acid production units, on or about September of 1992, Port Neal and Terra Industries commenced construction of a "modification" of an "existing facility" as those terms are defined in the NSPS, 40 C. F.R. §§ 60.2, 60. 14, at Port Neal #1 nitric acid production unit at the Port Neal Facility .... The modification increased the acid production capacity of the Port Neal Facility  ...."
According to the commenter, the EPA has knowledge that existing nitric acid plants can and have been modified since Subpart G was promulgated.
Commenter 0098 feels that the EPA's assumptions regarding the "modification" of existing sources that may trigger the new standard will lead to a standard that is arbitrary and capricious. The commenter notes that constructing a new NAP is more expensive than modifying an existing NAP. Because of the expense of construction materials and equipment, the commenter asserts that NAPs are frequently moved throughout the world, and are typically upgraded instead of building a new NAP. Thus, from an economic perspective, the commenter believes that EPA's assumption of no modifications or reconstructions of existing NAPs appears ungrounded.
Commenter 0093 urges the EPA to recognize that the standard could create a disincentive for facility modification due to difficulties in meeting such a standard. According to the commenter, the modifications to older nitric acid plants needed to achieve the proposed low NOX limits may be technically infeasible and prohibitively expensive as they might require reconfiguration of the design of the plant. The commenter notes that these facilities may have to make significant changes to ensure that the level of NOX exiting the absorber is low enough for tail gas NOX control equipment.
Commenter 0098 states that the EPA must evaluate the achievability of the proposed standard to modified and reconstructed NAPs in addition to new NAPs. According to the commenter, the EPA is incorrect in its conclusion that existing NAPs will not undergo modifications or reconstructions and, as a result, the Nitric Acid NSPS Proposal fails to (1) discuss the achievability of the 0.50 lb NOX/ton HNO3 standard for modified or reconstructed NAPs, and (2) evaluate the costs associated with such a modification or reconstruction. As such, before applying Subpart Ga to modified and reconstructed NAPs, the commenter asks that EPA perform the requisite CAA § 111 analysis.
According to Commenter 0098, the NSPS modification trigger is a low hurdle for sources and any number of minor changes at an existing NAP could subject it to Subpart Ga. The commenter notes that existing NAPs frequently undergo projects that could result in an increase in NOX emissions, subjecting the units to Subpart Ga. 
Response to Comment 6: One of the commenters requested that EPA explain this statement from the preamble: "it is our understanding that [nitric acid producers] would build new production units rather than modify or reconstruct existing units." 
During multiple meetings in 2009, industry representatives indicated that new nitric acid plants were unlikely to be built and that growth in the industry would be managed through modifications of existing units. See EPA-HQ-OAR-2010-0750-0022 and EPA-HQ-OAR-2010-0750-0017. Despite requests for specific information about the plants that had been modified, including costs and NOX emissions reductions, no other information was ever made available to the EPA prior to proposal. 
On June 3, 2010, the EPA had a conference call with The Fertilizer Institute to discuss subcategorization of low and medium pressure units from high pressure or dual pressure units, especially related to modifications or reconstructions of existing nitric acid plants. This conference call is documented in the docket. See email entitled Short Summary of Teleconference with TFI re: Subcategorization/Nitric Acid plants from Bill Neuffer, dated June 3, 2010.
During the course of the phone call, the EPA asked when the NSPS had been triggered by modifications or reconstructions. The response was that there had been no recent modifications that triggered NSPS but more information would be gathered. No further information on modifications or reconstructions was received by the EPA prior to proposal. However, the EPA did have information on the planned construction of a new nitric acid plant in Idaho. A permit to construct was issued for Southeast Idaho Energy, which will include a new nitric acid plant, in November 2009. With no information on the existence of modifications to nitric acid plants during the previous three year period, the EPA was left to presume that any future growth in nitric acid capacity would be met through construction of new nitric acid plants, like the one at the Southeast Idaho Energy facility.
With respect to the commenters' statements about the June 2011 consent decree for Terra Industries, the EPA disagrees that this information could have been used for this rulemaking. The timing of the consent decree makes it highly unlikely that any changes to the proposal could have been made and evaluated prior to the court-ordered proposal date.
However, now that some information on modifications has been shared by industry, the EPA agrees that the commenters' concerns about the achievability of the standard by nitric acid plants that have been modified or reconstructed warranted further analysis. The commenters identified a few nitric acid plants that fit those definitions, and we performed further evaluation of the NOX CEMS data for such plants. Based on our technical evaluation of the data, we have determined that the final standard is achievable for modified or reconstructed sources. See Section V of the final rule preamble for discussion of the achievability of the standard. Also see Section VI of the final rule preamble for discussion of the cost impacts of the standard for a modified plant.
 7. Subcategorization
Comment 7.1:  Commenters (0091, 0093, 0095, 0097, 0098) ask EPA to subcategorize nitric acid facilities based on controls, size and/or process, and have different requirements for each. The commenters recommend separate standards for small, single pressure nitric acid plants.
Commenter 0098 states that under CAA §111, EPA has the authority and discretion to create and distinguish among classes, types, sizes, of sources when establishing a NSPS for a source category. The commenter believes that the information provided in its member survey, the information gathered by EPA through the CAA § 114 process for this rulemaking, and the information EPA has gathered during its enforcement initiative demonstrates that there are differences between existing NAPs in terms of design, process, size, and operation, as well as between newly constructed NAPs, and older NAPs that directly and substantially impact a NAPs ability to meet a uniform standard. The commenter recommends that EPA evaluate and adopt NSPS limits in three categories: (1) newly constructed NAPs; (2) existing NAPs without NSCR that undergo a modification or reconstruction; and, (3) existing NAPs with NSCR that undergo a modification or reconstruction. While the commenter does not offer specific recommendations regarding NSPS limits within these categories, with respect to the latter two categories, the commenter refers EPA to its TFI Member Survey, which provides information on all NAPs in the industry.
Commenters (0093, 0095) state that if a small nitric acid facility utilizing mono-pressure needed to make a modification, it would be extremely unlikely that such a facility would be able to meet the proposed 0.50 pounds of NOX per ton of nitric acid standard, even if it was possible to install an SCR. According to the commenters, these facilities may have to make significant changes to ensure that the level of NOX exiting the absorber is low enough for tail gas NOX control equipment. As such, commenter 0093 recommends that EPA collect and analyze data from smaller nitric acid plants utilizing mono-pressure and either NSCR or SCR, which reflects the typical nitric acid facility in chemical service. Commenter 0095 notes that Nitric acid plants that currently utilize mono-pressure would have trouble meeting the proposed 30-day rolling emission rate even with similar air emission control equipment because the frequency of replacing the precious metal catalyst increases as the operating pressure increases, leading to more unplanned shutdowns and startups.
Commenter 0097 points out the differences between the dual pressure units EPA used for determining the new standard are fundamentally different from the single pressure systems often employed at chemical plants for process use. According to the commenter, dual pressure systems would not be economically feasible for smaller capacity plants being used to generate intermediate chemicals for use in on-site chemical plants. The commenter states that peroxide addition systems would be inherently unsafe for many typical chemical plant operations. The commenter notes that peroxide addition may also require high absorber pressure (which may not be possible for some existing single pressure plants) and may also require a new chiller system (which adds cost). The commenter asserts that smaller plants (typical for chemical plants) would not have available steam capacity for SCR preheat (such as at EI Dorado) and thus would need to build air heaters for SCR preheat (costs which EPA has not considered).
Commenter 0097 asks that EPA have a separate standard for small, single pressure nitric acid plants, both new and existing. The commenter states that EPA should evaluate data from small, single pressure plants using SCR without peroxide addition to establish a standard for new or existing small, single pressure plants. One of the commenter's existing small, single pressure units with SCR is performing in the range of 1.2 - 1.5 lb NOX/ton acid on average, including startups and shutdowns.
Comment 7.2:  Commenters (0091, 0093, 0095) state that EPA's assumption that the 30-day average emission limit that includes periods of startup and shutdown could be consistently met is flawed and the commenter believes a higher emissions limit is warranted for these operations that are integrated with downstream production process. Commenters 0091 and 0093 state that the Agency assumes that startup and shutdown events will be infrequent, and that the 30-day rolling emissions will fall below the proposed emission limitation. According to the commenters, large commercial nitric acid producing plants may have very few startups and shutdowns, but smaller plants integrated with downstream production processes incur frequent startups and shutdowns, and therefore will have higher 30-day rolling NOX emissions. The commenters note that more frequent startups mean more thermal cycling of high temperature equipment, which leads to shorter equipment life and failures, which in turn can increase the load on the NOX control equipment. Commenter 0093 notes that EPA has suggested HPI as a way to reduce emissions during startup, during some cases, but it may not be a viable option for chemical plants where the addition of peroxide could negatively impact acid quality for process use and pose additional safety hazards during downstream processing. Commenter 0091 understands that CEMS data that was submitted to EPA for the development of the Proposed Rule confirms this distinction. According to the commenter, EPA cannot, under section 111, promulgate an NSPS applicable to such smaller nitric acid plants that are integrated with downstream production processes based on an assumption of infrequent startup/shutdown that is inaccurate for that type of facility.
Comment 7.3:  Commenter 0097 disagrees with EPA's assertion that modified or reconstructed sources will be able to meet the proposed limit. The commenter states that facilities with existing small, single pressure units (if reconstructed or modified) which currently employ SCR (BSER) without the appropriate additional process technology (e.g. dual pressure design) may be required to install a new dual pressure unit (versus a retrofit) in order to comply.
Response to Comments 7.1-7.3: EPA disagrees that subcategorization is required or appropriate. As described in Section V of the final rule preamble, we have decided to promulgate a limit of 0.50 lb/ton based on our technical evaluation of the data available to the agency. The commenters have not offered data that would support changing the emission limit beyond this 0.50 lb/ton level. The data also do not support establishing a subcategory.
We reviewed data submitted by the American Chemistry Council (ACC) after proposal. The NOX emissions were higher than the final emission limit. However, none of the data were from nitric acid plants controlled by BSER. Therefore, these data have not impacted the level of the standard in the final rule. More details are contained in Docket Number EPA-HQ-OAR-2010-0750. See Teleconference with American Chemistry Council on January 11, 2012, and Analyses of Data Received Between Proposal and Promulgation of Part 60, Subpart Ga.
The EPA disagrees that information from the EPA's enforcement initiative or the TFI Member Survey must drive the level of the standard in the nitric acid NSPS review. According to the meeting minutes from a meeting with TFI on June 30, 2009, EPA staff stated that "[the enforcement initiative] focused on the high emitters and would not likely provide the information needed for NSPS review." Section 111 of the CAA requires that NSPS reflect the application of the best system of emission reductions which (taking into consideration the cost of achieving such emission reductions, any nonair quality health and environmental impact and energy requirements) the Administrator determines has been adequately demonstrated. Focusing on the high emitters does not demonstrate the best system of emission reductions.
In order to be useful for EPA's analysis, the TFI Member Survey should have included facility-specific information, which was never supplied. A summary of findings across the surveyed facilities does not demonstrate the best system of emission reductions.
The Agrium North Bend nitric acid plant, which is described further in Section V of the final rule preamble, is a small single pressure operation which has also been modified or reconstructed within the past 10 years. Based on the re-analysis of the CEMS data for this modified or reconstructed source and other single pressure operations supplied by ACC, we conclude that 0.50 lb NOX/ton acid is achievable for nitric acid plants that undergo modification or reconstruction. See memos entitled Statistical Evaluation of CEMS Data to Determine the NOX emission Standard (Updated Memo for Final Standard), Agrium North Bend Analyses, and Analyses of Data Received Between Proposal and Promulgation of Part 60 Subpart Ga.
We have updated our cost analysis to include the costs for installation of an SCR if an existing facility is modified or reconstructed and triggers NSPS subpart Ga. The capital cost for an SCR installed was estimated to be roughly $2,700,000 ($370,000 annualized cost, assuming a 20 year capital recovery period). This facility achieved emissions reductions of nearly 300 tons of NOX per year. The cost effectiveness for the addition of this control device was roughly $1,200 per ton of NOX. The costs for installation and operation of this control device are reasonable and cost effective. See memo entitled Impacts of Nitric Acid NSPS Review-NOX (Updated Memo for Final NSPS). See also Section VI of the final rule preamble. 
According to the CEMS data received from the Section 114 questionnaires, JR Simplot in Helm, California, Dyno Nobel in Deer Island, Oregon, and Coffeyville Resources in Coffeyville, Kansas are achieving low NOX emissions. All of these plants are monopressure nitric acid plants. The JR Simplot and Dyno Nobel Deer Island nitric acid plants are controlled by NSCR and the Coffeyville Resources nitric acid plant is controlled by SCR. Due to the low emissions from these monopressure nitric acid plants, subcategorization is not justified for this final rule.
The use of hydrogen peroxide injection at Agrium North Bend is summarized in Response to Comments 5.1-5.5. The full analyses are discussed in detail in the memo entitled Agrium North Bend Analyses.
   With regard to the cost issues raised by the commenters, refer to Response to Comments 10.4-10.9 and the memo Impacts of Nitric Acid NSPS Review -- NOX (Updated Memo for Final NSPS). See also Section VI of the final rule preamble for the estimated cost impacts of the final rule. The EPA considers these costs to be reasonable and supported by recent NSPS for other categories.

 8. Startup, Shutdown, and Malfunction 
8a. Numerical emission limitations should not apply during SSM
Comment 8.1:   Commenter 0091 asks that EPA justify applying the same emission limitations during startup and shutdown as during normal operations. The commenter states that there is ample precedent for EPA applying a different standard during SSM events. According to the commenter, both prior to the Clean Air Act Amendments of 1977 (which defined "emission limitation" and "emission standard" to be something that "limits the quantity, rate, or concentration of emissions of air pollutants on a continuous basis") and since that time, EPA has not required sources to meet NSPS emission limitations under CAA section 111 established for normal operations, during SSM events. See 40 C.F.R. § 60.8(c). The commenter notes that there is nothing in the legislative history of the 1977 amendments to the CAA that suggests Congress intended to overturn that practice. Rather, the commenter contends that the "continuous basis" language inserted in 1977 related to a debate in Congress about whether sources should be allowed to use temporary or intermittent pollution control technologies. The commenter notes that legislative history indicates Congress was aware that alternative emission limitations might at times be necessary, even though the emission limitations were established based on the capability of "continuous controls" like scrubbers. The commenter also notes court decisions both before and after the Clean Air Act Amendments of 1977 that have affirmed the appropriateness of including special SSM provisions in standards issued under section 111. According to the commenter, it is clear that the D.C. Circuit's Sierra Club decision does not compel or even support EPA's adoption of the SSM provisions (or lack thereof) in the Proposed Rule. The commenter notes that the decision said nothing about whether EPA could apply different emission limitations during SSM events, nor did it consider whether EPA could address SSM events through work practice standards.
Commenter 0091 states that an emission limitation that applies during SSM events does not meet the requirement of CAA section 111(a) that EPA establish "standards of performance" under that section that are "achievable" with the "best" emission control technology, "taking into consideration the cost of achieving such emission reduction, and any nonair quality health and environmental impact and energy requirements" that has been "adequately demonstrated" if in fact EPA has not demonstrated that the limitation is reasonably "achievable" with available technology during malfunction events.
The commenter asserts that the plain-language reading of the applicable statutory requirements is echoed by extensive case law. The commenter states the following:
         The Weyerhaeuser Co. v. Costle decision EPA cites in the preamble to the Proposed Rule, 590 F. 2d 1011 D.C. Cir. 1078), does not support EPA's position. See 76 Fed. Reg. 63,883 col. 3. In that case, the court was discussing a "technology forcing" standard, rather than one, like NSPS, that is to be based on what has already been demonstrated to be achievable. Also, the SSM events that EPA acknowledges are expected to occur at sources subject to the Proposed Standards are a far cry from the "'uncontrollable acts of third parties,' such as strikes, sabotage, operator intoxication, or insanity" that the Weyerhaeuser Court was considering in the passage quoted by EPA, see id. Industry is not requesting that EPA establish NSPS that can be met even during those exceedingly unusual types of events. Perhaps most importantly, the Weyerhaeuser decision came long before NRDC v. EPA, 859 F.2d 156 (D.C. Cir. 1988) which, as noted above, affirmed the need for an upset provision to address circumstances where compliance with effluent limitations is impossible through no fault of the permittee, and which endorsed Marathon Oil.
Commenter 0091 provides several options for NSPS that would apply during malfunction events:
   oo The commenter suggests EPA establish an emission limitation that applies at all times, but that has an averaging time of sufficient duration that short, infrequent spikes in emissions due to malfunctions would not cause the source to exceed the emission limitation (while at the same time ensuring that the source does not operate in a way that causes frequent, lengthy excursions above the normal controlled emission rate).
   oo The commenter suggests EPA use the flexibility accorded by CAA section 302(k) to address emissions during malfunction events through operational requirements rather than by applying the same limits on pollutant concentrations in exhaust gases that apply during normal operations. According to the commenter, EPA could have concluded that it has grounds to exercise its authority under CAA section 111(h) to promulgate a design, equipment, work practice, or operational standard, or combination thereof, because it is not feasible to prescribe or enforce an emission standard. The commenter also notes that EPA might use several of these approaches in combination.
According to commenter 0091, there is no indication in the Proposed Rule that EPA gave much, if any, consideration to these types of options. In short, the commenter believes that there are ample reasons to reject EPA's conclusory assertions that it cannot take malfunctions into account when setting NSPS for Nitric Acid Plants. According to the commenter, EPA's failure to evaluate these options thoroughly renders the Proposed Rule arbitrary and requires EPA to develop a new proposal.
Commenters (0091, 0092) state that EPA offers no explanation of its contradictory position that, even though it believes malfunctions are not a distinct operating mode, emissions during malfunctions should not be used to characterize the source's operating mode, which should be characterized only by a source's emissions during normal operations, and not "account for the innumerable types of malfunction events in setting emission standards." According to the commenters, asserting that malfunctions are part of normal operations, but then excluding emissions during malfunctions when determining emission limitations for the source's operation, makes no sense.
Commenter 0091 notes that the EPA acknowledges that properly designed and operated equipment will sometimes exceed emission limitations based on steady-state operation due to malfunctions. According to the commenter, the operators of these processes and equipment must treat malfunctions as very distinct events from steady-state operations, depending on the severity of the malfunction, requiring anything from shutdown of the unit to emergency fire response actions. In addition, the commenter notes that it is often impossible to gather emission data during malfunctions  -  either for standard-setting or for compliance-demonstration purposes. Malfunctions are by definition unexpected, so according to the commenter, it is not possible to plan to have test equipment in place to measure emissions when one occurs. Even if test or monitoring equipment is in place, the commenter notes that emissions during malfunctions often are not routed to a stack where they can be measured, and upsets during stack testing invalidate the test results under EPA's approved test methods.
Rather than supporting EPA's decision to ignore the fact that SSM events can lead to higher emissions even at well-operated facilities with the best control equipment, commenter 0091 notes that these findings should lead EPA to its authority under CAA section 111(h) to prescribe alternative design, equipment, work practice, or operational standards where it is not feasible to set or enforce a numerical emission limit. According to the commenter, EPA cannot rationally defend its articulated view that applying the concept of "best system of emission reduction which ... has been adequately demonstrated" is inconsistent with a source experiencing a malfunction. The commenter notes that EPA ignores that there are work practices  -  such as monitoring of operating parameters to identify a malfunction and stopping or cutting back the process accordingly  -  that represent the best practices for minimizing emissions during a malfunction. While the measures that represent these best practices will depend on facility-specific issues, such as process design, pollution control train, and other factors, the commenter notes that they are nonetheless the "best system of emission reduction which ... has been adequately demonstrated."
Commenter 0091 states that in the preamble to the proposed rule, EPA does not say that it is precluded from adopting different emission limitations that would apply during startup and shutdown periods. Rather, the commenter notes that EPA has acknowledged that NOX emissions are higher during startup and shutdown than during normal operations, but EPA "believe[s]" that sources can meet a 30-day rolling emission limitation that is based on performance achievable during normal operations, even if there are startup or shutdown events during those 30 days, because startup and shutdown events at nitric acid plants are "relatively short," meaning "generally" only a few hours. 
   As incorporated by reference, commenter 0091 argued in docket EPA-HQ-OAR-2009-0559 as follows:
   
