American
Iron and Steel Institute
25 Massachusetts Avenue, NW Suite 800 Washington, DC 20001 Phone-____ BC6____ i Fax 202.452-1039 E-maii tgibson@steel.org
www.steei.org
President and CEO
May 15, 2017
Ms. Samantha K. Dravis Office of Policy U.S. Environmental Protection Agency 1200 Pennsylvania Avenue, NW Washington, DC 20460
Re: Response to Environmental Protection Agency's Request for Comment on the Evaluation of Existing Regulations (Docket ID No. EPA-HQ-OA-20170190)
Dear Ms. Dravis:
The American Iron and Steel Institute (AISI), on behalf of its U.S. producer member companies, appreciates this opportunity to comment on the Environmental Protection Agency's (EPA) Request for Comment on the Evaluation of Existing Regulations (82 Fed. Reg. 17793)(April 13, 2017). AISI serves as the voice of the North American steel industry in the public policy arena and advances the case for steel in the marketplace as the preferred material of choice. AISI also plays a lead role in the development and application of new steels and steelmaking technology. AISI is comprised of 18 member companies, including integrated and electric furnace steelmakers, and approximately 120 associate members who are suppliers to or customers of the steel industry. AISI's member companies are classified in the North American Industry Classification System (NAICS) in codes 3311 (Iron and steel mills and ferroalloy manufacturing) and 3312 (Steel product manufacturing from purchased steel). Additionally, member companies that mine iron ore are included in code 2122 (metal ore mining).
We reviewed the request for comment and have gathered information from our member companies detailing the impact of existing federal regulatory requirements on their ability to operate, construct, expand or modify facilities in the U.S. The detailed comments below provide:
1. Background on the number and types of permits our members need to obtain to construct, operate and/or expand facilities in the U.S.;
2. Proposed changes to existing permitting regulations that would help streamline the process and reduce overall costs for companies; and
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00001
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 2
3. A broader list of regulations outside of the federal permitting structure that our members consider an added and unnecessary burden to domestic manufacturing.
1. Manufacturing Permitting Process
The construction, operation and expansion of a steel, coke and/or mining operation requires various permits under different existing federal statutes regulating air emissions, discharges to water and handling and processing of waste products. The total permits needed to construct and operate a facility will vary depending on the particular operation but could be as high as 12 or more permits for a single facility across various state and federal agencies. As will be outlined further in our comments, the high costs and timing uncertainties in the numerous permit application processes create significant impediments and often frustrate future planning of new or expanded facilities in the U.S.
Under the Clean Air Act (CAA) before a facility is constructed or expanded the company is required to go through the New Source Review (NSR)1permitting process that ensures that facility will employ up-to-date pollution control technology.
Facilities opening or expanding in an area of the country considered in attainment with current National Ambient Air Quality Standards (NAAQS) must obtain a pre- construction permit under the CAA's prevention of significant deterioration (PSD) provisions. The PSD permitting process requires a showing that the facility is using the best achievable control technology (BACT) to ensure projected emissions increases will not cause an area to exceed NAAQS for any of the listed criteria pollutants. The PSD process is generally applied for and issued through state environmental agencies with oversight and approval from U.S. EPA Regional offices. A PSD permit may be required for various processes or emission units at a single facility. AISI members report that the PSD permitting approval process can take from a little over a year to up to two years in some EPA Regions despite the CAA
1 NSR permitting technically refers to permitting of facilities in CAA nonattainment areas, and PSD refers to permitting of facilities in CAA attainment areas. Generally, the term NSR is often used broadly to refer to the permitting program, including both NSR and PSD permitting. These terms are used throughout this document according to that standard approach.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00002
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 3
requirement that completed permit applications be granted or denied not later than one year after the date of filing.2
In addition to PSD permitting, facilities must also obtain a CAA Title V Operating Permit through its relevant state agency. The Title V permit includes the pollution control requirements from federal or state regulations that apply to a source and is subject to reviews by U.S. EPA, state environmental agency, federal land managers and public comment and hearings. Similar to the PSD permitting process, AISI members report varied timelines for completing the Title V review and approval process depending on the state regulatory agency and EPA Regional office, taking up to three years to receive the final permit and costs of several millions of dollars for each operating permit needed. Title V permits also must be renewed every five years, during which time a facility needs to demonstrate compliance with the permit conditions by reporting its air emissions over the five year period.
The Clean Water Act (CWA) administered by both U.S. EPA and U.S. Army Corps of Engineers (Army Corps) contains various permitting requirements that most industrial facilities must meet. CWA Sections 401 and 404 deal with certification and permitting for constructing, operating or expanding an industrial facility where federal waters will be dredged or filled. Most Section 401 water quality certifications are handled by the states and can be used to condition or deny Section 404 permits issued by the Army Corps. The Section 401 certification and Section 404 permit process can take between a year to up to four years to complete. Discharges of either industrial process waters or storm waters into federal jurisdictional water requires National Pollutant Discharge Elimination System (NPDES) permits. These permits are issued by state environmental agencies with oversight from EPA Regional offices. Members report that NPDES permits can take six months or more to obtain depending on the state or region.
2. Proposed Changes to Streamline Existing Permitting Regulations
Once a facility is constructed and in operation, some of the most onerous permitting processes that facilities have to go through to continue to function as well as to seek any efficiency improvements or expansions are those relating to obtaining the NSR/PSD construction permits described above.3 The loss of productivity and
2 42 U.S.C. 7475(c). 3 See also 40 C.F.R. Parts 51 and 52.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00003
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 4
costly, time consuming permitting efforts associated with the current NSR process are without any real benefit to the environment and have the unintended consequence of actually discouraging installment of new efficiencies.
EPA should seek to streamline NSR permitting for modifications conducted at existing steel facilities to facilitate and stimulate productivity increases, innovation and efficiency in domestic steel manufacturing. This effort is necessary because the current NSR procedures have the unintended consequences of often forcing decreases in production. EPA proposed many of the reforms advocated in 2002, 2003, 2004 and 2005. Although some of these reforms were stayed as a result of court rulings, AISI believes that EPA can still implement the reforms through a careful review of the provisions and could yield beneficial NSR reforms.4 AISI stands ready to help EPA engage in this process.
a. Routine Maintenance Repair and Replacement Reforms
The NSR rules provide that certain physical and operational changes at a plant do not constitute "major modifications" and thus do not need to undergo the burdensome NSR process.5 One of the categories of exempt activities are those considered "Routine maintenance, repair and replacement" (RMRR).6 EPA applies the RMRR exclusion primarily on a case-by-case basis, using a multi-factor test for determining whether a particular change falls within or outside the exclusion. This case-by-case analysis involves several different time consuming, expensive and legally risky options a company must weigh to potentially qualify for the RMRR exemption: (i) in advance of a facility submitting a construction permit the company can request a formal applicability determination (which generally takes a minimum of three months); (ii) an RMRR argument for exemption can be made as part of the construction permitting process, which also can take many months; (iii) a facility can proceed "at risk" without prior regulatory approval and with potentially serious enforcement consequences; or (iv) a facility can forego installing state-of-the-art equipment or undertaking efficiency projects and continue operations with older, less efficient equipment.
4 See New York v. EPA, 443 F.3d 880 (D.C. Circuit 2006). 5 40 C.F.R. 51.165(a). 6 Id.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00004
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 5
EPA previously recognized the need to reform the RMRR determination process due to inherent uncertainty, costs and legal risks imposed on domestic facilities, stating that the process:
" .. .results in lost capacity and lost opportunities to improve energy efficiency and reduce air pollution."
And also requires permitting authorities:
"...to devote scarce resources to make complex determinations, including applicability determinations, and consult with other agencies to ensure that any determinations are consistent with determinations made for similar circumstances in other jurisdictions."
EPA goes on to say of the currently-existing NSR process:
"...the effect is to discourage plant owners or operators from engaging in replacements that are important to restoring, maintaining and improving plant safety, reliability and efficiency...this effect is exacerbated by...the uncertainties inherent in the case-by case approach."7
These same deficient permitting processes exist today and present the same challenges to domestic manufacturing that existed more than a decade ago. EPA should revisit certain portions of the RMRR reforms from 2003, specifically the provisions known as the "Equipment Replacement Provisions" (ERP). These provisions would exclude from the definition of "major modification" at 40 C.F.R. Section 51.165, projects that replace existing, permitted equipment or components thereof with like-kind or functionally equivalent equipment, provided the replacement components or equipment do not alter the basic design of the process unit and would not result in an exceedance of the permit limits.
Additionally, EPA could identify specific activities at steel facilities and other industrial sector operations that should be presumptively considered to constitute RMRR. Such a list could be added to the regulatory language or could be issued as clarifying guidance. As an initial example, the following list of modifications common to the steel industry could be considered presumptively as RMRR rather than be
7 69 Fed Reg. 61250 (October 27, 2003).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00005
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 6
subjected to the current costly case-by-case analysis: replacement of furnace shells, oxy-fuel burners, oxygen lances, tundish pre-heaters, tundish burners, transformers, capacitors, regulators, lime or carbon injection equipment and other equipment that is commonly and routinely replaced in the industry as part of normal, continual maintenance activities. AISI and its members welcome any opportunity to discuss these recommended changes to the RMRR process with the administration or answer any questions.
b. Exclude Pollution Control Projects from Modification Definition
Another potential regulatory change to the existing definition of "major modifications" triggering NSR permit requirements that could streamline the permitting process for existing manufacturing facilities is to exclude Pollution Control Projects ("PCPs") that result in net overall reductions of air pollutants, involve upgraded or rebuilt pollution control equipment or devices, or allow raw material/fuel substitutions that do not result in exceedances of permit limits. Similar to the RMRR revisions described above, EPA explored excluding PCPs from NSR review in the past on the basis that the exclusion "allows sources to install emissions controls that are known to be environmentally beneficial...thus offer flexibility while improving air quality."8 The prior EPA effort listed specific technologies that the agency believed should be presumptively considered PCPs, but further provided that PCPs not listed could nevertheless qualify for the exclusion if the permitting authority determined on a case-by-case basis if the non-listed PCP was environmentally beneficial, and provided that PCPs entitled to the exemption could include in addition to equipment and technology, work practices, process changes or other pollution prevention strategies. AISI again recommends EPA revisit the PCP exclusion and publish in either regulatory text or through guidance a list of technologies, work practices and/or strategies employed in the steel sector that would be presumptively considered PCPs and exempt from NSR review.
c. Match the NSR Definition of Major Modification with the NSPS
EPA should revise the NSR emissions test definition to match the NSPS definition. This change would provide greater opportunities for facilities to make efficiency improvements to their operations without the significant costs and permit approval delays that are currently associated with the NSR trigger. EPA proposed a
8 67 Fed. Reg. 80190 (December 31, 2002).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00006
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 7
similar change in definitions which, had it been finalized, would have applied only to electric generating units (EGUs).9
AISI believes that that statutory language behind the NSR program fully allows for this change, and the benefits of such a change would significantly aid streamlining the permitting process currently faced by our members' facilities.
d. Adopt Plant-Wide Applicability Determinations
EPA should promote and facilitate a voluntary option that a facility can establish Plant-Wide Applicability Limitations ("PALs") (basically emissions limits that apply facility-wide) through a permitting process, allowing such facility to change, modify and upgrade equipment and operations and add new equipment without triggering major modification NSR review, provided the changes do not result in exceeding the established PAL emissions limits.10 When EPA explored establishing PALs in the past, the agency stated "[w]e believe that the added flexibility provided under a PAL will facilitate your ability to respond rapidly to changing market conditions while enhancing the environmental protection afforded under the program."11 AISI believes EPA's conclusion remains true today, however, these provisions were stayed in litigation. EPA could revisit this issue and work with industry to minimize the burdens of establishing PALs, so that steel plants and other domestic manufacturing facilities would be able to make needed changes to equipment and processes more quickly than under the existing burdensome NSR process. This would allow steelmaking facilities and other manufacturers to have flexibility to respond to market conditions and make the U.S. manufacturing industry more competitive in the global marketplace. With a PAL, steel facilities would, without increasing emissions above the levels already deemed acceptable by the permitting authority in the PAL permit, be able to invest in equipment and technologies without the risks and delays inherent in the existing NSR permit process.
e. Specific NSR Permit Issues
9 See 70 Fed. Reg. 61081 (October 20, 2005). 10 See 40 C.F.R. 51.165(a)(l)(v). 11 67 Fed. Reg. 80189 (December 31, 2002)
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00007
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 8
There are a number of additional important NSR permit related matters that would benefit greatly from increased clarity and flexibility:
EPA should reconsider policy reasons for co-located sources being considered the responsibility of the host company or facility. A co-located facility is on leased ground and has its own operating permit, employees, equipment and is responsible for maintaining compliance. EPA maintains the position that the "host" company has control and responsibility for any co-located company's actions, which is not the case. Separately permitted entities should not be a part of permitting evaluations for each other.
EPA should provide for better flexibility for emissions offsets from energy conservation and reuse projects. For instance, over-restrictive netting requirements force many limits on a new boiler because the rules do not allow for direct emission offsets without additional federally enforceable limits to be set, like a minimum blast furnace gas (BFG) limit or a boiler specific emission limit. While a natural gas limit would make sense for a new, efficient boiler, the other requirements do not when looking at the overall reduction in emissions.
EPA policies that reversed in-project netting should be established to give needed flexibility for modifications so that a permittee does not have to pull in every source (including on-site contractors) for netting calculations.
EPA should modify rules related to data collection to allow for more current techniques. Certain rules list prescribed means of data collection that are now archaic and not easily maintained as they are obsolete (e.g., paper chart strip recorders). Likewise, rules related to equipment calibration should be modified. These rules are outdated in that they do not recognize the newer generations of electronic instruments that cannot be calibrated in a traditional sense and either work, fail or internally diagnose and alert user of required attention. Many units also selfcalibrate. Facilities now create a "calibration" to satisfy the rule requirement but the work is just a paper trail with no real meaning or added value for instrument performance.