         The commenter stated that EPA incorrectly claimed that its authority to prescribe unique standards for SSM periods is constrained by Sierra Club v. EPA, 551 F.3d 1019 (D.C. Cir. 2008). The commenter stated that EPA has failed to account adequately for emissions that occur during SSM periods. The commenter argued that the D.C. Circuit's Sierra Club decision manifestly is not binding on EPA's establishment of performance standards for SSI units under CAA sections 111 and 129. The commenter contended that the Sierra Club decision interpreted section 112, not section 129 (which incorporates by reference section 111), and pointed out that this interpretation is not merely a technical distinction. The commenter referred to the language that the D.C. Circuit considered dispositive in interpreting EPA's standards-setting authority under section 112 (the statement in the definition of emission limitation and emission standard that it limits the quantity, rate, or concentration of emissions of air pollutants on a continuous basis) as having been in the statute since 1977. The commenter pointed out that since 1977, EPA has exempted emissions during SSM events from compliance with NSPS under CAA section 111 (referenced 40 CFR §60.8(c)). The commenter argued that congress enacted the continuous basis language in section 302(k) knowing that EPA's emissions standards under section 111 exempted SSM periods, and pointed out that there is nothing in the legislative history of the 1977 amendments to the CAA that suggests congress intended to overturn that practice. The commenter referenced 551 F.3d at 1027, citing Kamp v. Hernandez, 752 F.2d 1444, 1452 (9th Cir.1985), and stated that it had nothing to do with limitations applied during SSM nor with EPA's established practice of exempting excess emissions during SSM events from its performance standards. 
         
         Also, as incorporated by reference, commenter (0091) argued that the Sierra Club decision does not prevent EPA from adopting emission standards for SSM periods that are different from those required during periods of normal operation. The commenter stated that while it is true that a blanket exemption from any standard may be inconsistent with the Sierra Club holding, the opinion does not prohibit EPA from applying different, even non-numerical, standards during SSM events from those standards that apply during steady-state operations. The commenter stated that the Sierra Club decision only rejected EPA's assertion that it had discretion to decide not to impose any emission standard whatsoever during SSM periods (referenced id. at 1027-28, 1030); and in fact, Sierra Club acknowledged that the definition of emission standard in §302(k) indicates that any one standard need not apply at all times.
     Also as incorporated by reference, commenter 0091's statement from docket EPA-HQ-OAR-2010-0600 is summarized as follows:
         According to the commenter, the proposed SSM provisions are not required in order to be consistent with Sierra Club v. EPA. The commenter notes that EPA suggests that its treatment of excess emissions during SSM events in the Proposed Standards is appropriate, perhaps compelled, in order to make the standards "consistent with" the D.C. Circuit's decision in Sierra Club v. EPA, 551 F.3d 1019 (D.C. Cir. 2008), cert. denied, 130 S. Ct. 1735 (2010), which vacated the exemption for excess emissions during SSM events contained in the 40 C.F.R. part 63, subpart A General Provisions for emission standards for hazardous air pollutants under CAA section 112. See, e.g., 75 Fed. Reg. at 65,074 col. 2; 65,075 col. 3 - 65,076 col. 1. However, according to the commenter, the D.C. Circuit's Sierra Club decision does not, however, compel or even support EPA's adoption of the Proposed SSM Provisions.
           
         According to the commenter, the Sierra Club decision did not say that the same emission limitations that EPA has derived for normal operations must also apply during SSM events. According to the commenter, there is ample precedent for EPA applying a different standard during SSM events. The commenter notes that the language that the D.C. Circuit considered dispositive in interpreting EPA's standards-setting authority under section 112  --  the statement in the definition of "emission limitation" and "emission standard" that it "limits the quantity, rate, or concentration of emissions of air pollutants on a continuous basis"  --  has been in the statute since 1977. Throughout that time, the commenter noted that EPA has not required sources to meet NSPS emission limitations under CAA section 111 established for normal operations during SSM events. See 40 C.F.R. § 60.8(c). In fact, the commenter noted that Congress enacted the "continuous basis" language in section 302(k) knowing that EPA's emissions standards under section 111 exempted SSM periods. According to the commenter, there is nothing in the legislative history of the 1977 amendments to the CAA that suggests Congress intended to overturn that practice.
         
         According to the commenter, the Sierra Club decision did not address whether EPA could use a "design, equipment, work practice or operational standard," as authorized under CAA section 112(h) and included in the definition of "emission limitation" and "emission standard" in CAA section 302(k), in lieu of a numerical emission limitation during SSM events. According to the commenter, EPA told the Court that the General Provisions SSM exemption struck down in Sierra Club was not an alternative standard based on that work practice, etc. authority. See 551 F.3d at 1028. Indeed, the commenter noted that EPA argued in that case that section 112(h) was irrelevant to its authority to exempt excess emissions during SSM events. Id. at 1030 (Randolph, J. dissenting). 
         
         Thus, the commenter stated that EPA cannot hide behind the Sierra Club decision as a justification for ignoring the inability of even the "best performers" to achieve during SSM events the emission limitations EPA has established for normal operations.

Response to Comment 8.1: For the reasons stated below, the EPA disagrees that there should be a separate NOX standard during startup and shutdown.
   The EPA does not claim that the Sierra Club case constrains its authority to prescribe unique standards for SSM periods or directly addresses the legality of SSM exemptions for section 111 standards. The EPA's view is that the reasoning of the Sierra Club decision calls into question the legality of source category-specific SSM exemptions in rules promulgated pursuant to section 111. The EPA`s approach to malfunctions is consistent with section 111 and is a reasonable interpretation of the statute.
    
   The EPA disagrees with commenter's suggestion that the existence of an SSM exemption in rules implementing section 111 in 1977 when Congress enacted the "continuous basis" language in the definition of "emission standard" is evidence that Congress approved of that regulatory SSM exemption. The commenter fails to cite legislative history or any other evidence supporting that Congress was aware or approved of that exemption. 
   
The EPA also disagrees with the commenter that contends that the Sierra Club decision addresses in any way the EPA's current approach to malfunctions. Consistent with Sierra Club, we have issued standards that apply at all times, including during malfunctions.
8b. The standard is not achievable
Comment 8.2: Commenter 0091 notes that the EPA ignores the undisputed existence of malfunctions even at best-performing sources, and claims falsely that the best-performing sources "achieve" emission levels that they undisputedly do not achieve part of the time. Since the EPA describes malfunctions as being sometimes unavoidable or "not reasonably preventable," despite proper design and maintenance of equipment, the commenter believes there is no basis for the EPA's conclusion that malfunction events are not representative of best-performing sources. According to the commenter, the EPA must acknowledge the fact that those sources nevertheless experience malfunction events, rather than assume those emissions away.
Commenters (0091, 0095, 0098) state that the 30-day rolling compliance period does not properly account for the increased emissions at startup and shutdown. 
Commenter 0098 notes the EPA's use of Test Plants that can avoid the increased emissions at startup and shutdown due to their atypical design, process, operation, and emissions controls to set a 30-day compliance period thoroughly undercuts the EPA's reasoning when considered in light of the entire industry that the vast majority of NAPs cannot avoid such increased emissions. 
Commenter 0098 believes that the 30-day rolling compliance period does not properly account for the increased emissions at startup and shutdown. Specifically, the commenter believes that multiple startups and shutdowns within a 30-day operating period can drastically skew the data for any given period. As such, and consistent with EPA's BACT determinations, the commenter urges the EPA to consider a 365-day rolling emission rate rather than the 30-day approach. According to the commenter, this allows the facilities to more appropriately represent emissions over a longer time period, thereby accounting for SSM events as well as balancing seasonal impacts to emissions (i.e., higher summer emissions and lower winter emissions).
Commenter 0091 states that using long-term averaging periods for emission limitations is one way of addressing temporarily higher emission rates that even plants with the best demonstrated control technology may experience during startups and shutdowns, provided the EPA establishes the long-term emission rate using data that adequately represent emissions during startup and shutdown events. The commenter emphasizes that under CAA section 111, EPA must determine that emission limitations that are based on steady-state operation but will apply during startups and shutdowns are "achievable" using emission reduction technology that the Administrator determines is the "best" and has been "adequately demonstrated." According to the commenter, there is no indication in the preamble to the Proposed Rule that EPA has conducted a sufficient analysis to demonstrate the achievability of the proposed emission limit during startup and shutdown events. The commenter is skeptical that EPA has done so, because adequate emissions data for such an analysis often are not available for periods of startup and shutdown.
Commenter 0098 states that the EPA's estimate of SSM events at three to four hours per month grossly underestimates such events. The commenter suggests that the EPA seek actual SSM event information from the industry in order to properly account for such events when setting the proposed NSPS. According to the commenter, SSM events are highly variable at NAPs, as some are scheduled and, of course, others are not. For one of the commenter's organizational members, such events are scheduled for gauze changes (approximately every 140 days) during which the NAP can be down for nine hours. Moreover, the commenter believes that it is not unusual to have one or more additional SSM events between gauze changes. Another member typically experiences eight hour startups (with numerous startups per month), and has power loss (due to the grid) of three to four hours. Additionally, according to the commenter, older NAPs typically change out catalysts every 60 to 90 days, which can lead to additional SSM events each month. Thus, in the commenter's experience, NAPs typically experience many more hours of SSM events during a 30-day period than the three to four hour average estimated by the EPA, which, in turn, bears directly on the achievability of the proposed NSPS, which is required continuously. 
Commenter 0092 states that even with selective catalytic reduction (SCR) emissions control equipment similar to dual pressure plants, mono-pressure nitric acid plants would have more difficulty meeting the proposed 30-day rolling emission rate because of start-up, shutdown and more frequent maintenance issues. According to the commenter, mono-pressure plants typically start up and shut down more frequently because the precious metal catalyst life cycle is shorter, typically about three months while dual pressure plants typically change catalyst every six to nine months. Therefore, according to the commenter, the proposed rolling emission rate of 30 days is too short. According to the commenter, more frequent thermal cycling of high temperature equipment also shortens the expected operating life of mono-pressure plants and increases the frequency of re-starts for maintenance issues. If both mono and dual pressure plants will be regulated under a single new NSPS standard, the commenter asks that a longer rolling average time period be used.
Commenter 0095 states that smaller plants that produce Nitric Acid for internal demand driven consumption incur very frequent startups and shutdowns and, therefore, have higher average NOX emissions. The assumption that the 30-day rolling emission rate that includes periods of startup and shutdown will compensate for the excess hourly emissions that occur during these periods with typically low emissions during steady state operations may be true for large commercial nitric acid producing plants, but according to the commenter, not the smaller ones. The commenter's review of CEMS data that was submitted to the EPA under the 114 request indicates these emissions trends.
Commenter 0091 states there are several reasons why adequate data during SSM events often do not exist. According to the commenter, the applicable regulations generally prohibit testing during SSM conditions, and require that data not be used for compliance purposes if obtained during a startup, shutdown, or malfunction event. To the extent the EPA evaluates emissions data collected through continuous monitoring, the commenter notes that the applicable regulations often require or allow the source to exclude from its reporting of continuous monitoring data those data reflecting SSM conditions. Also, according to the commenter, atypical pollutant concentrations and other stack conditions that may exist during startup and shutdown can result in the continuous monitoring system producing unusable data because the pollutant concentration may be outside of the monitoring equipment's span or the stack conditions may not meet monitoring system QA/QC parameters, or the data may be truncated on the high end because of limitations of the monitoring equipment. According to the commenter, these factors would cause the EPA to calculate an emission limit that understates emissions occurring during startup and shutdown. 
Additional concerns of commenter 0091 are: 
   oo If unusual stack conditions during a startup or shutdown event are outside the QA/QC parameters for a continuous monitor that is installed and calibrated to monitor normal operations, the commenter is concerned that the source may be in violation of the NSPS because of monitor unavailability; and
   oo The commenter suggests that the EPA clearly demarcate the conditions that should be characterized as either malfunctions or as startup or shutdown events.
   oo The commenter states that the EPA could better ensure that the standards applicable during startup and shutdown comply with the criteria for NSPS in CAA section 111. 
The commenter recommends conducting stack testing during startup and shutdown events for certain types of sources and stack configurations, to better delineate the variations in emissions. (According to the commenter, stack sampling procedures may not apply or may not produce accurate results during the abnormal flow and concentration conditions present during startup or shutdown.)
The commenter recommends analyzing data from plants with continuous monitoring for startup and shutdown events at more facilities.
The commenter recommends deferring promulgation of emission limitations that apply during periods of startup or shutdown until after EPA obtains such data. 
The commenter recommends studying facilities' operational logs to be sure it has chosen an averaging time for an emission standard sufficiently long that typical startups and shutdowns would not cause an exceedance of the standard.
The commenter recommends determining, based on the infeasibility of collecting representative data during startup and shutdown events, that it is necessary and appropriate to promulgate a design, equipment, work practice, or operational standard, pursuant to CAA section 111(h) and/or section 302(k), in lieu of deriving limitations on the mass or concentration of pollutants emitted during startup or shutdown.
    
 While the EPA calculated the proposed emission limits for nitric acid plants using continuous monitoring data that the Agency asserts included startup and shutdown events, it is not apparent to commenter 0091 whether or how that analysis adjusted for the limitations in monitoring data for periods of startup and shutdown described above, or that the data are representative of the magnitude of emissions during startup and shutdown. The commenter suggests that the EPA would need to analyze monitoring data for an extended period of time and for numerous facilities to try to establish emission limitations that could be met during startup and shutdown (with the level of control technology provided in CAA section 111). Expressing an emission limitation as a long-term average, while helping to address higher emissions during startup and shutdown, the commenter argues that the average could fail to account adequately for the potential wide variation in emissions during startup and shutdown events, which could be masked if the EPA simply assumes that calculating a long-term-average emission rate from existing monitoring data will adequately address startups and shutdowns.
Commenter 0091 states that a particular problem with respect to characterizing emissions during startup and shutdown events for Nitric Acid plants is that the current NSPS for Nitric Acid plants only requires a continuous NOX monitor, which provides NOX emissions in parts per million by volume (ppmv), but does not provide mass emissions (i.e., lb/hr), and yet the NSPS NOX standard is in terms of mass emissions (lb/ton of nitric acid produced). Under the existing (and proposed) NSPS, the commenter notes that a conversion factor is established during the initial performance tests to allow conversion of the ppmv values into lb/hr NOX emissions. According to the commenter, there does not appear to be any evidence, however, that these factors, which the EPA presumably used to set the proposed revised NSPS emission limit and are calculated based on normal steady-state operation, are applicable or valid for startup/shutdown events. (According to the commenter, most existing nitric acid plants do not have stack gas flow rate monitors, since the current NSPS does not require one, and therefore, there is no actual data on lb/hr emissions during startup/shutdown events. Therefore, according to the commenter, the EPA likely has no valid lb/ton emission rate data during startup/shutdown events.)
Response to Comment 8.2: Commenters have argued that the emission limit is not achievable during malfunction events because testing is prohibited during SSM conditions. First of all, the commenter is referring to the RATA test, where the emission factor is determined for use with the NOX concentration CEMS to calculate emissions in units of the standard. This prohibition is not in place under normal operating procedures.
   Also commenters argued that data are not used for compliance purposes if obtained during a startup, shutdown, or malfunction. However, we note that the general provisions in Part 60 require facilities to keep records of the occurrence and duration of any startup, shutdown or malfunction (40 CFR 60.7(b)) and either report to the EPA any period of excess emissions that occurs during periods of startup, shutdown or malfunction (40 CFR 60.7(c)(2)) or report that no excess emissions occurred (40 CFR 60.7(c)(4)). If the CEMS data do not represent these periods of startup, shutdown, and malfunction, it is unclear to the agency how nitric acid plants are complying with these requirements. Also see Section V of the final rule preamble for further discussion of the final standard.
   
Commenters noted that mono-pressure nitric acid plants would have more difficulty meeting the proposed 30-day rolling emission rate because of start-up, shutdown, and more frequent maintenance issues. However, the EPA has obtained data from monopressure plants that showing compliance with the final standard of 0.50 lb NOX/ton acid has been demonstrated.
The requirements of Subpart G are such that the equipment used to measure the NOX concentration has a span value of 500 parts per million (ppm). Commenters have indicated that the NOX concentrations during startup and shutdown can be higher than the monitor can measure. However, the one of the facilities upon which the standard was based uses a dual-span NOX monitor allowing us to determine the magnitude of emissions during periods when concentrations exceeded 500 ppm.
Additional data received included nitric acid plants that are not currently subject to subpart G and are generally not employing the best system of emissions reductions. As expected, these facilities may need to install additional controls to comply with subpart Ga if they are modified or reconstructed and trigger the standard. We have analyzed the cost of such controls and find them appropriate and cost effective. Also see the memo entitled Analyses of Data Received Between Proposal and Promulgation of Part 60, Subpart Ga in Docket ID No. EPA-HQ-OAR-2010-0750.
   Commenters also noted that the vast majority of NAPs cannot avoid such increased emissions during startup and shutdown. The EPA notes that this standard does not apply to existing sources unless they are modified or reconstructed. This final emission limit of 0.50 lb NOX/ton acid will only be applicable to new sources (which includes reconstructed or modified sources). We note that periods of startup and shutdown make up a very small fraction of total operating time. See Agrium North Bend Analyses for further analysis of startup and shutdown periods. We have shown that SCR is the best system of emission reductions currently in use because it achieves very low emissions, is cost effective, and produces minimal secondary environmental impacts. 
   
    NSPS does not prescribe a particular technological system that must be used to comply. Sources remain free to elect whatever combination of measures will achieve equivalent or greater control of emissions.
   
8c. Issues with SSM analysis
Comment 8.3: Commenter 0098 suggests that as the EPA reevaluates its analysis regarding the impact of SSM on the achievability of a proposed NSPS that the EPA should provide technical definitions of the terms startup and shutdown that are specific to NAPs so that NAPs can determine when they are considered to be operating for purposes of compliance. Some of the commenter's facilities are subject to differing definitions of these terms in consent decrees, create substantial disparity between how these terms are defined, which can have serious impacts as to how emissions are measured and evaluated.
For example, the commenter notes that in its analyses of the Test Plants, the EPA assumed that ammonia feed represented operating hours of the NAP, and that all other hours were not to be included in the 30-day rolling emission rate. Thus, the commenter notes that its analysis of the PCS Geismar 5 data only included the hours where ammonia feed was greater than zero. According to the commenter, some NAPs define operation as when the NOX compressor is operational, while others consider the plant to be operating only when ammonia is flowing to the AOR. Also, the commenter notes the Consent Decree with Terra, where the EPA agreed that "startup begins one hour prior to the initiation of the feed of ammonia to the Covered Nitric Acid Plant and ends no more than five hours after such initiation of the feed of ammonia." Additionally, the commenter notes that the Consent Decree states that "shutdown begins at the time the ammonia feed ceases and ends three hours later." According to the commenter, these definitions are not consistent with the method the EPA used to calculate emissions at PCS Geismar 5, which is noted above.
Because there are differences in the industry regarding when startup and shutdown periods begin and end, and thus, when a NAP is considered operating, commenter 0098 recommends that the EPA define startup and shutdown such that NAPs are considered to be operational only when ammonia flow into the AOR is greater than zero. Thus, according to the commenter, startup would be defined as when ammonia is first introduced into the AOR, and shutdown as when ammonia flow to the AOR ceases. After defining startup and shutdown, the commenter requests that the EPA reevaluate the impact of such events on NAPs' emissions, and their ability to achieve the proposed NSPS.
Commenter 0098 requests that the EPA revise the definition of an "operating day" for purposes of Subpart Ga. Frequently, according to the commenter, NAPs will begin startup or shutdown operations in the evening meaning that one or perhaps two days of the 30 consecutive operating day period will be devoted to startups or shutdown emissions and these emissions may only last for a few hours. The result, according to the commenter, is that the emission rate, calculated as E30-day (§60.75a, Eq. 2), will be skewed with higher emissions relative to nitric acid production. In order to obtain more consistent data reporting periods, the commenter asks that a minimum number of hours should be established to constitute an "operating day" to eliminate operating days consisting only of startup, shutdown or malfunction events. As an alternative approach, if the EPA continues to believe a 30-day compliance period is appropriate, the commenter recommends that the EPA consider a 720 hour compliance period, as opposed to a 30-day period.
Response to Comment 8.3: The EPA disagrees with the commenter's statement that the EPA's analysis of PCS Geismar Train 5 was incorrect due to using data only where ammonia feed flow rate was greater than zero pounds per hour (lb/hr). The data received from PCS Geismar stated that the train is operational when the ammonia feed flow rate is greater than 1 lb/hr. The EPA's analysis contained a slight error because ammonia feed flow rates greater than 0 lb/hr were used to indicate whether the train was in operation. However, excluding emissions data that occurred while ammonia feed flow rates were between 0 and 1 lb/hr, the data analysis did not change significantly and the resulting conclusions did not change. Also, the EPA disagrees that a "NAP can be down for nine hours" is the same as the NAP undergoing nine hours of startup, shutdown, malfunction (SSM) events. When the NAP is down (in cold shutdown and not producing nitric acid), the process is not operating and there are no emissions. When the NAP is undergoing SSM events, there are emissions.
However, the EPA agrees that a correction should be made to the way that the NOX emissions are calculated over the 30 day compliance period. The final rule calculation will be based on how many hours the unit operated over each 30 day period, up to 720 hours. Each hour is weighted equally whether this hour is a normal operating hour or an hour where startup, shutdown, or malfunction occurred during a 30 day period. 
Also, the EPA agrees that some facilities may undergo more (or fewer) periods of startup and shutdown during a typical 30 day period compared to the facilities for which we had emissions data. Nonetheless, we still believe these periods make up a very small fraction of total operating time. Further, the EPA has performed hypothetical calculations in which the number of SSM periods during a month was doubled, and the standard was still achieved. Additionally, the updated method for calculating the 30 day rolling emission rate should prevent these infrequent and short duration periods from having artificially inflated impacts on the 30 day rolling emission rate values. See the memo Agrium North Bend Analyses, and the memo Statistical Evaluation of CEMS Data to Determine the NOX Emission Standard (Updated Memo for Final Standard). Also see Section V of the final rule preamble for more discussion of the final standard.
Also, the EPA agrees with the commenter's statement about monitor unavailability. In order to lessen the likelihood that a monitor installed and calibrated to monitor normal operations will not be able to be used to monitor emissions during SSM events, the EPA has added a requirement to use dual span monitors. This ensures that the monitor has been calibrated for the ranges of emissions expected during normal operations and during SSM events.
The EPA disagrees with the commenters that definitions for startup and shutdown are needed. Due to the fact that the emission standard for nitric acid plants must be met at all times, including startup and shutdown, there is no need to define startup and shutdown specifically for nitric acid plants. Therefore, the definitions contained in the General Provisions for Part 60 are appropriate. 
8d. Numerical emission limit should apply during SSM
Comment 8.4:   Commenter 0104 states that the EPA's proposed 30-day averaging period may be adequate if periods of startup and shutdown are included and malfunctions are not excused. As a general principle, the commenter states that averaging times should not be stretched simply to accommodate whatever emissions may result from these events. Rather, the commenter urges the EPA to evaluate their causes and determine whether specific requirements should be included within the NSPS performance standards for each industry to reduce the frequency and severity of such episodes. For example, when determining the NSPS performance standard, the commenter recommends that the EPA consider the impact of downtime and accident history on emission levels from the different pollution control options it is evaluating before making its selection.
The commenter recognizes that starting up and shutting down units for repair in response to market conditions is an inevitable feature of plant operation. So long as the EPA's 30-day rolling emission rate is based on actual hours of operation, the commenter notes that it may be reasonable in light of the limited information available in the docket. 
It is difficult to respond to the EPA's invitation to comment on whether the compliance period should be shorter, as the commenter notes that the information in the docket does not always distinguish between emissions resulting from malfunctions and those that are driven by startup and shutdown. Regardless of that distinction, the commenter believes that the data does indicate that, at the very least, the 30-day compliance period should be more than adequate to accommodate startups and shutdowns, and that a shorter compliance period may not have a significant impact on emissions. 
According to the commenter, the EPA has recognized that there is no basis in law for excusing malfunctions, either directly or by manipulating averaging times in anticipation that pollution control equipment will fail. Such a loophole would, in the opinion of the commenter, reward those companies that choose not to invest in maintenance or otherwise operate their facilities in a negligent manner. According to the commenter, the CEMS data available indicate that malfunctions are relatively infrequent and should have no impact on compliance with the 30-day limit that EPA has proposed. In addition, the commenter notes that EPA is basing its standard on emission levels that plants built long ago are already achieving 99% of the time; newer units should be expected to achieve higher performance levels.
Response to Comment 8.4: The EPA agrees with the commenter that the NOX emission standard should apply during startup, shutdown, and malfunction. 
There is nothing in CAA section 111 that directs the Agency to consider malfunctions in determining the level "demonstrated" by the application of the "best system of emission reduction" when setting emission standards. Moreover, while the EPA accounts for variability in setting emissions standards consistent with the CAA section 111 case law, nothing in that case law requires the Agency to consider malfunctions as part of that analysis. Clean Air Act section 111 uses the concept of "best system of emission reduction" in defining the level of stringency that CAA section 111 performance standards must meet. Applying the concept of "best system of emission reduction" to a unit that is malfunctioning presents significant difficulties, as malfunctions are sudden and unexpected events.
As explained above, the EPA has determined that CAA section 111 does not require that emissions that occur during periods of malfunction be factored into development of CAA section 111 standards.
8e. Impact of DC Circuit Court's Sierra Club decision
Comment 8.5:   Commenters (0091, 0098) state that the EPA failed to properly consider the impact of SSM events when setting the proposed NSPS.
Commenters (0091, 0098) state that the EPA conducted no analysis of what impact malfunctions have on the achievability of its proposed NSPS, other than to state that malfunction events were covered by the data it analyzed with respect to the Test Plants. The commenters note that the reasoning is in error for the same reason the EPA's reasoning regarding startup and shutdown events is in error. 
The commenters state that the EPA did not address what impact the D.C. Circuit Court's Sierra Club decision has under CAA §111 in this rulemaking. Instead, without comment or explanation, the commenters state that the EPA proceeded to draft a proposed NSPS that requires continuous compliance, including during SSM events, a departure from over 40 years of contrary requirements. Such an action, alone, according to the commenters, is arbitrary and capricious. The commenters assert that the EPA made no attempt to analyze what impact SSM events have on the achievability of the proposed NSPS. Commenter 0098 objects to the proposed accounting of startup and shutdown periods by using a 30-day compliance period. The commenters state that the EPA provides no analysis or data to support this claim, but instead apparently believes that a 30-day compliance period is sufficient based on the emissions data from the Test Plants. The commenter states that the 30-day compliance period is too short.
Response to Comment 8.5:  The EPA does not claim that the Sierra Club case constrains its authority to prescribe unique standards for SSM periods or directly addresses the legality of SSM exemptions for section 111 standards. The EPA's view is that the reasoning of the Sierra Club decision calls into question the legality of source category-specific SSM exemptions in rules promulgated pursuant to section 111. The EPA also disagrees with the commenter that contends that the Sierra Club decision addresses in any way the EPA`s current approach to malfunctions.
Periods of startup, normal operations, and shutdown are all predictable and routine aspects of a source's operations. However, by contrast, malfunction is defined as a "sudden, infrequent, and not reasonably preventable failure of air pollution control equipment, process equipment, or a process to operate in a normal or usual manner * * *"(40 CFR 60.2). 
Nothing in CAA section 111 directs the Agency to consider malfunctions in determining the level "demonstrated" by the application of the "best system of emission reduction" when setting emission standards. Moreover, while the EPA accounts for variability in setting emissions standards consistent with the CAA section 111 case law, nothing in that case law requires the Agency to consider malfunctions as part of that analysis. Clean Air Act section 111 uses the concept of "best system of emission reduction" in defining the level of stringency that CAA section 111 performance standards must meet. Applying the concept of "best system of emission reduction" to a unit that is malfunctioning presents significant difficulties, as malfunctions are sudden and unexpected events.
As explained above, the EPA has determined that CAA section 111 does not require that emissions that occur during periods of malfunction be factored into development of CAA section 111 standards. Also refer to Section V of the final rule preamble.
8f. Safety issues during SSM
Comment 8.6:   Commenter (0098) states that the EPA failed to consider safety issues associated with its proposal, as required by CAA §111(a)(1). According to the commenter, the EPA must account for these safety concerns when it attempts to bring startups and shutdowns within its NOX standard.
According to the commenter, there are two major safety issues surrounding startup and shutdown procedures that require very low levels of NOX emissions. The commenter notes the first issue is the creation and accumulation of ammonia nitrite in the absorption tower. If a NAP is started slowly and the primary gauze is not heated to its appropriate levels immediately, the commenter notes that there is a risk of ammonia "slip" from the AOR down-stream into the process stream. Under conditions present in the stack, the commenter states that the ammonia and NOX that have slipped into the flue gas can react, forming ammonium nitrite, an unstable ammonia salt which is explosive at temperatures greater than 60-70°C (140-158°F). Depending on conditions, the commenter notes that the ammonia nitrite can also be formed and deposited, which also creates a risk of explosion.
According to the commenter, the second safety concern involves the accumulation of ammonia within the SCR unit itself. In order to ensure that all NOX has been destroyed, the commenter notes that the SCR unit can be run without the plant being operational. Without appropriate levels of NOX being supplied to the SCR by means of the AOR and/or the compressor, the commenter states that ammonia will accumulate within the SCR unit, which can lead to the disintegration of the SCR catalyst and can create an explosive environment within the SCR shell, which is especially dangerous considering the high levels of heat involved in the process.
Response to Comment 8.6: We recognize that safety hazards can occur if certain conditions are present in the control device. However, there are methods to prevent these conditions, such as hydrogen peroxide injection (HPI) or startup heaters that are in use in the industry. For small plants, startup heaters can be purchased and installed for approximately $400,000.
If the above methods cannot be used, the facility needs to exercise caution when starting up control equipment to ensure that these conditions do not occur. However, the emission standard is based on a sufficiently long compliance period that if higher emissions occur during startup and shutdown because control devices cannot be operated in the same manner in which they're used during steady state operations, the impact of those emissions should be sufficiently small that they will not create significant compliance problems.
8g. Affirmative defense is unnecessary
Comment 8.7:   Commenter 0095 states that the affirmative defense is unnecessary and commenters 0098 and 0104 ask that the EPA eliminate the affirmative defense.
Commenter 0098 notes that nothing in CAA §111 expressly commands EPA to account for malfunctions and the EPA cannot ignore such events when setting an NSPS. According to the commenter, it is the EPA's duty to ensure that an NSPS is set at an achievable level, and that its analysis necessarily includes considering events that impact emissions at NAPs. The commenter asserts that the EPA has failed to meet that duty when setting the proposed NSPS. Moreover, the commenter notes that EPA's proposal to soften its failure to account for malfunctions by suggesting it will exercise appropriate enforcement discretion or by creating an affirmative defense does not suffice. According to the commenter, substituting enforcement (whether based on an affirmative defense or the exercise of enforcement discretion) for actual emission limitations was long-ago rejected by the D.C. Circuit and is not an option. [The commenter references Portland Cement, 486 F.2d at 399; and also Nat'l Lime, 627 F.2d at 431 n.46 (enforcement does not render an NSPS standard achievable), 434 n. 54 (suggesting that in evaluating achievability and representativeness, the CAA 1977 amendments' requirement that sources meet emissions limitation continuously may have repealed any flexibility EPA had to exclude consideration of SSM events when setting NSPS).]
Commenter 0104 states that the EPA's proposed "affirmative defense" to civil penalties based on malfunction events is unlawful, would likely reduce compliance with the proposed standards, and would have harmful consequences for public health and the environment.
According to commenter 0104, the statute makes clear that only courts and judges are authorized to assess civil penalties, whether a case is brought by the EPA or a citizen. Moreover, in the opinion of the commenter, Congress expressed a clear intent as to how judges should determine the size of civil penalties whenever they are sought, and thus Congress flatly barred the EPA from limiting when civil penalties can be assessed. Indeed, according to the commenter, the Clean Air Act itself spells out the only limits that Congress intended to impose on citizens' ability to seek and recover penalties in enforcement suits. By attempting to impose additional agency-created limits, the commenter believes that the EPA exceeds its authority.
Commenter 0104 notes as an example--in an enforcement action against a nitric acid plant for a violation of the NSPS performance standards, the owner might argue that it is exempt from paying civil penalties so long as the owner satisfied the requirements set forth in the EPA's proposed affirmative defense regulations. According to the commenter, the owner of a facility must not be able to evade the civil penalties that apply when the congressionally mandated factors in the statute are met. The commenter states:
         See 42 U.S.C. § 7413(e) (listing factors). Note that the proposed exemption would also be barred under Chevron step two or found to be arbitrary and capricious since, even if there exists some slight ambiguity, it is unreasonable to construe the statute as permitting EPA to short-circuit the consideration of specifically listed factors. See Chevron, 467 U.S. at 843 (explaining that, if the statute does not answer the question at issue, "the question for the court is whether the agency's answer is based on a permissible construction of the statute"); S. Coast Air Quality Mgmt. Dist. v.EPA, 472 F.3d 882, 895 (D.C. Cir. 2006) ("We further hold that EPA's interpretation of the Act in a manner to maximize its own discretion is unreasonable because the clear intent of Congress in enacting the 1990 Amendments was to the contrary."); see also Gen. Instrument Corp. v. F.C.C., 213 F.3d 724, 732 (D.C. Cir. 2000) (explaining that "an arbitrary and capricious claim and a Chevron step two argument overlap"); Motor Vehicle Mfrs. Ass'n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (explaining that an agency acts in arbitrary and capricious manner if it fails to consider "relevant factors" or "entirely fail[s] to consider an important aspect of the problem"). By "upset[ting] the statutory balance struck by Congress," as discussed above, the affirmative defense is unreasonable under Chevron step two. Int'l Alliance of Theatrical & Stage Employees v. N.L.R.B, 334 F.3d 27, 35 (D.C. Cir. 2003).
The commenter continues by saying it is improper for a court to fail to consider these factors, or to fail to make its own determination of what civil penalties are "appropriate" under Section 304(a), and the EPA should not ask a court to ignore its legal duty. According to the commenter, it is simply impermissible for the EPA to displace the civil penalty factors of the Clean Air Act or alter their significance by creating a bar to penalties if certain agency-defined considerations are met instead. The commenter states:
         In another proceeding, EPA has argued that it may promulgate an affirmative defense as an interpretation of the `catchall' provision of Section 113(e)(1), which directs "the Administrator or the court, as appropriate," to consider the enumerated factors and "such other factors as justice may require." Partial Approval of Texas SIP, 75 FR 68989, 68999 (Nov. 10, 2010). As noted above, however, with respect to civil penalties, this provision represents a Congressional delegation of discretion to judges, not EPA. Moreover, even if EPA had authority to interpret "other factors," EPA's proposed affirmative defense would require a court to elevate these additional factors above and in contradiction to those enumerated by the statute, and this is not a reasonable or permissible interpretation.
According to Commenter 0104, the Clean Air Act grants EPA minimal discretion that only applies to administrative penalties, allowing the EPA to "compromise, modify, or remit, with or without conditions, any administrative penalty which may be imposed under [subsection 113(d)]." According to the commenter, there is no similar grant of authority, however, for the EPA to compromise, modify, or limit civil penalties that a court may impose under Section 113(e) or Section 304. According to the commenter, Section 304(a) grants courts the sole authority "to apply any appropriate civil penalties" in citizen suits. The explicit reference to the EPA's ability to modify penalties in one subsection and its absence in the other subsection of the same provision, according to the commenter, can only be understood as an intentional delineation of EPA authority by Congress that EPA may not contravene by rule.
Commenter 0104 urges the EPA not to adopt in its final rule any affirmative defense because a regulatory limitation on civil penalties intrudes upon the authority Congress granted solely to courts to impose civil penalties for violations of the Clean Air Act.
 Commenter 0104 states that Congress plainly intended for citizens to be able to enforce emission standards under the Clean Air Act using the full range of civil enforcement mechanisms, subject only to the limitation that government not be "diligently prosecuting" its own civil enforcement action. Under the proposed rule, according to the commenter, citizen participation in Clean Air Act enforcement will be hindered, in violation of citizens' rights to protect themselves from pollution and in direct conflict with congressional intent. The affirmative defense proposed here by the EPA would, according to the commenter, likely be used on a routine basis by polluters seeking to avoid penalties, just as the malfunction exemption was. As a result, the commenter believes that citizens who seek the assessment of civil penalties against polluters in order to protect themselves and achieve the Act's goals may be forced to engage in fact-intensive disputes over the cause of emission violations and the adequacy of responsive measures -- an outcome Congress intended to prevent with the simple straightforward enforcement and penalty provisions in the Clean Air Act. As a result, the commenter states that compliance with the Act could suffer, for civil penalties provide a powerful deterrent to violators as Congress intended. According to the commenter, the Supreme Court has explained: "To the extent that [civil penalties] encourage defendants to discontinue current violations and deter them from committing future ones, they afford redress to citizen plaintiffs who are injured or threatened with injury as a consequence of ongoing unlawful conduct."
Commenter 0104 continues: Thus, the affirmative defense runs counter to the clearly expressed intentions of Congress because the burden it places on citizens makes it less likely that they will enforce the Act, and several of the factors at issue in the affirmative defense undercut Congress's intent that citizen suit enforcement should avoid re-delving into "technological or other considerations." According to the commenter, both result from the technical burden the EPA is proposing to place on citizens with the affirmative defense, and both render the affirmative defense impermissible.
Commenter 0104 states that Congressional intent on civil penalties is clear -- they are a remedy available for enforcement by citizen plaintiffs, and the Act gives judges a list of factors to consider in assessing them. The commenter asserts that the EPA cannot write regulations contravening that intent. By shifting this careful balance and contravening these mandates, the EPA's proposed affirmative defense for nitric acid plants will, according to the commenter,  impermissibly chill citizen participation and hinder their ability to win an effective, deterrent remedy in Clean Air Act enforcement actions.
According to the commenter, much like citizens, neither the EPA nor state agencies have the resources to investigate, much less contest, affirmative defenses. The commenter asserts that promulgating this affirmative defense is equivalent to giving polluters "get out of jail free" cards for serious violations. According to the commenters, polluters are likely to claim that any violation of the standard is due to a malfunction in order to evade the applicable requirements and the consequences of violating them. the commenter claims that allowing polluters to evade financial penalties -- which are the real teeth of the standards -- through this type of measure is likely to lead to polluters simply ignoring or factoring potential standard violations into their cost of doing business, rather than actually trying to prevent malfunctions and violations of the standards as a way to avoid financial losses from the application of penalties.
Commenter 0099 disagrees with the EPA's proposal to "add an affirmative defense to civil penalties for exceedances of emission limits that are caused by malfunctions," and urges the EPA to withdraw its proposed creation of an affirmative defense in its entirety.
The commenter states that creation of such a new loophole is illegal and unwarranted, and the proposal does not support the agency's duty to protect public health. The commenter notes that the Clean Air Act specifies how the courts are to assess civil penalties, whether a case is brought by the EPA or a citizen. According to the commenter, Congress plainly intended citizens to be able to enforce emission standards under the CAA using the full range of civil enforcement mechanisms available to the government. The commenter states that the EPA's proposal, by purporting to shift this balance and contravening these mandates, violates the statute and impairs the ability of affected communities to enforce their right to protection from harmful pollution.
Moreover, commenter 0099 believes that the EPA's proposed standard already accounts for periods of malfunction: the EPA believes "that sources will be able to operate their plants in compliance with the standard even if they experience malfunctions." The commenter believes the proposed standard is insufficiently stringent even for NOX, and it makes no sense to add an additional loophole on top of a standard that already accounts for the purported inevitability of malfunction at nitric acid plants.
 Commenter 0095 states that the affirmative defense provisions are not necessary for a NSPS rule since the Part 60 Rules were not impacted by the D. C. Circuit's decision in Sierra Club vs. EPA, which vacated the exemption for excess emissions during SSM events contained in 40 CFR Part 63, subpart A. In fact, according to the commenter, the NSPS rule structure does not require sources to meet NSPS emission limitations during SSM events in 40 CFR Section 60.8(c). The commenter quotes:
      "Operations during periods of startup, shutdown, and malfunction shall not constitute representative conditions for the purpose of a performance test nor shall emissions in excess of the level of the applicable emission limit during periods of startup, shutdown, and malfunction be considered a violation of the applicable emission limit unless otherwise specified in the applicable standard."
Response to Comment 8.7: The EPA included an affirmative defense in the final rule in an attempt to balance a tension, inherent in many types of air regulation, to ensure adequate compliance while simultaneously recognizing that despite the most diligent of efforts, emission standards may be violated under circumstances beyond the control of the source. The EPA must establish emission standards that "limit the quantity, rate, or concentration of emissions of air pollutants on a continuous basis." 42 U.S.C. § 7602(k)(defining "emission limitation and emission standard"). See generally Sierra Club v. EPA, 551 F.3d 1019, 1021 (D.C. Cir. 2008). Thus, the EPA is required to ensure that section 111 emissions standards are continuous. The affirmative defense for malfunction events meets this requirement by ensuring that even where there is a malfunction, the emission limitation is still enforceable through injunctive relief. 
While "continuous" limitations, on the one hand, are required, there is also caselaw indicating that in some situations it is appropriate for the EPA to account for the practical realities of technology. For example, in Essex Chemical v. Ruckelshaus, 486 F.2d 427, 433 (D.C. Cir. 1973), the D.C. Circuit acknowledged that in setting standards under CAA section 111 "variant provisions" such as provisions allowing for upsets during startup, shutdown and equipment malfunction "appear necessary to preserve the reasonableness of the standards as a whole and that the record does not support the `never to be exceeded' standard currently in force." See also, Portland Cement Association v. Ruckelshaus, 486 F.2d 375 (D.C.Cir. 1973). Though intervening caselaw such as Sierra Club v. EPA and the CAA 1977 amendments calls into question the relevance of these cases today, they support the EPA's view that a system that incorporates some level of flexibility is reasonable. The affirmative defense simply provides for a defense to civil penalties for excess emissions that are proven to be beyond the control of the source. By incorporating an affirmative defense, the EPA has formalized its approach to upset events. In a Clean Water Act setting, the Ninth Circuit required this type of formalized approach when regulating "upsets beyond the control of the permit holder." Marathon Oil Co. v. EPA, 564 F.2d 1253, 1272-73 (9th Cir. 1977). But see, Weyerhaeuser Co. v. Costle, 590 F.2d 1011, 1057-58 (D.C. Cir. 1978) (holding that an informal approach is adequate). The affirmative defense provisions give the EPA the flexibility to both ensure that its emission limitations are "continuous" as required by 42 U.S.C. § 7602(k), and account for unplanned upsets and thus support the reasonableness of the standard as a whole.
Further, the EPA's view is that the affirmative defense is consistent with CAA sections 113(e) and 304. Section 304 gives district courts jurisdiction "to apply appropriate civil penalties." Section 113(e)(1) identifies the factors that the Administrator or a court shall take into consideration in determining the amount of a penalty to be assessed only after  it has been determined that a penalty is appropriate. The affirmative defense regulatory provision is not relevant to the amount of any penalty to be assessed under section 113(e) because if a court determines that the affirmative defense elements have been established, then a penalty is not appropriate and penalty assessment pursuant to the section 113(e)(1) factors does not occur.
   Additionally, the affirmative defense is reasonable. The EPA's judgment is that the affirmative defense criteria capture the appropriate considerations in determining whether penalties are appropriate when a violation occurs as the result of a malfunction. As noted above, the affirmative defense criteria overlap to some extent with the penalty assessment criteria set forth in section 113(e), but are not identical. For example, size of business is one of the factors listed in section 113(e), but is not reflected in the EPA's affirmative defense. This reflects the EPA's view that when a violation is caused by a malfunction, the size of the business is not relevant to whether penalties should be excused. If the violation was unavoidable and could not have been prevented, the EPA's view is that it would be unfair to impose a penalty no matter the size of the business. 
   