EPA should address PSD projects by allowing the permittee to revisit and/or take on new or additional limits when transitioning from the source modification/construction permit to the operating permit before closing the PSD time-period if the project lasts more than some minimum time threshold (e.g., 18 months or two years from the initial application to the facility start-up). Facilities should have the right to true-up or make
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00008
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 9
adjustments with the netting balances prior to operating the new equipment in case of previously unforeseen changes in scope, so long as the permittee does not change the significance determination of the project in the Technical Support Document (TSD).
f. Align Implementation Guidance with NAAQS Revisions
Under current EPA policy, when NAAQS are revised for any of the criteria pollutants, existing NSR permit applicants must often redo the required modeling analysis using the new standard. This is often difficult and costly for both the affected facility and the state permitting authority as EPA has increasingly revised NAAQS before providing the states with implementation or modeling guidance. This lag time between revised standards and the implementing guidance, which often spans many years, can cause states to delay PSD permit approvals for facilities until EPA provides the needed data. The simple fix is for EPA to ensure it has the necessary implementation and modeling guidance in place before it issues a revised standard or at the very least have guidance on track to be available before a revised standard comes into effect.
g. Grandfather Existing NSR Permits When Adopting New NAAQS
EPA should adopt regulations requiring that NSR permit applications be subject to the NAAQS standards that exist on the date the application is submitted to EPA. Manufacturing facilities that have triggered NSR review spend significant resources modeling emissions to demonstrate that the NAAQS will not be exceeded when BACT controls are employed. If any of the NAAQS are changed while the permit is being processed EPA often requires new modeling based on the new standard. A consistent approach that grandfathers existing applications in connection with any future NAAQS revisions would provide companies with certainty and ensure companies do not spend time and resources on NSR permitting only to have the goal posts moved at the last minute and find themselves back at square one. This approach could also help alleviate EPA resources from having to develop the detailed guidance states need to implement a revised standard in a rushed timeframe.
h. Single-Source Photochemical Grid Modeling
The recent final rule amending Appendix W requires that, in the absence of an analysis allowing a source to screen out of a requirement to conduct and report the
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00009
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 10
results of air quality modeling, applicants seeking PSD permits must model impacts of precursors to ozone and PM2.5 using a "chemical transport model" such as a photochemical grid model.12 EPA does not, however, specify a preferred model for this purpose. Although the CAA requires applicants for PSD permits to model air quality, EPA has long recognized that "it was not technically sound to designate with particularity specific models to be used to assess the impacts of a single source of ozone" or PM2.5 and instead allowed a permit applicant, working with the permitting authority, to choose a "method" to conduct the required air quality analysis. EPA has evaluated whether it is feasible to designate a single model for ozone and PM2.5 for use by all major sources applying for a PSD permit and concluded that it could not make such designations. EPA should, therefore, have retained the existing requirements applicable to sources emitting precursors of ozone and PM2.5. What EPA has done, however, is to effectively eliminate the exploring of other methods for analyzing the impact of a proposed source on air quality. EPA should return to its long-standing prior approach to air quality analysis to support PSD applications for sources that emit precursors to these pollutants. At a minimum, EPA should adopt a moratorium on single-source modeling of at least three years to further develop costeffective models and screening techniques. EPA can reevaluate the state of technology at the close of that period to determine whether the moratorium should continue.
i. Use of Probabilistic Modeling
Tighter margins between background pollutant concentrations and increasingly-stringent NAAQS make it more and more difficult to demonstrate compliance using overly-conservative modeling assumptions. Current EPA guidance requires "deterministic" air quality models that use a facility's maximum operating rate and maximum allowable emissions as opposed to realistic actual operating and emissions rates at a facility that are often well below the maximum allowables. This frequently results in modeling data substantially overstating the effects a new or modified facility will have on ambient air quality for the region. EPA should adopt more probabilistic approaches to modeling to address this conservatism, especially given ever stricter NAAQS approaching background levels that makes it difficult to attain permits for expansions, construction or efficiency improvements. Probabilistic modeling, for example, would be allowed to take into account the variability of both
12 82 Fed. Reg. 5182 (January 17, 2017) (to be codified at 40 C.F.R. Part 51, Appendix W 5.3.2(c)).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00010
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 11
background air quality and emission rates for modeled sources. While numerous commenters suggested this approach to EPA during consideration of the final Appendix W rule, EPA declined to take such action. EPA is using such probabilistic modeling in other programs, so should be able to make it workable and reliable under NSR.
j. Air Emission Modeling Improvements Needed
EPA has recognized that problems exist with its preferred models, but designated fixes to those problems are considered "non-default BETA option." Use of "BETA" fixes requires Model Clearinghouse approval, which substantially limits the utility of a fix. While fixes may require some period of technical assessment, prolonging non-default BETA designation of proven fixes does not foster the balance between the economy and environment called for by the CAA's PSD provisions.
EPA should make improved modeling tools a higher priority, including through allocating more funding for model development. Furthermore, EPA should conduct rulemakings to update Appendix W more frequently. Before the recent revision released in late 2016, EPA had not revised Appendix W since 2005, even though EPA has revised NAAQS for nearly every criteria pollutant in that same period. At a minimum, EPA should review Appendix W when it reviews NAAQS. Conducting such revisions concurrently with identification of improvements to models would better achieve the PSD program's explicit statutory purpose of ensuring economic development while protecting air resources. Key modeling improvements include:
Receptor exclusions - strengthen EPA guidance to exclude receptor data from areas that the public does not have legal or physical access to;
PM fugitive emissions - use of a pre-processing step to account for the overestimate of fugitive emissions; and
Source Characterization - EPA should streamline the modeling protocol approval process. Case-specific refinements involving source characterization should not trigger a non-guideline modeling approach. Substantial objective evidence and alternative source characterization protocols have been submitted to EPA to support the use of intermediate steps (without changes to AERMOD) to improve the accuracy of air dispersion modeling demonstrations for unique, non-traditional air
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00011
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 12
emission sources (e.g., fugitive emissions, sources with large fugitive heat releases, unique stack configurations, etc.). EPA should authorize the use of source characterization techniques such as LIFTOFF, AERMOIST, AERLIFT, urban characterization, pre-processing step for fugitive PM, etc. Furthermore these "source characterization" techniques should be considered for routine application without a need for a non-guideline model approval if adequate documentation of the effects is provided.
EPA's AERMOD model is known to over-predict and controls should not be required based on modeling and sources should not be responsible for the cost of adding new ambient air monitors. AISI submitted a study to EPA in 2011 that documents AERMOD's over prediction of emissions by more than 10 times compared to monitored data. When modeling shows nonattainment, sources can add new monitors per the SO2 Data Requirements Rule instead of relying on modeling. EPA or sta tes, not industrial sources, have historically been responsible for the cost of new and existing monitors and this cost should not be shifted to facilities. Modeling results that show an overprediction compared to monitored data should be allowed to be adjusted or calibrated based on the monitored data, instead of having industry install expensive controls based on known inaccurate modeled results.
k. Base NSR BACT Reviews on Domestically Proven Technology
In addition to the specific NSR regulatory reforms identified above that provide clarification on when PSD permitting is required, AISI suggests other changes to EPA policy once the PSD review process is triggered. BACT controls and limits considered during the PSD analysis should be based only on existing proven technologies and limits clearly demonstrated as reducing emissions at facilities located in the U.S. The PSD BACT evaluation process, spelled out through EPA guidance, should not include unproven technologies employed in other countries that have not been demonstrated as commercially feasible or effective at controlling emission in the U.S. Requiring domestic facilities to conduct technology reviews and costly feasibility analyses of technologies utilized in countries that do not have the same rigorous air pollution control and permitting requirements, places unreasonable permitting demands and delays on the already lengthy U.S. permitting process.