In further response to the commenter that stated the affirmative defense provisions are not necessary for NSPS standards, the reference to 40 CFR Section 60.8(c) affirms the EPA's position on SSM for Section 111 sources. The paragraph can be interpreted to allow "emissions in excess of the level of the applicable emission limit during periods of startup, shutdown, and malfunction [to] be considered a violation of the applicable emission limit" if "specified in the applicable standard." 
It follows that the language "unless otherwise specified in the applicable standard" means that the EPA has the right to override the provisions of 40 FR Section 60.8(c) simply by stating it in the applicable subpart, which the EPA has done by finalizing Subpart Ga of Part 60.
   	8h. Affirmative defense-shifting the burden of proof
Comment 8.8:   Commenter 0091 questions the EPA's legal authority in the CAA to shift the burden to the regulated community of proving (or disproving) essential elements of an alleged violation. The commenter believes that the statute is silent as to the issue and "the ordinary default rule [is] that plaintiffs bear the risk of failing to prove their claims." Shaeffer v. Weast, 546 U.S. 49 (2005), quoting McCormick on Evidence §337, at 412. 
      ("The burdens of pleading and proof with regard to most facts have and should be assigned to the plaintiff who generally seeks to change the present state of affairs and who therefore naturally should be expected to bear the risk of failure or proof or persuasion"); C. Mueller & L. Kirkpatrick, Evidence § 3.1, p. 104 (3d ed. 2003) ("Perhaps the broadest and most accepted idea is that the person who seeks court action should justify the request, which means that the plaintiffs bear the burdens on the elements in their claims"). 
The commenter notes that while the Supreme Court has recognized exceptions such as affirmative defenses, the commenter asserts that courts retain the authority to establish such rules unless Congress acts to delegate that authority. In this instance, according to the commenter, the EPA has not provided any statutory authority, nor any real justification, for requiring a source to prove its innocence; moreover, to fully demonstrate its innocence within 30 days of the event, without even being charged. The commenter states that if the EPA adopts an approach along the lines of the proposed affirmative defense, it should be stated instead in terms that, once a source has claimed that its excess emissions were related to a malfunction, it will not be considered to be in violation of the standards unless the enforcement authority demonstrates that the source is not entitled to claim the malfunction.
Response to Comment 8.8: The EPA disagrees with comments that criticize the affirmative defense for shifting the burden of proof. The affirmative defense does not require a facility to prove its innocence rather than requiring an enforcement authority to prove a violation of the CAA or change the burden of proof with respect to establishing a violation. The affirmative defense applies to penalties and thus is only utilized where a violation has been established. The burden of proof remains with the plaintiff in an enforcement action. See, e.g., 40 C.F.R. 22.24. If a violation has been established and a source wishes to assert the affirmative defense with respect to penalties, the source does bear the burden of establishing that the elements of the affirmative defense have been met. This burden-shifting is appropriate because the source is in a better position to determine the facts required to establish the defense. See, e.g., Arizona Pub. Serv. Co. v. EPA, 562 F.3d 1116, 1120, 1129-30 (10th Cir. 2009) (rejecting industry challenge to the EPA's use of an affirmative defense to address excess emissions during malfunction events). 
8i. Affirmative defense -- injunctive relief
Comment 8.9:   Commenter 0091 notes that The Proposed Rule states that: "The affirmative defense shall not be available for claims for injunctive relief." According to the commenter, the preamble to the Proposed Rule does not give any explanation for why the affirmative defense would not apply to injunctive relief. If in fact the excess emissions associated with the equipment or process failure are not reasonably preventable, the commenter asserts that there is no apparent reason why injunctive relief should be available either. According to the commenter, as a matter of law, injunctive relief may not be available in cases where a civil penalty cannot be imposed. See Sierra Club v. Otter Tail Power Co., 615 F.3d 1008 (8th Cir. 2010) (under concurrent remedy doctrine, injunctive relief for a CAA violation is barred when civil penalty is barred by statute of limitations).
The commenter states that maintaining liability for injunctive relief renders the affirmative defense particularly ineffective with respect to citizen suits. According to the commenter, if the source is even potentially subject to injunctive relief, and therefore could be required to pay the citizen-plaintiff's attorneys fees even if the source successfully demonstrated that it otherwise qualified for the affirmative defense, then the affirmative defense would not accomplish EPA's stated objective of providing relief in situations where the emission limitations cannot be met despite proper design and operation of process and control equipment.
The commenter notes that the EPA has not addressed these and other apparent limitations and shortcomings of the affirmative defense, which make it an entirely inadequate substitute for setting NSPS that include provisions for SSM events. Moreover, according to the commenter, the EPA has provided no analysis that would supersede its long-standing determination that it is not desirable to rely on enforcement, rather than regulatory language, to address the inability to comply with technology-based standards during SSM events. See 37 Fed. Reg. 17,214, 17,214 (Aug. 25, 1972) (establishing SSM provision in NSPS). The  commenter states that courts have adopted the same view. See National Lime, 627 F.2d at 431 n.46 ("the flexibility appropriate to enforcement will not render `achievable' a [NSPS] which cannot be achieved on a regular basis, either for the reasons expressly take into account in compliance determination regulations (here startup, shutdown and malfunction), or otherwise.). See also Marathon Oil Co. v. EPA, 564 F.2d at 1273 (explaining why EPA's proffered assurances that it would not take enforcement action against sources that exceeded effluent limitations because of upset events was "not an adequate response" to the argument that standards that cannot be met during unavoidable upsets fail to reflect available technology).
At a minimum, the commenter asks that the EPA state that the affirmative defense applies to civil penalties, civil administrative penalties, noncompliance penalties, and injunctive relief, in an action brought by the EPA, a state, or a citizen-suit plaintiff. The commenter also asks that the EPA reword the "affirmative defense," so that it states that a source "will not be deemed in violation of" the NSPS for excess emissions or other deviations from the standards, associated with a startup, shutdown, or malfunction event, unless the event, and the source's response to the event, do not meet the criteria spelled out in the regulations. Configured in that way, the commenter believes that this provision should be called something other than an "affirmative defense," such as an "alternative standard for SSM events."
Response to Comment 8.9: In exercising its authority under section 111 to establish emission standards (at a level that meets the stringency requirements of section 111), the EPA necessarily defines conduct that constitutes a violation. The EPA's view is that the affirmative defense is part of the emission standard and defines two categories of violation. If there is a violation of the emission standard and the source demonstrates that all the elements of the affirmative defense are met, only injunctive relief is available. All other violations of the emission standard are subject to injunctive relief and penalties.
8j. Errors and Inconsistencies in Affirmative Defense language
Comment 8.10:   Commenter 0098 notes errors and inconsistencies in the affirmative defense language.
Commenter 0098 states that the definition of a "malfunction" in the proposed affirmative defense provision is broader than that set forth in EPA's NSPS definition of a "malfunction." In proposed §60.74a(a)(1)(i), the commenter notes that a malfunction includes "excess emissions" that "[w]ere caused by a sudden, infrequent, and unavoidable failure of air pollution control and monitoring equipment, process equipment, or a process to operate in a normal or usual manner." (emphasis added). By comparison, the commenter notes that the definition of a malfunction set forth in the §60.2 definition is "a sudden, infrequent, and not reasonably preventable failure of air pollution control equipment, process equipment, or a process to operate in a normal or usual manner." (emphasis added). According to the commenter "unavoidable" in proposed § 60.74a(a)(1)(i) sets a much higher bar than "not reasonably preventable" in §60.2. If EPA wants to change the definition of a malfunction for purposes of the NSPS program, or impose a more stringent definition on a particular source category, the commenter demands that EPA explain its departure from the long-standing agency position.
Commenter 0098 states that the affirmative defense also contains several references to "excess emissions." According to the commenter, this term is neither defined in the general NSPS provisions nor in proposed Subpart Ga. The commenter asks that the EPA propose a definition of "excess emissions" before moving forward with the affirmative defense.
Response to Comment 8.10: Subpart Ga has not been changed to include a definition for "excess emissions." Changes were made to the affirmative defense language to refer to "violations" rather than "excess emissions." 
The EPA disagrees that the criteria of the affirmative defense changes the definition of malfunction (as defined in 40 CFR 60.2) for purposes of the NSPS program. With respect to the wording difference between the §60.2 definition of "malfunction" and the criterion stated in §60.74a(a)(1)(i), the EPA does not intend that the affirmative defense will be available in every instance of a malfunction.
8k. Revisions to Affirmative Defense Language
Comment 8.11:  Multiple commenters offered suggestions for changing the affirmative defense language.
Commenter 0095 suggests the following revisions to §60.74(a):
         (a) (2) Repairs were made as expeditiously as practical possible when the applicable emission limitations were being exceeded. Off shift and overtime labor were used, to the extent practical to make these repairs; and
The commenter notes that responding to malfunctions many times is a case-by-case situation depending on the specifics of the malfunction and equipment to be repaired, and the language regarding off-shift and overtime labor is unnecessary and confusing. According to the commenter, the use of overtime or off-shift labor is only one consideration in how to address a malfunction event. If the EPA wants to emphasize that the use of overtime or off-shift labor must be used, the commenter recommends this language be included in the preamble rather than the rule itself.
         (a) (7) All of the actions in response to the excess emissions were documented by properly signed, contemporaneous operating logs; and
The commenter notes that electronic media to document actions in response to excess emissions is used in the commenter's plant that will be subject to this rule. The commenter notes that imposing a requirement to print and sign a record of such action is outdated if one can document the response to actions in an electronic format.
Commenter 0095 recommends that if the EPA chooses to retain the affirmative defense provisions in the regulation:
For malfunctions that result in only small levels of excess NOX emissions, instead of requiring the regulated entity to establish and prove an affirmative defense, the commenter asks that the EPA establish a work practice requirement that requires the owner/operator to stop the emissions, stop the malfunction, and to return to normal operations as soon as practical.
In Section 60.74a, the EPA proposes that nine requirements be met in order for a facility to claim an affirmative defense for a malfunction; according to the commenter, some of these requirements are unreasonable, difficult to demonstrate, and subject to varying interpretation.
Commenter 0100 suggests the following modifications to the language to make it more usable: Requirements in § 60.74a are impossible to meet due to the use of ambiguous terms such as "careful," "proper," or "better." Until these are defined, the commenter thinks it is impossible to determine whether these criteria have been met. The commenter suggests the following modifications to the language to make it more usable if the EPA keeps the affirmative defense concept: 
The commenter asks that the EPA drop the reference to "any" activity in § 60.74a. According to the commenter, there are also several references to "All" that would make it difficult to ever satisfy the requirements of an affirmative defense.
According to the commenter, the language in the provision is contradictory; in paragraph (a), the phrase "preponderance of evidence" is used while later in that paragraph (iii), the language refers to "any activity."; this same trend occurs in paragraphs (5)  -  "All possible," (6) "All," (7) "All of the actions," and (8) "At all times." According to the commenter, "all" would include "preponderance," but "preponderance" does not mean all of the time. The commenter suggests that the phrase "preponderance of evidence" is adequate and the references to "all" and "any" in the later paragraphs should be modified.
The commenter suggests that the EPA consider making the following modifications (using strikeout to show text deleted and underline to show text added) to the regulatory language in § 60.74a to address the concerns mentioned above and to make an affirmative defense a more useful tool.
         § 60.74a Affirmative Defense for Exceedance of Emission Limit During Malfunction.
         In response to an action to enforce the standards set forth in paragraph § 60.72a, you may assert an affirmative defense to a claim for civil penalties for exceedances of such standards that are caused by malfunction, as defined at 40 CFR 60.2. Appropriate penalties may be assessed, however, if you fail to meet your burden of proving all of the requirements in the affirmative defense. The affirmative defense shall not be available for claims for injunctive relief.
         (a) To establish the affirmative defense in any action to enforce such a limit, you must timely meet the notification requirements in paragraph (b) of this section, and must prove by a preponderance of evidence that:
         (1) The excess emissions:
         (i) Were caused by a sudden, infrequent, and unavoidable failure of air pollution control and monitoring equipment, process equipment, or a process to operate in a normal or usual manner, and
         (ii) Could not have been reasonably prevented through careful planning, proper design or better operation and maintenance practices; and
         (iii) Did not stem from any activity or event that could have been reasonably foreseen and avoided, or planned for; and
         (iv) Were not part of a recurring pattern indicative of inadequate design, operation, or maintenance; and
         (2) Repairs were made as expeditiously as possible when the applicable emission limitations were being exceeded. Off-shift and overtime labor were used, to the extent practicable to make these repairs; and
         (3) The frequency, amount and duration of the excess emissions (including any bypass) were minimized to the maximum extent practicable during periods of such emissions; and
         (4) If the excess emissions resulted from a bypass of control equipment or a process, then the bypass was unavoidable to prevent loss of life, personal injury, or severe property damage; and
         (5) All possible Reasonable steps were taken to minimize the impact of the excess emissions on ambient air quality, the environment and human health; and
         (6) All eEmissions monitoring and control systems were kept in operation if at all possible consistent with safety and good air pollution control practices; and
         (7) All of the aActions in response to the excess emissions were documented by properly signed, contemporaneous operating logs; and
         (8) At all times, tThe facility was operated in a manner consistent with good practices for minimizing emissions; and
         *  *  *
Response to Comment 8.11: The EPA has no plans to make any changes to the language at this time. We believe the language is clear, not contradictory in nature, and provides an owner or operator a systematic and thorough approach for establishing an affirmative defense in the event of a malfunction that results in a violation of a standard. Many of the conditions were modeled after the conditions of the affirmative defense in the EPA's SIP SSM policy, which several states have adopted into their SIPs. We do not have any indication that parties to enforcement proceedings have had any significant difficulties applying the terms of these SIP affirmative defenses. (See, e.g., State Implementation Plans: Policy Regarding Excessive Emissions During Malfunctions, Startup, and Shutdown (Sept. 20, 1999); Policy on Excess Emissions During Startup, Shutdown, Maintenance, and Malfunctions (Feb. 15, 1983)). Other conditions were modeled after a Federal Implementation promulgated by EPA. ((40 CFR 50.1312). The EPA's view is that use of consistent terms in establishing affirmative defense regulations and policies across various CAA programs will promote consistent implementation of those rules and policies.
8l. Affirmative defense -- general comments
Comment 8.12:  Commenter 0104 states for the sake of argument that if the EPA does have authority to promulgate any type of affirmative defense to penalties for malfunctions, the commenter questions whether there is any basis in the record that the EPA has developed for establishing an affirmative defense to an emission standard that is based on (a) a 30-day rolling emission rate, and (b) performance already being achieved at older plants, rather than emission levels that ought to be expected from new or modified units using state of the art pollution controls. The commenter understands that certain acts of God or force majeure events may be impossible to anticipate, but it is not clear that the affirmative defense discussed in the EPA's preamble is limited to these extreme circumstances. According to the commenter, malfunctions should be of extremely limited duration, and the EPA has not shown why any malfunction that is serious enough to cause a 30-day rolling emission rate to be violated ought to qualify for an affirmative defense. The commenter notes that these defenses were originally designed for short-term standards that had to be met every three hours, not emission limits that are based on much longer compliance time periods.
If the EPA can provide some legitimate rationale for adopting an affirmative defense, the commenter suggests that the EPA (1) limit the availability of the affirmative defense to true acts of God or force majeure events; (2) provide that the affirmative defense cannot be used by a specific facility or company more than once within a set period of time, such as 10 years; and (3) promulgate specific public reporting and notification requirements for all malfunctions and violations of which an operator is aware, including publication on the EPA's website and other means to eliminate the need for any person to submit a FOIA request for such a report.
Comment 8.13:  Commenters (0091, 0098) note that the EPA's suggestion that a NAP can avoid high emissions related to a malfunction by shutting the NAP down makes little sense, given the high emissions associated with shutdowns.
Commenter 0091 states that this approach is not consistent with CAA section 111. According to the commenter, neither the NSPS General Provisions nor past EPA practice requires a plant to shut down whenever a malfunction results in excess emissions. The commenter states that basing NSPS emission limitations on an assumption that a plant will shut down during a malfunction is inconsistent with the statutory requirement that NSPS reflect emissions that are "achievable" by facilities using the best system of emissions control.
Comment 8.14:   Commenters (0091, 0093) state that the proposed affirmative defense is not a substitute for addressing malfunction events in the performance standards themselves. According to the commenters, the EPA acknowledges that the sources subject to the Proposed Rule sometimes will be unable to comply with the Proposed Standards because of malfunctions, even if their equipment is properly designed and maintained, through no fault of the source. The commenters asks that the EPA promulgate a standard of performance that eliminates that situation, so that the regulated emission sources will be subject to differentiated requirements, achievable with the identified best demonstrated technology, during malfunction events, rather than an "affirmative defense." According to the commenters, the criteria to be met and actions taken by the source bear no direct relation to the statutory factors for standards of performance under CAA section 111. The commenters state that the EPA should set emission standards that are achievable during malfunctions. According to the commenter, the Proposed Standards, incorporating the affirmative defense, do not represent emission limitations that have been "adequately demonstrated" to be "achievable" through use of the best control technology under CAA section 111(a)(1). Commenter 0093 suggests that Section 111(h) of the Clean Air Act provides EPA with several options for setting standards to be met during periods of startup, shutdown or malfunction, including the use of work practices. The commenter believes that the EPA should utilize the flexibility accorded to the Agency by various sections of the CAA to develop standards that reflect the demonstrated performance that can be achieved during periods of SSM with technology that has been adequately demonstrated.
Comment 8.15:  Commenter 0100 recommends that the EPA modify the affirmative defense provisions so that it is a "rebuttable presumption." The commenter agrees that some form of enforcement discretion is needed for malfunctions, and the commenter supports the EPA maintaining a regulatory provision for malfunctions such as an affirmative defense. The commenter expresses concern that by labeling this as an affirmative defense, it implies that the facility is guilty until proven innocent. The commenter notes that the last sentence in 40 CFR 22.24(a) states that the "respondent has the burdens of presentation and persuasion for any affirmative defenses" while the first sentence in this section states that the EPA has the burden of presentation and persuasion. The commenter is concerned that by calling something an affirmative defense even before it has been established to be a deviation improperly shifts the burden to the facility. The commenter suggests that the EPA establish a rebuttable presumption (rather than affirmative defense) where it is presumed that the facility did everything in their power to minimize emissions during these events, unless the Agency proves certain facts that are enumerated in the rules. If the Agency wants to challenge these activities, the commenter notes that the burden of proof would be on them to show that the facility did not undertake reasonable actions to minimize emissions. 
Comment 8.16:   Commenter 0091 states that it is unclear how the affirmative defense would apply to enforcement actions by state and local governments, or to private citizen enforcement actions under CAA section 304. The commenter notes that the preamble to the Proposed Rule speaks only in terms of application of the affirmative defense in an EPA enforcement action. According to the commenter, an affirmative defense should clearly state that it is applicable to enforcement actions by states or citizen-suit plaintiffs, as well.
Comment 8.17:  Commenter 0100 notes that it is impossible to eliminate the causes for certain malfunctions (e.g., lightning strikes).
Comment 8.18:  According to Commenter 0091, the EPA states that in the preamble that, if a source cannot prove its entitlement to the affirmative defense, "appropriate penalties could be assessed in accordance with section 113 of the CAA (see also 40 CFR part 22.77)." 76 Fed. Reg. at 63,884 col. 2. The commenter notes that this does not answer the questions stated above. Moreover, the commenter notes that there is no "part," nor section, 22.77 in the July 1, 2010, edition of Title 40 of the Code of Federal Regulations (the most recent version available for Part 22). The Proposed Rule, according to the commenter, is unclear on this key issue of what the affirmative defense is intended to cover.
Commenter 0091 disagrees with the conditions of paragraph §60.74a(b) and states that it is novel at best for a person to be determined to have acted unlawfully unless the person has submitted his entire defense before he is even notified of a potential enforcement action. In many cases, the commenter believes it would be obvious to the enforcement authority, based on the kind of short malfunction or deviation report sources already submitted under many air programs, that an exceedance of the Proposed Standards resulted from an unforeseen and unavoidable equipment failure or process upset. The commenter further states that it is extremely inefficient and burdensome, for both sources and regulators, to require a complete justification of the affirmative defense before the enforcement authority has indicated any need for further investigation. 
According to commenter 0091, a number of the conditions for establishing an affirmative defense use phrases that are subject to a wide range of interpretations, and that on their face do not recognize any need for reasonableness or cost-effectiveness. The commenter asks, "how will the enforcement authority, or a judge, determine whether `proper design' or `better operation and maintenance practices' could have prevented a malfunction (section 60.74a(a)(1)(ii)), whether a recurring malfunction is a result of `inadequate design' (section 60.74a(a)(1)(iv)), whether repairs were made `as expeditiously as possible' (section 60.74a(a)(2)), whether the source took `all possible steps' to minimize the impact of the excess emissions (section 60.74a(a)(5)), and whether emissions monitoring and control systems `were kept in operation if at all possible' (section 60.74a(a)(6))? The commenter notes that in many, if not most cases, it may have been possible to avoid the malfunction, or to do more to reduce the excess emissions, if the source had spent huge amounts of money or had imposed economically impracticable constraints on its operation. The commenter believes that the affirmative defense, as proposed, leaves open the possibility that a source will be considered to be in violation because the enforcement authority decides that in one or more respects it would have been "proper" or "possible" for the source to take further steps to prevent or minimize the malfunction. In effect, the commenter states that the EPA or the court may impose an extreme version of best demonstrated technology, during malfunction periods. At a minimum, the commenter is concerned that the vague and unqualified descriptors in the criteria for demonstrating the affirmative defense will inevitably lead to varying conclusions as to whether a violation has occurred, resulting in inconsistency from one jurisdiction to the next.
According to commenter 0091, several of the conditions for establishing an affirmative defense in proposed 40 C.F.R. § 60.74a list numerous items which may be appropriate considerations in determining whether excess emissions during a malfunction should be considered a violation, but which ought not to be listed as mandatory criteria, all of which have to be met in order for an exceedance to be excused. For example, the commenter notes that proposed section 60.74a(a)(1)(i) requires that the excess emissions must have been "caused by a sudden, infrequent, and unavoidable failure of air pollution control and monitoring equipment, process equipment, or a process to operate in a normal or usual manner...." The commenter notes that there is no apparent reason why a "sudden" and "unavoidable" equipment failure or process upset leading to emissions greater than the NSPS should be considered a violation of the standards if it is not "infrequent," even though it is "not part of a recurring pattern indicative of inadequate design, operation, or maintenance" and "[c]ould not have been prevented through careful planning, proper design or better operation and maintenance practices." See 60.74a(a)(1)(ii) and (iv).
According to commenter 0091, proposed section 60.74a(a)(6) requires, as a condition for the affirmative defense, that all emissions "control systems were kept in operation if at all possible." The commenter states that the phrase "if at all possible" should not be used because it is an extreme term that bears no relation to good air pollution control practices. Additionally, the commenter asks that this provision (as well as the "general duty" language the EPA proposes to add to the Nitric Acid Plant NSPS, as discussed below), should be qualified, as the EPA has qualified similar provisions in the NSPS General Provisions in 40 CFR §60.11(d) and the NESHAP General Provisions in 40 CFR §63.6. According to the commenter, these conditions should be qualified with caveats that the operation must be consistent with safety and good air pollution control practices, that it does not require the source to make further efforts to reduce emissions below what the standards require, and that it does not require regular operation of backup or standby pollution control equipment. [As noted by the commenter, the EPA agreed (40 CFR §63.483(a), 76 Fed. Reg. 22,566, 22,583, 22,588 (April 21, 2011)) that it was appropriate to include such a qualifier in comparable language in NESHAPs.] The commenter notes that the EPA has long recognized, in the General Provisions applicable to NSPS, that it is appropriate to require sources to operate the affected facility and related air pollution control technology "to the extent practicable ... consistent with good air pollution control practice for minimizing emissions" during SSM periods, not "if at all possible." See 40 CFR §60.11(d). The commenter asserts that the EPA cannot abandon those rational approaches and adopt the kind of absolute requirements implied by the Proposed Rule, without an explanation of why it is necessary and appropriate to do so.
According to commenter 0091, proposed section 60.74a(a)(4) would preclude a facility from taking advantage of the affirmative defense if the malfunction involved bypassing control equipment or a process and the bypass was not "unavoidable to prevent loss of life, personal injury, or severe property damage." The commenter asserts that this language is stated in such strong terms that it may be difficult or impossible for a source to demonstrate that it met this criterion, even though bypassing the control equipment or the process was an appropriate exercise of good air pollution control practices. For example, the commenter notes that a bypass can constitute the best air pollution control practice in response to an upset in order to prevent excess emissions, e.g., to avoid fouling of pollution control equipment media that in turn would result in reduced pollution control equipment efficiency or increased pollution control equipment downtime. According to the commenter, there can be substantial room for disagreement about what constitutes "severe" property damage. Besides the unclear and subjective nature of these criteria, the commenter notes that there is nothing inherent to standards of performance under CAA section 111 that requires a source to avoid bypassing control equipment to such a degree. Lastly, it is not apparent to the commenter at all why the CAA would disfavor bypassing "a process" in this way.
According to the commenter, under proposed section 60.74a(a)(5), a source claiming the affirmative defense must prove that: "All possible steps were taken to minimize the impact of the excess emissions on ambient air quality, the environment and human health." In addition to the subjective term "all possible steps," the commenter notes that this provision again presents a disconnect between the absolute and extreme requirement of the affirmative defense and the provisions of the CAA designed to attain ambient air quality standards and protect human health and the environment. The commenter asserts that the CAA does not require sources to take "all possible steps" to control emissions, even to minimize the impact of hazardous air pollutant emissions on human health under CAA section 112(f). In addition, it is unclear to the commenter how this criterion for qualifying for the affirmative defense differs from proposed section 60.74a(a)(3), which requires that the frequency, amount, and duration of excess emissions "were minimized to the maximum extent practicable." Unless the EPA explains what additional showing would be needed by section 60.74a(a)(5), the commenter asks that it be eliminated.
According to the commenter, the requirement proposed in section 60.74a(a)(8), that the source must show that "[a]t all times, the facility was operated in a manner consistent with good practices for minimizing emissions" (emphasis added), could be read to go way beyond an analysis of why equipment may have malfunctioned and place an overwhelming burden on the source to meet a "good practices" standard (undefined) for the whole industrial plant or commercial premises. The commenter notes that this factor potentially could allow a regulatory enforcement official to judge the entirety of operations at a site against a multitude of arbitrary "good practices" and subjectively find the stationary source lacking, thereby denying the source the ability to assert an affirmative defense for NSPS applicable to a particular operation at the site. If the EPA retains this condition for the affirmative defense, of demonstrating operation consistent with good practices for minimizing emissions, the commenter asks that the Agency clarify that this condition applies to the "affected facility," as that term is defined in 40 C.F.R. § 60.2. If the EPA really does intend for this condition to apply to operation of the entire site, and not just "the apparatus to which a standard is applicable" (see section 60.2 definition of "affected facility"), then the commenter asks the EPA to explain why it has the statutory authority to shift the focus of the Proposed Standards from the apparatus subject to the NSPS to the plant as a whole.
Commenter 0091 feels that the proposed affirmative defense as written is unreasonable and impracticable. The commenter notes that although a source believes it qualifies for the affirmative defense, it may be considered to have violated the standards -- and may have to report violations, certify noncompliance, etc. -- until there has been an enforcement proceeding and the source has successfully asserted the affirmative defense. The commenter asks that the affirmative defense be substantially modified for it to provide any significant relief. First, the commenter asks that the affirmative defense state clearly that a source that qualifies for the affirmative defense shall not be deemed to have violated the applicable standards during that time. If the EPA does that, the commenter notes that it may be unnecessary to state also that the affirmative defense relieves the source from liability for all types of penalties and injunctive relief (save criminal penalties), but that should be the clear effect of qualifying for the affirmative defense.
Commenter 0091 states that it is not clear what the affirmative defense covers. In the Proposed Rule, the commenter notes that the EPA states that the affirmative defense is "to a claim for civil penalties for exceedances of such standards that are caused by malfunction, as defined in section 60.2." Proposed section 60.74a. The commenter asks, "Is the term `civil penalties,' which is not defined in the Proposed Rule, intended to apply as well to a `civil administrative penalty' imposed by the EPA under CAA section 113(d)? (According to the commenter, the term "civil penalty" in other contexts means only penalties imposed by a court.) The commenter asks, "does the affirmative defense apply to `noncompliance penalties' under CAA section 120 (which apply, inter alia, to noncompliance with a section 111 NSPS)? According to the commenter, to meet the purported purpose of the affirmative defense, which is to provide relief from emission limitations that cannot be met at times even with equipment that is properly designed and maintained (see 76 Fed. Reg. at 63,884 col. 1), the affirmative defense would need to apply to civil and administrative penalties, including noncompliance penalties.
Response to Comments 8.12-8.18: We agree with the commenters concerning the use of "affected facility" versus "affected source." As a result, the EPA has changed "affected source" to "affected facility" within the "general duty" clause for Subpart Ga.
We also agree with the commenter that noted that there is no "part," nor section, 22.77 in the July 1, 2010, edition of Title 40 of the Code of Federal Regulations (the most recent version available for Part 22). This was a typographical error in the proposed affirmative defense language. For the final rule, this reference has been corrected to read 22.27.
For malfunctions, the EPA is finalizing the proposed affirmative defense language for violations of the numerical emission limits that are caused by malfunctions with minor revisions as discussed below. The EPA recognizes that even equipment that is properly designed and maintained can fail and that such failure can cause a violation of the relevant emission standard. The EPA is including an affirmative defense in the final rule as we have in other recent section 111, section 112, and section 129 rules so as to balance the tension, inherent in many types of air regulation, to ensure adequate compliance while simultaneously recognizing that despite the most diligent of efforts, emission limits may be exceeded under circumstances beyond the control of the source. The EPA must establish emission standards that "limit the quantity, rate, or concentration of emissions of air pollutants on a continuous basis." 42 U.S.C. § 7602(k)(defining "emission limitation and emission standard"). See generally Sierra Club v. EPA, 551 F.3d 1019, 1021 (D.C. Cir. 2008) (emissions limitations under must both continuously apply and meet minimum stringency requirements, even during periods of startup, shutdown and malfunction). Thus, the EPA is required to ensure that emissions limitations are continuous. The affirmative defense for malfunction events meets this requirement by ensuring that even where there is a malfunction, the emission limitation is still enforceable through injunctive relief. 
While "continuous" limitations, on the one hand, are required, there is also caselaw indicating that in some situations it is appropriate for the EPA to account for the practical realities of technology. For example, in Essex Chemical v. Ruckelshaus, 486 F.2d 427, 433 (D.C. Cir. 1973), the D.C. Circuit acknowledged that in setting standards under CAA section 111 "variant provisions" such as provisions allowing for upsets during startup, shutdown and equipment malfunction "appear necessary to preserve the reasonableness of the standards as a whole and that the record does not support the `never to be exceeded' standard currently in force." See also, Portland Cement Association v. Ruckelshaus, 486 F.2d 375 (D.C.Cir. 1973). Though intervening caselaw such as Sierra Club v. EPA and the CAA 1977 amendments calls into question   the relevance of these cases today, they support the EPA's view that a system that incorporates some level of flexibility is reasonable. The affirmative defense simply provides for a defense to civil penalties for excess emissions that are proven to be beyond the control of the source. By incorporating an affirmative defense, the EPA has formalized its approach to upset events. In a Clean Water Act setting, the Ninth Circuit required this type of formalized approach when regulating "upsets beyond the control of the permit holder." Marathon Oil Co. v. EPA, 564 F.2d 1253, 1272-73 (9th Cir. 1977). But see, Weyerhaeuser Co. v. Costle, 590 F.2d 1011, 1057-58 (D.C. Cir. 1978) (holding that an informal approach is adequate). The affirmative defense provisions give the EPA the flexibility to both ensure that its emission limitations are "continuous" as required by 42 U.S.C. § 7602(k), and account for unplanned upsets and thus support the reasonableness of the standard as a whole.
Further, the EPA's view is that the affirmative defense is consistent with CAA sections 113(e) and  304. Section 304 gives district courts jurisdiction "to apply appropriate civil penalties." Section 113(e)(1) identifies the factors that the Administrator or a court shall take into consideration in determining the amount of a penalty to be assessed only after  it has been determined that a penalty is appropriate. The affirmative defense regulatory provision is not relevant to the amount of any penalty to be assessed under section 113(e) because if a court determines that the affirmative defense elements have been established, then a penalty is not appropriate and penalty assessment pursuant to the section 113(e)(1) factors does not occur. 
In exercising its authority under section 111 to establish emission standards (at a level that meets the stringency requirements of section 111), the EPA necessarily defines conduct that constitutes a violation. The EPA's view is that the affirmative defense is part of the emission standard and defines two categories of violation. If there is a violation of the emission standard and the source demonstrates that all the elements of the affirmative defense are met, only injunctive relief is available. All other violations of the emission standard are subject to injunctive relief and penalties. The CAA does not require that all violations be treated equally. 
Further, a citizen suit claim under section 304 allows citizens to commence a civil action against any person alleged to be in violation of "an emission standard or limitation under this chapter." The CAA, however, allows the EPA to establish such "enforceable emission limitations." Thus, the citizen suit provision clearly contemplates enforcement of the standards that are defined by the EPA. As a result, where the EPA defines its emissions limitations and enforcement measures to allow a source the opportunity to prove its entitlement to a lesser degree of violation (not subject to penalties) in narrow, specified circumstances, as the EPA did here, penalties are not "appropriate" under section 304. 
The EPA's view is that an affirmative defense to civil penalties for exceedances of applicable emission standards during periods of malfunction appropriately balances competing concerns. On the one hand, citizen enforcers are concerned about additional complications in their enforcement actions. On the other hand, industrial sources are concerned about being penalized for violations caused by malfunctions that could not have prevented and were otherwise appropriately handled (as reflected in the affirmative defense criteria). The EPA has utilized its Section 301(a)(1) authority to issue regulations necessary to carry out the Act in a manner that appropriately balances these competing concerns.
The EPA disagrees that the affirmative defense provision will hamper citizen enforcement. First, injunctive relief is still available and the threat of penalties would not deter violations in cases where all of the conditions of the affirmative defense have been satisfied because the affirmative defense criteria ensure that all reasonable steps were taken to prevent a malfunction that causes excess emissions. Further, litigating whether a source has met the affirmative defense will not burden citizen group any more or less than would litigating the appropriate penalty amount in the penalty assessment stage of a citizen suit enforcement action, because the 113(e) penalty assessment criteria and the affirmative defense criteria are similar and in fact overlap. For example, the requirement that the Administrator or the court consider "good faith efforts to comply" is bound to generate the type of fact-intensive disputes that the commenter complains of. In addition, several of the affirmative defense criteria are exactly the type of criteria the Administrator or Court might consider in determining whether a source made "good faith efforts to comply." For example, to take advantage of the affirmative defense, the source must prove by a preponderance of the evidence that, among other things, the excess emissions "were caused by an unavoidable failure of air pollution control and monitoring equipment, process equipment, or a process to operate in a normal or usual manner" and "could not have been prevented through careful planning, proper design or better operation and maintenance practices" and "did not stem from any activity or event that could have been foreseen and avoided, or planned for." 
Thus, the EPA does not expect the affirmative defense provision to significantly alter the burden of bringing a citizen enforcement action. For those cases that do proceed to trial, even in the absence of this affirmative defense, sources generally raise equitable arguments to argue for a low penalty and citizens often rebut such arguments. Therefore, as a practical matter, the EPA does not expect the affirmative defense provision to materially affect the practice of CAA enforcement.
Additionally, the affirmative defense is reasonable. The EPA's judgment is that the affirmative defense criteria capture the appropriate considerations in determining whether penalties are appropriate when a violation occurs as the result of a malfunction. As noted above, the affirmative defense criteria overlap to some extent with the penalty assessment criteria set forth in section 113(e), but are not identical. For example, size of business is one of the factors listed in section 113(e), but is not reflected in the EPA's affirmative defense. This reflects the EPA's view that when a violation is caused by a malfunction, the size of the business is not relevant to whether penalties should be excused. If the violation was unavoidable and could not have been prevented, the EPA's view is that it would be unfair to impose a penalty no matter the size of the business. 
Further, the EPA disagrees with comments that criticize the affirmative defense criteria as being overly vague or unduly restrictive and complex. The EPA believes that courts are well equipped and often do evaluate and apply the type of criteria set forth is the affirmative defense. Many of the conditions were modeled after the conditions of the affirmative defense in the EPA's SIP SSM policy, which several states have adopted into their SIPs. (See, e.g., State Implementation Plans: Policy Regarding Excessive Emissions During Malfunctions, Startup, and Shutdown (Sept. 20, 1999); Policy on Excess Emissions During Startup, Shutdown, Maintenance, and Malfunctions (Feb. 15, 1983)). We do not have any indication that parties to enforcement proceedings have had any significant difficulties applying the terms of these SIP affirmative defenses. In addition, the EPA's view is that use of consistent terms in establishing affirmative defense regulations and policies across various CAA programs will promote consistent implementation of those rules and policies. 
The EPA does not find persuasive commenters' concerns with potential use of the root cause analysis against the source. The purpose of the root cause analysis is to determine, correct, and eliminate the primary causes of the malfunction and the excess emissions resulting from the malfunction event at issue. A source's ability to do so with the benefit of hindsight after a malfunction has occurred does not necessarily mean that malfunction could have been predicted or prevented.
8m. Affirmative defense should include startup and shutdown
Comment 8.19:   Commenter 0091 states that the EPA offers no defensible rationale for excluding startup and shutdown events from the proposed affirmative defense. The commenter notes that the EPA has recognized it may not be possible to meet normally applicable emission limitations during startup, shutdown, and malfunction events for many NSPS for decades. According to the commenter, if the EPA does not provide separate emission limitations for startup and shutdown of regulated units, and if excess emissions from those units cannot reasonably be avoided during a startup or shutdown, not because of failure of a process or equipment but because of the nature of conditions while starting up or shutting down the source, there is no justification for the EPA not to provide the same kind of affirmative defense that it proposes to provide for malfunctions. According to the commenter, the EPA offers that startup and shutdown periods at nitric acid plants "generally" are of relatively short duration (see 76 Fed. Reg. at 63,883 col. 2) and that the higher emissions during startup or shutdown would not cause an exceedance of the proposed 30-day NOX emission limitation because of that short duration. Even if this assumption is true, the commenter states that it would not justify excluding startup and malfunction events from the affirmative defense, for those situations where the startup or shutdown event is of such a duration or magnitude that it causes an exceedance of the 30-day emission limitation.
Response to Comment 8.19: The EPA is promulgating the affirmative defense for malfunctions only, not for periods of startup and shutdown. As explained in the proposed preamble, the EPA believes that malfunction events should be treated differently than periods of startup and shutdown, which are predictable and routine aspects of a source's operations. In contrast, the EPA has determined that CAA section 111 does not require that emissions that occur during malfunctions be factored into development of CAA section 111 standards. Because startup and shutdown periods are part of a source's normal operations, the same approach to compliance with, and enforcement of, applicable emissions standards during those periods should apply as otherwise applies during a source's operations. Further, as explained above, periods of startup and shutdown  -  but not malfunctions -- are taken into account when establishing section 111 emissions standards. For these reasons, the EPA does not believe it is appropriate to apply the affirmative defense provision to startup and shutdown periods. Also see Section V of the final rule preamble.
8n. Notification requirements for affirmative defense
Comment 8.20:   Multiple commenters noted that the notification requirements for affirmative defense do not allow sufficient time to provide the documentation required.
Commenter (0091) states that allowing only 45 days to provide the kind of extensive documentation required by the affirmative defense, including a completed root cause analysis, is unreasonable. According to the commenter, ninety days is the minimum time that should be allowed, unless EPA substantially streamlines the criteria for the affirmative defense, consistent with these comments. The commenter notes that the proposed 45-day period for submitting a written report demonstrating that the source qualifies for the affirmative defense commences on the date of "the initial occurrence of the exceedance of the standard." This presents several problems to the commenter: much of the required content of that written report could not be created until the malfunction event ceased -- which could be days or even weeks after the malfunction commenced. According to the commenter, this could make the amount of time the source would have, in practice, to prepare the report much less than 45 days. For the particular emission limitation in the Proposed Rule, which is a 30-day rolling average, the commenter notes that the source operator often will not know whether the malfunction will result in emissions that exceed the emission limit until much later, potentially almost 30 days after the malfunction occurred. According to the commenter, this makes the proposed requirements to notify the Administrator within 48 hours after, and submit a written report within 45 days after, the malfunction commenced virtually unworkable. The commenter asserts that EPA cannot claim credit for an "affirmative defense" that many sources will be precluded from using because of the disconnect between the averaging period and the short deadlines for phone/fax and written reporting.
Commenter (0091) states that the provision in the Proposed Rule for requesting and obtaining an extension of the reporting deadline of up to 30 days would provide little practical relief: the source would have to submit the report within 45 days unless the EPA Administrator (or his or her authorized representative) granted the extension request before the expiration of the initial 45-day period, which is not likely to happen until shortly before, if not after, the report would otherwise be due. At a minimum, the commenter asks that the rule provide that a request for extension of the reporting deadline that the Agency has not acted on within 10 days is considered granted.
Commenter (0091) requests that malfunction reporting be required on a semi-annual basis, for malfunctions that occurred during the preceding six months (the same frequency as similar reports pursuant to NESHAPs and Title V permit requirements). According to the commenter this would enable EPA to review the source's compliance history and the measures being taken to address malfunctions, without imposing unworkable reporting deadlines (or potentially encouraging sources to bombard the Agency with reports for malfunctions that may not ultimately result in the source exceeding the 30-day average emission limitation).
Commenter (0100) suggests the following modifications to the language to make it more usable: Faxing is an obsolete technology. The commenter asks that EPA allow notification by e-mail or other electronic means. The commenter suggests that EPA consider making the following modifications (using underline to show text added) to the regulatory language in §60.74a to address the concerns mentioned above and to make an affirmative defense a more useful tool.
         § 60.74a Affirmative Defense for Exceedance of Emission Limit During Malfunction.
          * * * * *
         (b) Notification. The owner or operator of the facility experiencing an exceedance of its emission limit(s) during a malfunction shall notify the Administrator by telephone, or facsimile (FAX) transmission, or electronic means as soon as possible, but no later than two business days after the initial occurrence of the malfunction, if it wishes to avail itself of an affirmative defense to civil penalties for that malfunction. The owner or operator seeking to assert an affirmative defense shall also submit a written report to the Administrator within 45 days of the initial occurrence of the exceedance of the standard in § 60.72a to demonstrate, with all necessary supporting documentation, that it has met the requirements set forth in paragraph (a) of this section. The owner or operator may seek an extension of this deadline for up to 30 additional days by submitting a written request to the Administrator before the expiration of the 45 day period. Until a request for an extension has been approved by the Administrator, the owner or operator is subject to the requirement to submit such report within 45 days of the initial occurrence of the exceedance.
Commenters (0091, 0095) suggest the following revisions to §60.74a:
         (b) The owner or operator of the facility experiencing an exceedance of its emission limit(s) during a malfunction must notify the Administrator by telephone or facsimile (FAX) transmission as soon as possible, but no later than 2 working days after the initial occurrence of the malfunction, if it wishes to avail itself of an affirmative defense to civil penalties for that malfunction. The owner or operator seeking to assert an affirmative defense must also submit a written report to the Administrator as part of the semi-annual periodic report in within 45 days of the initial occurrence of the exceedance of the standard in 63.72(a) to demonstrate, with all necessary supporting documentation, that is has met the requirements set forth in paragraph (a) of this section. The owner or operator may seek an extension of this deadline for up to 30 additional days by submitting a written request to the Administrator before the expiration of the semi-annual reporting 45 day period. Until a request for an extension has been approved by the Administrator, the owner or operator is subject to the requirement to submit such report within the required timeframe. 45 days of the initial occurrence of the exceedance.
Commenter (0095) recommends the immediate notification (within 2 working days) requirement as a pre-requisite for asserting the affirmative defense should be eliminated in section 60.74a(b) for at least the following reasons and replaced with a semi-annual reporting requirement:
      * The commenter asserts that it is burdensome and duplicative of emergency response reporting requirements under SARA/CERCLA/EPCRA and many state regulations for larger excess emission events; the reportable quantity for excess emissions of NO and NO2 from non-combustion sources is 10 lbs.
      * According to the commenter, this new requirement will contradict many existing reporting rules by requiring the reporting of emission quantities below CERCLA/SARA/State Agency Reportable Quantities. Thus, the commenter fears that agencies could be flooded with reports of very small excess emissions.
      * The commenter notes that the rule structure encourages reporting to the agency even if the owner/operator is not sure if an exceedance of an emission limit has occurred for fear of not being able to claim the affirmative defense. Thus, the commenter is concerned that agency reporting systems will become "clogged" with insignificant reports further complicating the ability of all resources to respond to a larger event.
      * According to the commenter, the rule is not clear to whom the reports should be submitted (e.g., the EPA Regional Office, State or Local Agency, LEPC).
Commenter (0098) is concerned with the requirement that owners and operators of NAPs must notify EPA "no later than two business days after the initial occurrence of the malfunction." Due to the 30-day averaging period specified in the Nitric Acid NSPS Proposal, the commenter notes that an owner/operator may not realize that there was a malfunction within the 30-day averaging period that resulted in "excess emissions" until after that 30-day period, which negates the ability to assert the affirmative defense. Further, given holidays, the commenter notes that a NAP owner/operator may not become aware of a malfunction until longer than two days after the malfunction occurred.
Response to Comment 8.20: We agree with your comments on the notification period related to the affirmative defense language. We also agree with your comments concerning the reporting requirements. The language has been changed in the final rule at paragraph 60.74a(b) to read as follows:
      (b) Report. The owner or operator seeking to assert an affirmative defense shall submit a written report to the Administrator with all necessary supporting documentation, that it has met the requirements set forth in paragraph (a) of this section. This affirmative defense report shall be included in the first periodic compliance, deviation report or excess emission report otherwise required after the initial occurrence of the violation of the relevant standard (which may be the end of any applicable averaging period). If such compliance, deviation report or excess emission report is due less than 45 days after the initial occurrence of the violation, the affirmative defense report may be included in the second compliance, deviation report or excess emission report due after the initial occurrence of the violation of the relevant standard. 