For the same reasons as above, the control technologies or limits considered under PSD BACT should be required to have been previously performance tested at a domestic facility to show an environmental benefit from the existing control
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00012
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 13
technology. Facilities seeking construction permits should not be required to evaluate technologies that are not already proven in the U.S. to reduce emissions. Because the current PSD BACT processes are unduly burdensome and time consuming for permit applicants and the reviewing authority, these procedures should be changed to reduce the economic burdens and time delays associated with BACT technology review.
1. Discontinue Use of Permit Performance Tests in Enforcement Actions
EPA has a history of using its enforcement powers to mandate control technologies and emissions limits in consent decrees as injunctive relief or mitigation measures. This practice of case-by-case enforcement measures creates significant disparities in the treatment of manufacturing facilities across the nation. Most importantly the enforcement strategy forces a small subset of facilities to bear the burdens of evaluating control technologies that EPA should be promoting through notice-and-comment rulemaking, or the permitting process. The enforcemen t burden on the targeted facilities can be extremely expensive and disruptive and often forces production limits resulting in a significant economic disadvantage of a target mill or group of mills, compared to their competitors. It also creates an additional corresponding burden and stigma of legal defense and enforcement in the facilities public record that the surrounding community may not fully understand.
EPA should also not utilize results from a post-construction performance test in an enforcement case seeking penalties or injunctive relief, provided the EPAapproved and permitted BACT technology installed was submitted with good faith pre-operational estimates of emissions. It is unfair and unproductive for EPA to review and approve technology-forcing BACT controls in a construction permit, only to file an enforcement action against a facility when it has invested in and installed the BACT controls in the event the controls do not meet the emissions reductions the facility and EPA anticipated.
AISI suggests that rather than pursue an enforcement action for penalties in the event EPA-approved BACT controls do not meet expected reductions, EPA should pursue a policy of modifying the construction permit based on the results of the performance testing. For emissions limits to be established as part of a BACT process, EPA should adopt a "test-and-set" process that establishes limits post construction, rather than establishing emissions limits in advance of a performance test.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00013
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 14
m. Discontinue Use of Enforcement Actions to Amend NSPS
EPA could also adopt an enforcement policy to discontinue the current practice of requiring a steel or iron ore production facility through an enforcement action to demonstrate compliance with limits on fugitive emissions from a plant production area or control device when limits are not set forth in the New Source Performance Standard (NSPS) for that source. Unless EPA changes the NSPS standards to specifically include numeric limits on fugitive emissions, it should not use language in enforcement action consent decrees to set emissions limits that were not adopted in NSPS notice-and-comment rulemaking. Currently, the NSPS does not have specific particulate matter numerical fugitive emissions limits, but rather sets opacity limits to ensure fugitive emissions are properly controlled. EPA is currently involved in a series of enforcement actions against several steelmaking facilities in Region 5 and has stated publicly that it does not like the published NSPS standards, and thus, is using the enforcement process to require Region 5 facilities to measure fugitive emissions in a manner in lieu of, and far beyond the NSPS-established opacity standards. This selective enforcement, in a single EPA Region, creates significant disparity in the applicability of the NSPS across the country and harms the competiveness of manufacturers operating in that Region.
Similarly, EPA should not use injunctive relief or mitigation as part of an enforcement action to require any steelmaking or iron ore processing facility to use a monitoring technology other than the EPA-approved Method 9 set forth in the NSPS, to evaluate compliance with the opacity standards. Again, EPA Region 5 is currently engaged in an enforcement strategy against steelmaking facilities in the Region seeking to have those facilities to install processes and technologies that are not set forth in the NSPS standards. This puts facilities in the Region at a legal and economic disadvantage to operations in other locations.
If EPA desires to change the NSPS requirements applicable to controls or fugitive particulate emissions from steelmaking facilities, it should amend the NSPS through public notice-and-comment rulemaking, rather than by selective enforcement which creates disparities in the industry and sidesteps the established rulemaking processes.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00014
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 15
n. EPA Should Update the NSPS Opacity Standard
The opacity limit EPA utilizes under Subpart AAa, 40 C.F.R. Section 60.272a (a)(2) and (3), which mandates a 3 percent opacity limit for emissions from a control device as well as a 6 percent limit that applies to a steel plant's production area to which the control device is attached should be updated to 10 percent. The changes are appropriate because EPA concedes that both of these limits were somewhat arbitrary when established originally and because of the practical limitations of reading a 3 or 6 percent opacity using EPA-approved Method 9 methodology. The use of a continuous opacity monitor (COM) as allowed under 40 C.F.R. 60.273a(c) also has a high margin of error in relation to the 3 and 6 percent limits, thus creating errors and misreading of opacity, without any environmental benefit. The standards setting organization ASTM, International has also identified COMs to be accurate only to the level of 10 percent opacity, making a 3 or 6 percent limit functionally impossible to accurately measure facility opacity for compliance purposes.13
o. Classification of Baghouse Dust as Hazardous Waste
Dust collected from the emission control devices employed during the primary production of steel in electric arc furnaces is listed as hazardous waste under the Resource Conservation and Recovery Act (RCRA)(K061) and requires use of a RCRA- permitted recycling facility to recover valuable commodities from the dust, such as zinc. AISI requests that K061 baghouse dust be delisted as a hazardous waste. The de-listing of K061 would allow for new, potentially significant opportunities for recycling and reuse of this waste stream without negative impacts on the environment. Baghouse dust would still be subject to being classified as a regulated waste if it is determined by analysis to have hazardous waste characteristics. While steel facilities currently manage their baghouse dust utilizing several RCRA-permitted recycling processes, AISI believes that if the waste was de listed, additional recycling markets would quickly develop for this large-volume waste stream. The development of additional uses for and recycling strategies for non-hazardous baghouse dust will potentially eliminate significant annual costs and management burdens (labeling, training, placarding, storage limits, transportation challenges) associated with listed waste, without environmental detriment.
13 See, ASTM D 6216-98, Standard Practice for Opacity Monitor Manufacturers to Certify Conformance with Design and Performance Specifications, ASTM International, West Conshohocken, Pa., 1998.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00015
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 16
p. U.S. Army Corps of Engineers Wetland Permitting
The Army Corps is responsible for administration of Section 404 of the CWA, which covers permitting associated with impact to wetlands under federal jurisdiction. The 404 permitting process is currently one of the most ill-defined processes for a regulated party to understand and thus to predict permit timelines. This absence of a schedule is counter to a regulatory structure that fosters a balance between environmental protection and economic development.
Specifically, the Army Corps' Section 404 process should be improved to include a publicly defined sequence of required steps and timelines to achieve each of them. As an example:
The Section 106 consultation process obligates the Army Corps to consult federally recognized tribal entities during the permitting process. This has been a completely open-ended process where the Army Corps allows the consultation to occur indefinitely. The administration should impose three-month duration to allow consultation to occur, which is more than ample time for interested parties to understand and provide input on a potential wetland impact (most public comment processes default to a 3060 day window).
Similar consultation processes are required with the U.S. Fish and Wildlife Service, the state Historical Preservation Offices and U.S. EPA. All of these processes can and should have defined windows of duration to build a sense of business certainty of how long it will take to get through a permitting process.
Elimination of the open-ended processes would also result in reduction of other collateral effects of the current permitting process, which industry has experienced taking two-to-six years to issue a new permit, and in one case, over seven years to simply amend an existing permit. Those collateral effects include things like repeated Endangered Species Act (ESA) reviews. Because the Army Corps process currently has no mandated or even target timelines, other processes required as part of the Section 404 permit, such as review for potential concerns with endangered species are never truly complete in the Army Corps eyes. When it takes the Army Corps anywhere from two-to-seven years to issue a permit, and species are added to the ESA list (e.g., endangered, threatened), removed and sometimes re
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00016
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 17
added to the ESA list, the Army Corps sees an obligation to go back to the drawing board and re-initiate the ESA analysis. Providing permitting structure and defining the process timeline for the consultation and permit processes improves permitting certainty, avoids delays and expending unnecessary resources and ultimately the prospects for economic.
3. Broader Regulatory Programs Impacting Domestic Manufacturing that are in Need of Reform
In response to EPA's request for broader existing or proposed regulatory programs that are outside the scope of facility permitting, AISI provides the following list of programs of most concern to its members. Where appropriate we provide potential alternatives but also stand ready to work with the appropriate EPA offices in crafting common sense, data-driven solutions. AISI also wishes to express its support for broader regulatory review and revision efforts the administration has already begun on matters of import to manufacturing in general, but also the domestic steel industry. These efforts include: EPA's recent decision to withdraw and reconsider or rewrite the 2015 final Waters of the U.S. rule; EPA's actions to place the litigation over the Clean Power Plan in abeyance and consider remand and review of the final regulations for new and existing EGUs; and EPA's actions to hold the litigation over the 2015 Ozone NAAQS in abeyance while the agency considers the appropriate actions in review of the final 70 ppb standard.
a. CERCLA Section 108b Financial Assurance Requirement for Hardrock Mining
On December 1, 2016, EPA proposed the first framework for financial assurance regulations under the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) Section 108b to address risks from hazardous substances.14 In the same notice, EPA applied that framework to the hardrock mining sector, proposing specific standards for financial assurance on the industry sector. EPA is under a court deadline to promulgate a final rule by December 1, 2017. EPA's proposal calls for classes of facilities to establish and maintain evidence of financial responsibility consistent with the degree and duration of risk associated with the production, transportation, treatment, storage or disposal
14 82 Fed. Reg. 3388 (January 11, 2017); 40 C.F.R. Part 320.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00017
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 18
of hazardous substances. This rulemaking is of significant interest across the mining sector and beyond, as it would very likely require significant financial resources to be moved out of operation and instead set aside for the potential future risk of hazardous contamination.
According to a notice that EPA issued in 2009 on this issue, hardrock mining is defined as the extraction, beneficiation or processing of metals (e.g., copper, gold, iron, lead, magnesium, molybdenum, silver, uranium and zinc) and nonmetallic, non-fuel minerals (e.g., asbestos, phosphate rock and sulfur). At that time, EPA decided to exclude 59 categories of mining from this definition, finding them to be low risk and thus not in need of financial assurance for CERCLA liability. Further, adequate state authority and regulations exist to cover any remediation costs associated with iron ore mining and preempt the need for redundant and oppressive federal financial assurance measures under CERCLA. AISI members in the iron ore mining sector have used EPA's own modeling to calculate the projected impact of this rule if it were to go final as proposed. The iron ore mining sector would likely incur over $8 billion in financial assurance obligations, which far outweighs the minimal risk of this mining sector poses. AISI and its impacted member companies in the iron ore mining sector believe the risk factors EPA used to exclude the other mining categories apply similarly to iron ore mining, which should also receive an exclusion from EPA.
b. EPA Draft Method for Specific Conductivity Criteria
On December 23, EPA issued its Draft Field-Based Methods for Developing Aquatic Life Criteria for Specific Conductivity ("Conductivity Guidance") (81 Fed. Reg. 94,370) (December 23, 2016), which purports to provide approaches for developing science-based conductivity criteria for flowing waters that reflect ecoregional- or statespecific factors. Once final, states and authorized tribes located in any region of the country may use the methods to develop field-based conductivity criteria for flowing waters. In 2011, EPA issued a similar conductivity "benchmark" for the Appalachian region - it was a chronic value of 300 micro-Siemens per centimeter (uS/cm). That equates to a Total Dissolved Solids (TDS) value of 192 ppm, which is an extremely low number that would be difficult to attain. The new EPA draft guidance takes the approach used in the Appalachian guidance and extends it to the rest of the country. Also, while the Appalachian guidance included only a chronic value, the new draft national guidance also includes acute values. Since the new document is actually a methodology, to be used with region-specific data, it does not provide specific
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00018
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 19
numbers for each region of the country, but it does contain four examples of how the methodology would be applied. The resulting values pose significant concerns for both AISTs iron ore mining and steel making members and their operations. In addition to the flawed scientific issues in the draft methodology, there are significant practical issues as to how a facility could comply with the low levels resulting from the methodology. As the draft guidance rests on a flawed scientific basis and would impose enormous compliance costs, EPA should formally withdraw the draft guidance and inform Regional offices to not apply the methodology going forward until the scientific flaws of the methodology are addressed.
c. Hazardous Air Pollutant Standards for Major Sources
EPA is currently under court ordered deadlines to complete 33 Risk and Technology Reviews (RTRs) as mandated under CAA Section 112.15 The deadlines to complete all 33 are staggered over approximately the next three years starting with an initial six RTRs by December 31, 2018, another 20 by March 13, 2020 and the final seven by June 30, 2020.16 AISI and its impacted members have been engaged with EPA in reviewing the current Section 112 NESHAPs and have provided substantial amounts of emissions data and invested time and capital resources to ensure the final result of the review process accurately captures the industry's risk profile.
Throughout the process of reviewing the Integrated Iron and Steel NESHAP (40 C.F.R. Part 63, Subpart FFFFF), EPA has insisted on collecting data and performing facility-wide risk review, including from non-category sources, i.e., sources within the facility not regulated under Subpart FFFFF. Many of these non category sources, such as coke ovens and steel pickling lines, have their own NESHAP standards making the additional risk modeling duplicative at best. AISI has numerous concerns with this approach and has expressed them to EPA in past meetings. Specifically, the data used to estimate risks from non-categorical sources in any facility-wide risk assessment is not subject to the same quality assurance and verification rigors as the data from Subpart FFFFF sources. For these reasons, AISI explained to EPA that the iron and steel industry has great concerns on the use of the156
15 42 U.S.C. 7412(d)(6), (f)(2). 16 See CA Communities Against Toxics v. EPA, l:15-cv-00512, U.S. Dist. Crt for DC Circ. (March
13, 2017) and Blue Ridge Environmental Defense League v. EPA, l:16-cv-00364, U.S. Dist. Crt for DC Circ. (March 22, 2017)(setting final rule completion deadlines for Integrated Iron and Steel Sector, 40 C.F.R. Part 63, Subpart FFFFF and Taconite Iron Ore Processing, 40 C.F.R. Part 63, Subpart RRRRR amongst 31 other sector categories).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00019
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 20
non-categorical data in assessing and aggregating the risks to develop a facility-wide risk, as such data are not necessarily representative of any actual risks from the sources in which the industry operates, especially since iron and steel facilities do not all share the same non- categorical sources. The emissions data AISI's members provided EPA pursuant to the initial Information Collection Request (ICR) prescribed specific sampling and analytical criteria as well as quality assurance requirements. Data from non- categorical sources is generally not subject to the same standards as the Subpart FFFFF data, is frequently outdated and could lead to misinformation. EPA has acknowledged the difference in the data quality in past meetings but still intends and is currently conducting risk modeling from the non-categorical sources. Given the concerns expressed above with the quality of the data, value and potential for misinterpretation of results and compounded with current EPA time and resource constraints given looming deadlines as outlined above, AISI again requests that noncategorical emissions and risk data be excluded from review.