Also in response to the request for electronic means of notification, the EPA accepts documents in electronic format, as long as the format is compatible with the requirements of the standards. For the affirmative defense provisions, the owner or operator of a facility experiencing an exceedance of its emission limit(s) during a malfunction must notify the Administrator by telephone or facsimile (FAX) transmission of the exceedance. However, the written reports required to demonstrate that the affirmative defense provisions have been met and requests for an extension of the deadline for these reports may be submitted electronically. 
          8o. Root Cause Analysis-suggestions for language changes
Comment 8.21:  Multiple commenters (0095 and 0100) suggested revisions to the root cause language.
Commenter (0095) suggests the following revisions to §60.74(a):
         (a)(9) A written root cause analysis has been prepared, the purpose of which is to determine, correct, and eliminate the primary causes of the malfunction and the excess emissions resulting from the malfunction event at issue. The analysis must also specify using best monitoring methods and engineering judgment, the amount of excess emissions that were the result of the malfunction.
The commenter notes that EPA should not require preparation and submission of a formal root cause analysis as this would have a chilling effect on internal communications regarding the root cause investigation and would detract from the goal of ensuring a candid inquiry into the event. According to the commenter, the cause of even smaller malfunction events is routinely investigated within the industry, and no regulation is necessary to require this to happen.
Commenter (0100) suggests modifications to the rule language regarding root cause analysis. To many engineers, the commenter notes that the term "root cause analysis" implies a formal process. According to the commenter, for many malfunctions, the cause is immediately obvious and a formal process for determining the cause is not needed. When a malfunction occurs, the commenter notes that the facility is expected to correct the problem as quickly as possible and return to their operating window. The commenter states that a formal root cause analysis is typically limited to very significant events or repeat events. The commenter notes that the proposed language assumes that all malfunctions are equally significant and need an identical degree of investigation. The commenter believes that a formal root cause analysis should only be used when other reasonable methods fail to show what caused the malfunction or when the serious nature of an event might make such an analysis necessary. Moreover, the commenter believes that other tools may be more appropriate (e.g., failure mode and effect, fault tree, etc.) or more powerful tools may be introduced in the future. According to the commenter, the facility is the only one that can and should decide what tool to use to determine the cause of the malfunction. The commenter states that part of this problem may be in communications. If EPA intends for the facility to investigate and fix the problem so that is it less likely to recur, the commenter supports that concept but suggests that the Agency use an alternative term that does not carry a specific meaning. However, if the Agency envisions a formal process for determining the root cause for every malfunction, no matter how simple, the commenter believes this is unnecessary and would result in excess efforts with no environmental gains. The commenter suggests that EPA consider making the following modifications (using strikeout to show text deleted and underline to show text added) to the regulatory language in § 60.74a to address the concerns mentioned above and to make an affirmative defense a more useful tool.
§ 60.74a Affirmative Defense for Exceedance of Emission Limit During Malfunction.
    * * * * *
(9) A written root cause analysis report has been prepared, the purpose of which is to determine, correct, and eliminate mitigate the primary causes of the malfunction and the excess emissions resulting from the malfunction event at issue. Facility personnel will determine the appropriate type of analysis required (may include but is not limited to root cause analysis, failure mode and effect, fault tree, etc.) to identify the cause of the malfunction. The analysis report shall also specify, using best monitoring methods and engineering judgment, the amount of excess emissions that were the result of the malfunction.
Response to Comment 8.21: We acknowledge your suggestions for changes to the language related to a root cause analysis. However, EPA has no plans to make any changes to the language at this time.
We believe the language is clear, not contradictory in nature, and provides an owner or operator a systematic and thorough approach for establishing an affirmative defense in the event of a malfunction that results in an exceedance of a standard. Many of the conditions were modeled after the conditions of the affirmative defense in the EPA's SIP SSM policy, which several states have adopted into their SIPs. We do not have any indication that parties to enforcement proceedings have had any significant difficulties applying the terms of these SIP affirmative defenses. (See, e.g., State Implementation Plans: Policy Regarding Excessive Emissions During Malfunctions, Startup, and Shutdown (Sept. 20, 1999); Policy on Excess Emissions During Startup, Shutdown, Maintenance, and Malfunctions (Feb. 15, 1983)). Other conditions were modeled after a Federal Implementation promulgated by EPA. ((40 CFR 50.1312). The EPA's view is that use of consistent terms in establishing affirmative defense regulations and policies across various CAA programs will promote consistent implementation of those rules and policies. 
8p. Root Cause Analysis -- legal obligation
Comment 8.22:   Commenter 0091 notes that the requirement in proposed section 60.74a(a)(9) that the source prepare a "written root cause analysis...to determine, correct, and eliminate the primary causes of the malfunction and the excess emissions resulting from the malfunction event at issue" does not implement the statutory criteria for standard-setting under section 111. The commenter claims that a source could never take advantage of the affirmative defense if the source was unable to determine the primary cause of the malfunction or was unable to correct that cause. According to the commenter, EPA has defined a malfunction as an event that is "unavoidable" and unforeseeable, and "not reasonably preventable." Therefore, the commenter expects that in many cases the primary causes of the malfunction will not be ascertainable, or it will not be possible to identify a way to ensure the malfunction will not recur. According to the commenter, EPA has long acknowledged this reality in the General Provisions applicable to NSPS, which requires that written reports of excess emissions include the "nature and cause of any malfunction, if known...." (See 40 C.F.R. § 60.7(b)(2)). In addition, requiring the facility to eliminate the primary causes of the malfunction, without regard to "taking into account the cost of achieving such" elimination and the "nonair quality health and environmental impact and energy requirements" associated with its elimination is, according to the commenter, unreasonable and entirely inconsistent with the criteria for NSPS established under CAA section 111(a)(1).
Response to Comment 8.22: A root cause analysis is not required for every malfunction - - only those malfunctions for which the source chooses to assert an affirmative defense. The EPA believes that sources seeking to minimize emissions will endeavor to find out what went wrong anytime there is a minor or major malfunction. The facility should decide what level of investigation is needed in each instance. We agree with the commenter that sources should use appropriate techniques of analysis to achieve the desired results and conclusions in the event of a malfunction.
The term "root cause analysis" was used generically to imply an analysis of sufficient depth and complexity to indicate whether a malfunction did indeed cause a failure to meet a standard, provide sufficient information on the nature and causes of a malfunction to determine whether the source had a malfunction that met the definition of a malfunction, and whether civil penalties are an appropriate sanction for the violation, if one occurred. The term "root cause analysis" is not defined in subpart A (the General Provisions) or in subpart Ga. The EPA did not intend to prescribe a specific methodology, given that "root cause analysis" is not a defined term in the applicable subparts of Part 60.
 9. NSCR Versus SCR Used as NOX Control Device
Comment 9.1:   Commenter 0104 notes that based on stack testing and monthly block averages submitted by nitric acid production facilities around the nation, many facilities are actually achieving in practice NOX emission levels significantly lower than BACT standards. According to the commenter, EPA has presented some evidence suggesting that facilities "may be able to achieve the proposed emission limit" using NSCR.
The commenter states that EPA required stack testing in 2010 for twelve nitric acid production lines equipped with either SCR or NSCR controls for NOX. Of the six lines with SCR, the commenter notes that four reported NOX emissions ranging from 0.04 to 0.46 lb/ton, well below the standard EPA has proposed. Of the six lines with NSCR, the commenter notes that four reported test results between 0.01 and 0.35 lb/ton, and a fifth measured NOX emissions at 0.52 lb/ton, just over the proposed limit. If conducted properly, the commenter asserts that test results should reflect peak hourly emissions during normal operations, i.e., long-term averages should be somewhat lower.
Finally, the commenter notes that the "youngest" plant in EPA's survey was built in 2000 and most of the remaining plants are at least 20 years old. According to the commenter, it is reasonable to expect any new unit built today to meet or exceed standards that are being met by production lines installed decades ago.
Comment 9.2:   Commenter 0104 states that EPA's suggestion that NSCR cannot meet or exceed the standard it has proposed is not supported by the record. The commenter notes that the analysis of 2010 stack test results prepared by RTI for EPA found that "[t]he lowest NOX emissions (0.01 lb/ton) originated from two high pressure trains ... controlled by an NSCR." The commenter also notes that conversely, the highest NOX emissions during testing came from PCS Geismar Unit 3, a production line controlled by SCR. After comparing data from units with NSCR to those with SCR, the commenter notes that the author concluded that "[b]oth abatement technologies are capable of emitting less than 0.05 lb/ton of NOX; therefore, the type of abatement does not necessarily result in higher NOX emissions." Moreover, the commenter notes that two other units equipped with NSCR (Dyno-Nobel Deer Island and J.R. Simplot-Helm) achieved monthly block averages between 0.01 and 0.35 lb/ton of NOX in 2009. 
While EPA has appropriately given weight to long-term CEMS data in developing the standard, the commenter notes that the two NSCR-equipped units included in the CEMS analysis (Agrium-Sacramento and PCS-Geismar Train 4) may not adequately represent performance at other units relying on the same control technology. For example, the commenter notes that stack test results from Agrium-Sacramento and PCS-Geismar Train 4 averaged 0.52 and 0.86 lb/ton, respectively, while results for four other plants with NSCR ranged from 0.01 to no higher than 0.35 lb/ton. This discrepancy suggests to the commenter that higher NOX emissions from Agrium-Sacramento and PCS-Geismar Train 4 may result from circumstances unique to those plants that are unrelated to NSCR performance.
Response to Comments 9.1-9.2: As discussed in the preamble to the proposed rule, the EPA chose SCR as BSER for a number of reasons. Based on the emissions data available to the agency, emissions from the ICR test plants controlled via SCR were lower than those using other technologies. Additionally, SCR was chosen as BSER because it is because it is more cost effective than NSCR for control of NOX emissions and it produces minimal secondary environmental impacts. 
Also, we evaluated continuous NOX emission data from two plants using NSCR. The analysis for Dyno Nobel St. Helens shows a maximum 30 operating day emission rate of 0.21 lb NOX/ton acid. Also, we had monthly averages from JR Simplot that ranged from 0.15 lb NOX/ton acid to 0.36 lb NOX/ton acid. Although monthly averages are not directly comparable to continuous hourly NOX emission data, there is a strong probability that this source could comply with 0.50 lb NOX/ton acid. Therefore, the standard of 0.50 lb NOX/ton acid limit is achievable for at least some nitric acid production units using NSCR.
Also, the EPA notes that the commenter is referring to emissions levels that were achieved during RATA testing. These data are gathered over a short period of time under steady state operating conditions and are not directly comparable to long term CEMS data. Long term CEMS data are more representative of the emissions that occur during normal operation. 
 10. Costs
10a. Marginal Costs and Benefits
Comment 10.1:  Commenter 0103 asserts that EPA could have set a more stringent revised standard. According to the EPA's analysis, the commenter notes that the change in the performance standard will result in no additional costs for nitric acid producers, new monitoring requirements will cost an average of forty‐five dollars per ton, but the revised performance standard, may result in little to no emission reductions "because the majority of control systems installed on future affected facilities would likely result in emissions at or below the proposed emissions limit even in the absence of these proposed revisions." 
Considering only average costs of reduced emissions, the commenter notes that EPA set a revised standard without taking into account whether further net benefits might be achieved by a more stringent rule. Given EPA discretion, statutory instructions, and executive orders, the commenter recommends that the agency use a cost‐benefit analysis to develop its nitric acid performance standards. According to the commenter, regulation should maximize social welfare, and cost‐benefit analysis is the best tool that agencies can use to achieve that end.
The commenter notes that any revisions to the nitric acid plant NSPS can and should be based on a cost‐benefit analysis that takes into account the costs of implementing the newest and best technologies for NOX emissions as well as the benefits of continued reduction and regulation of NOX pollutants. Specifically, the commenter asks EPA to set the performance standard at a rate for which the marginal cost of increasing pollution control equals the marginal benefit of increasing control.
Comment 10.2:  Commenter 0103 states that EPA assessed average costs of emissions reductions, but it failed to consider marginal costs. When estimating costs, the commenter feels that it is important for the EPA to accurately assess the baseline (i.e., costs under existing regulations) and to account for the potential for technological growth. Since the EPA originally issued the 1971 standards, the commenter notes that research has developed many newer, cheaper, and more effective means to reduce NOX emissions. 
According to the commenter, recent industry information indicates that nitric acid plants are fully capable of reducing their emissions to 95 to 98 percent below the uncontrolled facility emissions level. The EPA's analysis considered these current practices when determining the burden on the industry of using newer technologies to achieve emissions reductions, but the commenter notes that the EPA failed to consider the possibility that plants could meet the revised standard with existing technology and that further reductions may be possible. According to the commenter, this history illustrates the potential for industry to adapt to regulatory requirements by finding cheaper, more effective ways to comply; in other words, technology can bring down compliance costs over time. 
Comment 10.3:  Commenter 0103 suggests that the NOX Budget Trading Program could assist the EPA in setting an emissions standard for nitric acid plants that is cost‐benefit justified. Under a market system like the NOX Budget Trading Program, if the emissions budget is set efficiently, the commenter notes that the permit price will equal the marginal cost of abatement, which should also equal the marginal benefits of reducing emissions. According to the commenter, the benefit of reducing NOX emissions from sources within the trading program is roughly $592 per ton. Using this figure as a benchmark, the commenter believes that the EPA can approximate the benefits of reducing NOX emissions from nitric acid plants, and should design its nitric acid performance standards so that marginal costs equal marginal benefits. The commenter notes that the EPA should set the emissions standard such that nitric acid plants spend the necessary amount to comply with the regulation, up to the cost of the NOX Budget Trading Program permit price. In the Proposed Rule, the commenter notes that the EPA estimates only average costs, ignoring both marginal costs and marginal benefits of the proposed revision. Before issuing a final rule, the commenter believes that the EPA should also estimate the marginal costs of further reduction and compare those costs to the marginal benefits of further reduction; it should then set the revised performance standard at the level where marginal benefits equal marginal costs.
Response to Comments 10.1-10.3: The EPA has not performed an analysis of monetized benefits of NOX reductions in the context of this rulemaking. Such analysis is typically performed for rules with larger economic impacts than Subpart Ga. However, we note that the cost effectiveness for NOX reductions achieved through this final rule are reasonable and supported by previous NSPS for other source categories. See Section VI of the final rule preamble for more discussion of the economic impacts of this final rule.
10b. Costing Errors
Comment 10.4:  Commenter (0095) states that EPA's capital cost estimate of $39,000 for installation of a continuous emissions rate monitoring system is very low, based on the commenter's experience in procuring and installing flow rate monitors, especially on existing units. According to the commenter, it is always a challenge to install new equipment on existing systems, due to space constraints, alignment with data acquisition system, and having to calculate continuous emissions based on existing and new measurement devices.
Comment 10.5:   Commenter (0098) suggests that if existing NAPs are forced to purchase, install, operate, and maintain a second NOX monitor, or a dual span monitor in order to accurately capture emissions at all times (including SSM events), it comes at a significant cost. The commenter notes that such costs have not been considered by EPA.
Comment 10.6:  Commenter (0098) states that the commenter's members and their consultants have found that contrary to EPA's $15,000 estimate, a gas flow monitor and its attendant equipment and installation costs between $25,000 and $35,000. See Sample Price Quote at Enclosure A to the commenter's letter.
Comment 10.7:   Commenter (0098) states that EPA must take "into account the cost of achieving [the] reduction" mandated by the new NSPS. According to the commenter, EPA must consider all costs of complying with the new NSPS and whether such compliance is unduly precluded by such costs, as well as whether such costs unduly preclude future development of plants in the regulated source category, and any impacts the cost of compliance may have on the supply of the products produced by the regulated source category. The commenter states that EPA has failed to meet this statutory mandate and must begin its proposal and evaluation thereof anew with accurate cost estimates.
Comment 10.8:   Commenter (0098) states that EPA assumed that industry would choose not to modify or reconstruct existing NAPS and would construct new plants instead. According to the commenter, this error resulted in EPA limiting its consideration of costs to the six new NAPs it estimated would be constructed from 2011 to 2016, following the promulgation of the new NSPS. The commenter states that at no point in the preamble to the Nitric Acid NSPS Proposal or in the supporting materials in the docket did EPA collect, evaluate, or consider the cost of compliance on existing sources that subsequently are modified or reconstructed in a manner that constitutes a "modification" under EPA's NSPS Subpart A regulations, and thereby subjects those sources to the proposed NSPS under Subpart Ga.
The commenter offers the following examples of these omissions below: 
(1) The cost for a NAP with NSCR to make changes to its NSCR or to add to its train an SCR in order to comply (and any lost production related thereto); [The commenter also notes that existing NAPs likely will have space constraints for the installation of new control devices. Further, the existing temperature of the gas stream may limit where an SCR can be installed into the process. Moreover, if, based on space issues or where an SCR can be installed, the gas stream has to be heated or cooled prior to being introduced into the SCR, this will create additional costs and emissions.]
(2) The cost for a NAP to install an extended absorption tower(s) in an attempt to increase efficiency and thereby reduce emissions (and any lost production related thereto);
(3) The cost to a producer to abandon a NAP that cannot support the costs of compliance (as well as the impact of the loss of such facilities on the nitric acid market and the economy generally); and,
(4) The initial and long-term costs of additional labor and raw materials related to the installation and operation of larger, better, and/or additional control devices.
The commenter believes that EPA's failure to consider the costs related to existing or modified NAPs' compliance with the proposed NSPS also may impact its analysis regarding new NAPs constructed in the future. According to the commenter, this is because EPA based its conclusion that six new NAPs would be constructed between 2011 and 2016 on the market growth estimates based on past annual increases (in tons) of nitric acid production. The growth estimates were calculated in 2006 (based on estimates for the Greenhouse Gas Reporting Rule), and, it appears to the commenter that they were not calculated with any consideration of potentially negative impact the proposed NSPS for NAPs might have on the production. [The commenter notes that EPA appeared to use 2006 estimates instead of 2010 estimates to reflect actual production.] This in turn, according to the commenter, would impact EPA's projections regarding future construction of new NAPs, and any and all costs related thereto.
The commenter asserts that EPA must assess the costs of compliance related to existing NAPs that are modified or reconstructed and become subject to the proposed NSPS in Subpart Ga. The commenter is concerned that the expenditures and other non-monetary costs are not justifiable in light of the infinitesimal decrease in overall NOX emission gained by implementing the proposed limit.
Comment 10.9:   Commenter (0098) finds EPA's rationale regarding costs related to the control used in newly constructed NAPs is flawed, and EPA failed to properly consider monitoring equipment costs.
According to the commenter, in evaluating costs related to the projected six new NAPs to be constructed between 2011 and 2016, EPA projected costs as the incremental difference between complying with the current NSPS requirements of Subpart G, and complying with the proposed new NSPS. The commenter notes that EPA concluded that because there are many NAPs currently meeting the proposed emission limit of 0.50 lb NOX/tHNO3, there is no incremental increase in control costs for newly constructed NAPs. Thus, the only incremental costs to be incurred by the entire nitric acid industry to comply with the proposed NSPS, EPA concluded, are those related to the installation of six air flow monitors.
According to the commenter, EPA's decision to calculate the incremental costs of compliance from the current Subpart G NSPS to the proposed Subpart Ga NSPS, as opposed to calculating the actual costs of compliance with the proposed NSPS, makes little sense in light of EPA's (wrongful) conclusion that no existing sources would be modified or reconstructed (and thus never subject to the proposed Subpart Ga NSPS), and that only six newly constructed NAPS would be subject to the proposed Subpart Ga NSPS. Thus, the commenter believes that EPA calculated costs for a scenario they assumed would never exist.
As EPA has long recognized, the commenter notes that NAPs are not "one-size fits all operations." Instead, the commenter notes that they are finely tuned and designed based on desired production and pollution control requirements. The commenter believes that it is flawed to conclude in a vacuum that merely because some NAPs have achieved the proposed standard in the past that there will be no incremental cost for newly constructed sources to do so, particularly given EPA's estimates regarding the production capabilities of EPA's estimated six new plants. Moreover, according to the commenter, the fact that existing NAPs utilize a particular pollution control (e.g., SCR) does not mean that the installation and operation of the same type of control for a newly constructed NAP is not a cost of compliance for the new NAP.
The commenter notes that EPA concluded that there was no additional cost related to the installation of a NOX CEMS because the requirement of a NOX CEMS is included in the current NSPS. As above, the commenter asserts that the mere fact that a prior source installed a NOX CEMS does not remove the full cost of the installation and operation of this device as a cost of compliance for a newly constructed NAP.
Response to Comments 10.4-10.9: The EPA agrees with the commenters that dual span monitors should have been included in the cost analyses. See discussion below.
The EPA disagrees with the commenter's assertion that EPA did not include incremental increases in costs because there are many NAPs currently meeting the proposed emission limit of 0.50 lb NOX/tHNO3. The reason that incremental NOX CEMS installation costs for new plants were not included at proposal is because this equipment and the accompanying costs were already included in subpart G. New or modified facilities would incur the costs associated with use of a NOX CEMS regardless of whether subpart Ga existed or not.
The EPA also disagrees that the costs for the flow rate monitor are incorrect. Commenter 0098 supplied cost estimates for purchase and installation of flow monitors in the range of $25,000 to $35,000. Purchase and installation costs are typically considered capital costs. The EPA estimated the flow meters costs as $39,000 for capital costs and $15,000 for annualized costs per plant using the EPA CEMS Cost Model, available at http://www.epa.gov/ttn/emc/cem.html. These costs are higher than the commenter's estimated costs. Therefore, costs for flow rate monitors have not been changed since proposal. 
However, the proposed costs have been changed to account for one modified source and five new sources over the next five year period instead of six new sources over the same time period. Revisions include the addition of costs for installation and operation of an SCR system for the modified source ($2.7 million capital and $370,000 annualized), costs related to Appendix F to account for annual stack testing for all six sources ($72,000 annualized), and costs associated with the purchase and installation of a dual span NOX concentration monitor for all six sources ($138,000 capital and $53,400 annualized).
At proposal, the capital costs and annualized costs for Subpart Ga were $234,000, and $90,000, respectively. The cost effectiveness was estimated at $45 per ton of NOX. 
With the additions discussed above, the resulting total industry-wide final capital costs and annualized costs of $3.07 million and $585,000, respectively, remain reasonable and result in an overall cost effectiveness of approximately $300 per ton of NOX.
Comment 10.10:  Commenter (0098) states that EPA's projected costs fail to accurately assess the cost impact of its proposal to require a continuous emissions rate monitoring system (CERMS). According to the commenter, EPA's cost estimates are based on its assumption that each newly constructed NAP would be required to install an air flow meter only and that existing and installed NOX CEMS would not require modification. According to the commenter, this assumption appears to have been driven by EPA's calculation of incremental costs instead of actual costs. Of course, the commenter notes that such an assumption makes little sense for a newly constructed facility, which would need to install both the NOX CEMS and the air flow monitor. Nevertheless, even assuming the six future NAPs already had a NOX CEMS installed, the commenter notes that EPA's assumptions remain flawed.
According to the commenter, the majority of existing NAPs use NOX CEMS consisting of one of three analyzer technologies: (1) Chemiluminsescence analyzers; (2) Non-dispersive infrared analyzers; or, (3) Fourier Transform infrared analyzers. For the majority of NAPs in existence, the commenter notes that these analyzers are equipped with only a single span range (which monitor in the 0-500 ppm NOX range, in compliance with Subpart G). In order to capture SSM events, as required by the proposed NSPS, the commenter notes that the NOX analyzers will require replacement with dual range analyzers because SSM events result in emissions far greater than 500 ppm (producers find result in emissions from 2,000 to 5,000 ppm). Further, the new analyzers will be equipped with logic that will automatically switch the span range to the higher value so that SSM events (elevated NOX readings) are captured. According to the commenter, the purchase, installation, and maintenance of dual range CEMS are not included in EPA's cost considerations. The commenter's members and their consultants have found that the cost to switch a single span range analyzer to a dual span range analyzer is between $18,000 and 28,000.
In the more likely scenario, to perform the monitoring EPA seeks, the commenter believes that a new NAP would be required to purchase and install a comprehensive and integrated CERMS. Commenter members and their consultants have found that installed CERMS that would meet EPA's monitoring requirements for the proposed NSPS range between $140,000 and $300,000. [See the Attachment A to the commenter's letter.] Although not definitively determined, commenter members have estimated that ongoing O&M for a CERMS is approximately $100,000 per year, per NAP.
Comment 10.11:  Commenter (0097) states that adding the requirement for continuous flow meters also is unnecessarily costly for subcategorized small size nitric acid plants and the unnecessary costs to install and maintain them are unjustified.
   Response to Comments 10.10-10.11: We acknowledge your comments related to cost of CERMS. Much of the costs for CERMS supplied by the commenters related to NOX concentration CEMS (stack probe, data acquisition and handling systems, etc) are not associated with these NSPS revisions. The EPA has made some changes to the cost analysis. See memo entitled Impacts of Nitric Acid NSPS Review -- NOX (Updated Memo for Final NSPS). See also Response to Comments 10.4-10.9. Also as discussed in Section 7, no subcategorization in this source category is warranted.
   