The taconite mining industry is subject to the Taconite Iron Ore Processing NESHAP (40 C.F.R. Part 63, Subpart RRRRR). On March 22, 2017 the U.S. District Court for the District of Columbia issued an order setting a deadline for EPA to complete its NESHAP Risk and Technology Review (RTR) for taconite and twelve other NESHAP affected industries by December 31, 2018 or June 30, 2020, depending on EPA's category prioritization. Certain companies affected by the original Taconite NESHAP submitted a Petition to Delete (PTD) taconite as a source category regulated under Section 112 of the CAA in 2003 (along with supplemental information to EPA in 2006) because the HAP emissions associated with these facilities are low; do not represent an adverse impact on ambient air quality; and do not pose an unacceptable risk to human health or the environment as defined in Section 112 of the CAA. Therefore, as described in the PTD application, the taconite industry does not warrant regulation under the Section 112 NESHAP program. However, EPA has not completed its evaluation nor responded to the companies' PTD request. Prior to initiating the Taconite NESHAP RTR, EPA should review the information presented in the PTD request and determine that due to the industries' low risk, all or portions of the category should be deleted from the Taconite NESHAP RTR.
d. Clean Air Act Regulatory Changes
AISI requests the following additional changes be made to various CAA programs:
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00020
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 21
Add or strengthen EPA's specific requirement to consider costs and cost effectiveness when establishing new NAAQS, Maximum Achievable Control Technology (MACT) floor regulation, as well as State Implementation Plans (SIPs) control requirements.
The CAA should be amended so that NAAQS are not required to be reviewed every five years, but extended to at least 10-15 years or consider that revisions to the NAAQS are of such significant that they should be required to be passed by Congress and not EPA.
EPA should repeal the NSPS at 40 C.F.R. Part 60, Subparts N, Na. They are redundant and less restrictive than 40 C.F.R. Part 63, Subpart FFFFF. Subparts N and Na also require the use of obsolete technology (strip recorder charts) for tracking compliance data and reporting purposes. FFFFF is more effective at protecting the environment and recognizes available technologies exist and leave the means for tracking compliance data open.
EPA should repeal its position (not found in law or regulation, only in guidance via an FAQ document) that non-EGUs must maintain Part 75 Continuous Emission Monitoring Systems (CEMS) monitoring even if the states are following EPA's lead and not including non-EGU's in the Cross State Air Pollution Rule (CSAPR) trading program. States can demonstrate that state-wide thresholds for NOx are met through means other than Part 75 monitoring, such as the use of emissions factors. It is unreasonable, burdensome and costly to maintain and replace (as necessary) CEMS monitors that are not otherwise legally required on nonEGUs.
EPA should repeal the enforcement criteria in the Energy Star and other efficiency reward programs, which preclude a plant (or possibly even a corporation) from the program when one plant is in enforcement. Valid projects with environmental benefits should be considered because the enforcement is unrelated to the project. In addition, some enforcement actions take years to conclude, which makes plants ineligible for those same years.
e. NESHAP for Industrial, Commercial, and Institutional Boilers and Process Heaters
As noted above, many steel facilities have numerous major source emission units subject to CAA Section 112 NESHAP requirements, including industrial boiler
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00021
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 22
and process heaters (40 C.F.R. Part 63, Subpart DDDDD). The requirement to test/ tune/ test each burner of each applicable source is a burdensome exercise. At many steel making facilities there are multiple finishing lines with indirect heating furnaces that are comprised of hundreds of natural gas fired burners each below 5 MMBTU/hour. These units are considered cumulatively under the Boiler MACT and are therefore required to have annual tune-ups per 40 C.F.R. 63.7515(d). The annual tune-ups require excessive line outages and man hours. The annual requirement for testing and tuning of the many small burners can range up to $100,000 for a company with the time, equipment and proper skills to conduct the tuning. For natural gas sources with burner sizes less than a certain threshold, reducing the frequency of these tune-ups to every five years would significantly reduce the cost burden.
f. MACT "Once In, Always In" Policy
Facilities currently subject to CAA Section 112 NESHAP requirements should be afforded the opportunity to petition for removal of applicability if their potential to emit (PTE) has fallen below the major source thresholds in the CAA triggering requirements to employ MACT.17 For example, one member facility once had coalfired boilers that put the facility over the major source HAP threshold for Subpart DDDDD. After the boilers were permanently shut down, the PTE immediately fell below those thresholds, however the facility remains an affected source under the 40 C.F.R. Part 63, Subpart SSSS Surface Coating of Metal Coil. The permanent shutdown of the boilers brought the facility PTE down to Area Source thresholds, which require different emission control standards, specifically Generally Achievable Control Technology (GACT) and not the applicable MACT standards, which are more costly and directed at larger sources of air emissions.
Similarly, manufacturing facilities with emergency back-up generators onsite often find themselves subject to 40 C.F.R. Part 63, Subpart ZZZZ engine rules. These rules are intended to capture industrial engines that are much larger sources of HAPs than engines infrequently running back-up generators that typically produce minimal emissions.
17 See 42 U.S.C. 7412(a)(l)(defimng "Major Source" as sources with the potential to emit 10 tons per year or more of any hazardous air pollutant or 25 tons per year or more of any combination of hazardous air pollutants).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00022
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 23
One member facility whose back-up generators are subject to the Subpart ZZZZ NESHAP requirements reports the following emissions in 2016:
PMio: 0.0002832 tons per year (tpy) S 0 2: 0.0002639 tpy NOx: 0.0039906 tpy CO: 0.0008599 tpy VOC: 0.0003180 tpy Total: 0.005716 tpy
Having to monitor, report and maintain records for such a minimal emission source is not cost efficient. For a facility with five back-up generator engines it generally costs $500 ($100 per engine) per year to monitor, report and do mainten ance as EPA requires, which amounts to $90,000 per ton of emissions to comply with the regulations. AISI members believe this type of cost is an unjustified burden on manufacturers given the actual emissions from these sources and limited times these engines are running.
g. Restore Start-up, Shutdown, and Malfunction (SSM) Protections
From the inception of the CAA, EPA promulgated regulations and approved SIPs that granted exemptions and affirmative defense provisions for emissions during SSM events. But through an overly broad and unsupportable interpretation of two recent court decisions,18 EPA abruptly changed course and began a systemic process of eliminating existing SSM exemptions and affirmative defense provisions from various CAA regulations and previously- approved SIPs. Congress never intended for the EPA to eliminate SSM emergency exemptions that have existed and been successfully implemented for over 30 years. These exemptions provide certainty and flexibility for companies to lawfully maintain their equipment and, most importantly, to protect the safety of their workers and the surrounding communities without violating air permits.
AISI urges EPA to re-establish these important SSM exemptions that were removed from CAA regulations following EPA's strained interpretations of the Sierra Club and NRDC decisions. The administration's recent action to delay the ongoing
18 Sierra Club v. EPA, 551 F.3d 1019 (D.C. Cir. 2008); NRDC v. EPA, 749 F.3d 1055 (D.C. Cir. 2014).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00023
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 24
litigation over recalling 36 state SIPs based on SSM provisions in those plans is an important first step AISI applauds. We also support EPA's stated intention to review the legal rationale behind the decision to pull various SSM provisions from already approved SIPs.
h. Mandating Digital Camera Opacity Measurement Technique
EPA should not mandate industrial facilities to solely rely on the Digital Camera Opacity Technique (DCOT) for measuring opacity compliance as was done in the final Ferroalloys NESHAP and its reconsideration in January 2017. The existing Method 9 has been in place for many years and is a reliable methodology for measuring opacity at industrial facilities. The new DCOT technology has not been proven reliable or effective for measuring opacity from stack source emissions and should, therefore, not be extended to measure compliance for fugitive emissions from non-stack sources, such as the roof vents found at many of our members' facilities.
Additionally, the DCOT technology is currently provided by a single vendor. Given that DCOT, to our knowledge, is not currently being used by either of the two facilities in the U.S. that exist in the ferroalloys industry as the means of verifying compliance with a regulatory opacity standard, the price for using it as a compliance measure is unknown. If EPA retains DCOT as the sole-source required compliance method, then basic economics would say the price will surely rise. Given the long term historical use of Method 9 inside and outside the ferroalloys industry, it is widely available to the regulated community through a broad network of EPAapproved testing vendors. The large network of Method 9 vendors helps to maintain a price that is stable and not exorbitant.
EPA also made this change to a compliance measurement approach through the Ferroalloy rule that applies to only two facilities. However, Method 9's use for opacity standards has a much wider application than just for the ferroalloy industry. Therefore, if the agency wanted to make a change of this nature to such a central and widely used method, it should have more clearly done so as a wholesale change. This method change should have been formally proposed and received full public comment, with commenters aware of the full potential magnitude of this proposed change in compliance measurement. In seeking to establish DCOT for the ferroalloy industry, EPA did not follow its own guidance regarding broadly applicable test methods as spelled out in AISI's August 2016 comments on the reconsideration of the final Ferroalloy NESHAP. Given the concerns outlined here and in our formal
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00024
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 25
comments to Ferroalloy NESHAP reconsideration, AISI requests that EPA allow both DCOT and Method 9 as alternative opacity compliance measurement methods.
i. Formally Update EPA's Manganese Risk Level
EPA's long outdated inhalation reference concentration (RfC) for manganese and manganese compounds (Mn) that appears on the agency's Integrated Risk Information System (IRIS) website needs to be removed in accordance with EO 13777. Pending completion of a new IRIS review for Mn, AISI requests that EPA remove the Mn RfC from the IRIS website and replace it with a notation that refers to or incorporates the "minimal risk level" (MRL) developed by the U.S. Public Health Service's Agency for Toxic Substances and Disease Registry (ATSDR).
EPA developed the Mn RfC value in 1993, and it has not been revised or updated for more than two decades. Since that time, a substantial body of new scientific information has been developed concerning manganese toxicology. Several years ago when developing air toxics standards for the ferroalloys production sector, EPA recognized the outdated nature of the IRIS Mn RfC and the fact that it is no longer consistent with the latest and best available science. Accordingly, EPA ceased relying on the Mn RfC and opted instead to utilize the recently updated MRL developed by ATSDR. Further, in 2014, the EPA Office of Air Quality Planning and Standards revised its database of benchmark values for use in assessing risks from air emissions by replacing reference to the IRIS Mn RfC with the ATSDR MRL.
The agency's information quality (IQ) guidelines developed pursuant to the 2001 Information Quality Act impose a clear procedural and substantive obligation upon EPA to ensure the quality, objectivity, utility and integrity of all information disseminated by EPA. For the reasons outlined above, EPA's continued reliance upon the outdated Mn RfC on EPA's IRIS website does not meet those IQ obligations and it is therefore a prime target for elimination in accordance with the terms of EO 13777.
j. State Primacy in Regional Haze Program
In recent years, EPA has taken several actions that have dramatically increased the scope and reach of the visibility program by disapproving of SIPs to address regional haze and imposing Federal Implementation Plans ("FIPs") in their place. In doing so, EPA has unlawfully expanded the regional haze FIPs beyond the
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00025
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 26
limits imposed in the CAA and beyond the scope of EPA's own implementing regulations. These efforts, in turn, have dramatically increased the regulatory burden on stationary sources currently subject to regional haze requirements and thereby imposed economic burdens on domestic manufactures.
While AISI supports reasonable and cost effective measures to improve visibility in Class I areas, it is critical that EPA and the states do so in a simple, straightforward, and flexible manner that reduces regulatory burdens and minimizes costs. The visibility program differs from most CAA programs because it is focused exclusively on aesthetic rather than human health concerns. As a result, it is even more imperative that EPA provide an adequate justification for imposing costly regulatory requirements on states and regulated sources. In the past, EPA has failed to do so and instead has issued FIPs that imposed billions of dollars in costs to regulated sources in order to achieve visibility benefits that EPA concedes would be imperceptible to the naked eye.19
States, regulated entities, and the courts have all expressed concern over the legality of EPA's aggressive approach toward implementing the regional haze program. In fact, the Fifth Circuit recently issued a stay of EPA's federal implementation plan after concluding that petitioners had "a strong likelihood of showing that EPA exceeded its statutory authority by disapproving of Texas and Oklahoma's reasonable progress goals" and establishing costly federal reasonable progress goals to take their place. See State of Texas v. EPA, 5th Cir. Case No. 16-60118, Doc. No. 513595283. In addition, on June 14, 2013, the Eighth Circuit Court granted Petitioners (Cliffs Natural Resources, ArcelorMittal USA LLC, and the state of Michigan) a judicial stay of the Taconite Regional Haze FIP pending final decision on the merits.
AISI encourages EPA to continue to look for ways to streamline and improve the regional haze program so that it can be efficient and effective without adding unnecessary requirements and burdens on states whose current efforts have led to meeting program glide paths and ultimately long-term visibility goals without imposing uncertainty and unnecessary regulatory burdens on individual facilities.
k. Triggers for Supplemental Environmental Impact Statements (EIS)
19Visibility changes of less than 1 deciview cannot be perceived by the naked eye. 77 Fed. Reg. at 30,250.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00026
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 27
Triggers for Supplemental Environmental Impact Statements (EIS) Environmental Impact Statements have grown into review studies of staggering depth and scope and for the iron ore mining industry present years of time and millions of dollars in investment in preparation prior to initiating the permitting process. The breadth and depth required of these reviews should result in an assessment that is robust enough to serve for much of the activity undertaken by a mining operation. However, the federal language prompting the supplementing of an EIS grants enormous deference to the Responsible Governmental Unit to determine when such a study should be reopened and supplemented, a process which adds time and dollars again prior to the permitting process. The administration should adopt regulations that limit the conditions for supplementing an EIS to those where the existing state or federal permitting programs are unable to acknowledge any adjustments in project scope or environmental conditions during their respective permitting processes.