 11. General Supportive Comments
Comment 11.1:   Commenters (0092, 0093, 0095, 0096, 0098) agree with EPA's assertion that an opacity limit is not necessary for facilities. Commenter 0096 states that EPA's decision is appropriate since the facility is required to demonstrate compliance with NOX using a Continuous Emission Monitor. According to the commenter, the opacity requirement is not required as an additional method of demonstrating compliance with a NOX emission limit.
Response to Comment 11.1:  We acknowledge the support for EPA's decision to remove the opacity limit from subpart Ga.
Comment 11.2:   Commenters (0092, 0093, 0095, 0098) support EPA's inclusion of EPA Reference Method 7E of Appendix A-4 as an additional method available to sources for evaluating performance because it provides affected sources with flexibility in selecting the best performance test for its purposes. 
Commenter (0098) states many nitric acid plants have relied upon Method 7E in light of a March 2000 EPA determination. Although the commenter does not agree with EPA that Method 7E must include modifications to address moisture and a NO/NO2 split, the commenter nonetheless supports recognition in Subpart Ga of the use of a modified Method 7E.
Response to Comment 11.2:  EPA acknowledges the support for EPA's decision to allow the use of Method 7E in subpart Ga. 
   12. Wording and Typographic Changes
Comment 12.1:  Commenter (0105) questioned the placement of proposed method ASTM E1584-00(2005)e1 in §60.17(a)(93), stating that in the 3/21/2011 Federal Register, this paragraph was assigned to ASTM D6784‐02 (Reapproved 2008) for NSPS Subparts CCCC, DDDD, LLLL, and MMMM. EPA has proposed 24 new ASTM methods in other proposed amendments {10/9/2008  -  73 FR 59956, 12/28/2008  -  73 FR 78522, 5/3/2011  -  76 FR 24976}. None of them are ASTM E1584‐00 (2005).
Response to Comment 12.1:  The incorporation by reference is not official until the standards are final. When this rule is finalized, §60.17(a) will reflect the existence of ASTM E1584-11. Also note that the standard was revised in 2011 so the ASTM number has changed slightly from proposal.
    Comment 12.2:  Commenter (0098) notes that the proposed definition of "operating day" in the proposed §60.71a(c) is missing the word "operated": a 24-hour period beginning at 12:00 a.m. during which the nitric acid production unit [operated] at any time during this period.
   Response to Comment 12.2:  We agree with your recommendation. This change has been made to the final standards at Subpart Ga.

Comment 12.3:  Commenter (0105) submitted suggested changes to subpart Ga:
         Section numbers and headings should be in bold. Some of the Section numbers are not underlined with their heading. (May be done when signed document is converted to the Federal Register format.)
         §60.71a  -  remove the lettering "(a)", "(b)", "(c)", and "(d)" in front of the paragraphs of definitions. The current procedures for the definition section of the NSPS subparts are to the words alphabetically, not assigned a specific paragraph letter. The older NSPS subparts that have not been modified recently (≈ last 10 years) still have the lettering, and when their definition sections are modified the paragraph letter is removed and then sorted alphabetically. 
         §60.71a  -  the words being defined should be in italics. (May be done when signed document is converted to the Federal Register format.)
         §60.73a  -  Some of the major parts {(a), (b), (c), and (f)} of this section have individual headings which are underlined. The heading for (d) should also be underline. A heading should be added to (e) to be consistent for this section. {In other subparts, these headings are in italics.}
         §60.73a(d)(1) in the last sentence, change "lb NOX /ton" to "lb NOX/ton". To be consistent, the "x" should be capitalized. {Type NOX, then subscript the "X".}
         §60.74a(b)  -  Can the facility send an email notifying the Administrator of the emission exceedance? Most government agencies include emails as part of any FOIA review, and therefore have procedures for handling email submittals. Communications between government staff and facilities are more likely to be by email than faxes these days (don't remove fax from notification option). 
         §60.75a(a)(4) in Equation 2 definition for Md, change "lb NOX" to "lb NOX.".
         §60.76a(f)(2), suggest change "section 60.72a(b)," to "§60.72a(b)," to be consistent with similar language in the subpart.
         §60.77a(b)(3), suggest the following:
         change: (3) Reasons for noncompliance with the emissions standard; and description of corrective actions taken.
         to: (3) Reasons for noncompliance with the emissions standard; and (4) Description of corrective actions taken.
         §60.77a(f)(2), change "with 60.72a(b)," to "with §60.72a(b),"
Response to Comment 12.3:  EPA agrees with many of these recommendations. These changes have been made to Subpart Ga where appropriate. Most of the suggested changes occur when the document is converted to Federal Register format.
The agency understands and appreciates the commenter's suggestion regarding use of electronic mail, but at this time the rule has not been changed to allow it for notification purposes. As the agency works through security and routing issues, the rule may be revised in the future to allow for electronic mail notification.