1. NPDES Authority - Groundwater Conduit Theory
AISI requests clarification that the CWA does not govern discharges to groundwater even if there is a subsurface hydrologic connection between groundwater and surface water. The text and legislative history of the CWA indicate that it was never intended to regulate discharges to groundwater. EPA has never adopted any formal position interpreting the CWA to require an NPDES permit for the discharge of pollutants to groundwater that is hydrologically connected to surface water.
AISI requests that EPA, in its rewrite of the WOTUS rule or through administrative guidance clarify that the NPDES program does not regulate discharges to groundwater, even if the groundwater is hydrologically connected to surface water.
m. Inclusion of Manganese on Drinking Water Contaminant List
AISI opposes EPA's recent addition of manganese to the Drinking Water Contaminant Candidate List (CCL 4).20 Inclusion on CCL 4 unnecessarily opens manganese up to development of a new drinking water standard as EPA is required
20 See 81 Fed. Reg. 81099)(November 17, 2016).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00027
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 28
to make a determination to regulate or not regulate at least five substances on the CCL every five years. As EPA openly acknowledges, manganese is an essential nutrient that is subject to strict homeostatic control in the human body. Large amounts of manganese are naturally present in many foods consumed as a part of a normal diet, so manganese in drinking water is unlikely to add materially to the normal daily ingestion of manganese from diet.
The science EPA identified in the CCL 4 listing concerning manganese and any potential risk it might pose as a constituent of drinking water does not support the addition of manganese to CCL 4. The purported link between the consumption of drinking water and development neurotoxicity is not sufficiently robust to warrant the development of drinking water standard for manganese. AISI, through its coalition the Manganese Interest Group (MiG) submitted substantial comments to EPA's CCL 4 docket in 2015 outlining the flaws in EPA's science justifying manganese's inclusion on the list and incorporates those comments by reference again here. Accordingly, AISI respectfully requests that manganese either be removed from the CCL list or EPA affirmatively determine a drinking water standard is not warranted.
n. Modification or Removal of Designated Uses/Water Quality Variances
The federal water quality regulations at 40 C.F.R. Section 131.10(g) provide that states may modify or remove designated uses, and sets out six factors that may be considered to support modification or removal of a designated use. Section 131.14 provides for site-specific water quality variances. Such variances must be supported with use attainability analyses and at least one of the six factors listed at Section 131.10(g). Such variances must provide that the highest attainable condition be achieved considering all feasible alternatives.
While Section 131.10(g)(6) provides for consideration of costs for changes in designated uses and water quality variances, the cost threshold is so high (substantial and widespread economic and social impact) that this factor can only be used in rare and unusual circumstances. The regulations do not allow for considering a balancing of costs and environmental benefits. There are circumstances where designated uses may be partially attained (e.g., fish in attainment, macroinvertebrates not in attainment) and the investment and operating costs to attain full attainment are so high as to be unreasonable. The following addition to Section 131.10(g) is proposed to address this situation:
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00028
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 29
Section 231.10(g)(7). As determined by the Director, the costs to achieve full attainment of the designated use are wholly out of proportion to the environmental benefits that would be achieved from full attainment of the attainment use. The Director shall consider a site-specific use attainability analysis and the lowest cost to achieve the designated use for technicallyfeasible alternatives.
o. Mass and Concentration Effluent Limits
Title 40 C.F.R. Section 122.45(f)(2) provides that NPDES permit effluent limits shall be in terms of mass (e.g., lbs./day), and that permit writers may also establish effluent limits in terms of concentration. The categorical effluent limitations guidelines for many industrial categories are production-based, whereby a reasonable measure of actual production is used with the effluent limitations guidelines factors to calculate mass technology based effluent limits. The effluent limitations guidelines do not require attainment of any particular effluent flow or effluent concentration, so long as the mass technology based NPDES permit effluent limits are attained. When duplicative concentration effluent limits are established by permit writers without cause, the permittee is, in effect, placed in double jeopardy for effluent limit exceedances and the flexibility to use any combination of effluent flow and effluent concentration is removed.
To remedy this situation, the following modifications to 40 C.F.R. Section 125(f)(2) are proposed:
Section 122.45(f)(2). Pollutants limited in terms of mass additionally may be limited in terms of other units of measurementfor cause, and the permit shall require the permittee to comply with both limitations. The provisions of this section do not apply where production based effluent limitations guidelines and/or best professional judgment are used to determine the mass effluent limitations and the mass effluent limitations apply at an internal outfall or internal compliance monitoring point. In such cases, only the mass effluent limitation shall apply at the internal outfall or internal compliance monitoring point. Supplemental or additional concentration effluent limits may apply at the correspondingfinal outfall that discharges to a receiving water, as may be appropriate for water quality based effluent limits or other effluent limits derived by the permitting authority.
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00029
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 30
p. Other Water-Related Matters:
There are a number of other water-related actions that we would raise for specific comment:
We have concerns regarding a number of issues with the recently released guidance on conducting fish consumption surveys.21 We believe that this guidance should be reconsidered, and fully endorse the Federal Water Quality Coalition comments that were submitted on the draft guidance and are in the docket for that action.
EPA should modify the NPDES and pretreatment rules to require a permitting authority to act on a variance request within 120 days from the time of application, or the variance is deemed granted. Further, EPA should modify the NPDES permit regulations so that short-term concentrations are not an issue if overall the mass loading is reduced; otherwise, diluting the concentration results in more total pollution to the waterbody.
EPA should mandate public reviews during the Total Maximum Daily Load (TMDL) development process. Often facility managers do not see what the documents supporting the TMDL look like until they are final. It is unreasonable to expect permittees to comment on a TMDL during a permit review process, which gives the permittee little time to analyze and respond to the TMDL.
AISI has concerns with EPA's proposed NPDES Application and Program Updates proposed rule, issued in May 2016 and not yet finalized.22 We request that EPA issue a new proposal considering the significant comments received, especially regarding permit shields and permit reforms. We endorse the Federal Water Quality Coalition's comments submitted on this proposed rule, which are in the docket for this action.
AISI also has concerns about certain aspects of the draft recommended aquatic life criteria for selenium and implementation guidance. We endorse the Federal Water Quality Coalition's comments submitted on this draft guidance, which are in the docket for this action.
q. TSCA Chemical Data Reporting
21 See, https:/ / www.epa.gov/fish-teeh/gmdaBce-oendttcting-feh-corisurniption-surveys 22 See 81 Fed. Reg. 31343 (May 18, 2016).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00030
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 31
Under the Toxic Substances Control Act (TSCA) regulation, the Chemical Data Reporting (CDR) regulations require exceptionally detailed monitoring, recording, and reporting of the chemical make-up of our members' steel and steel coatings, raw materials, downstream uses and downstream users, among other things.23 Having to track all of the different industry sectors, customers and out processors that our steel touches is simply an inefficient use of valuable time without a concurrent benefit. It is overly burdensome to the steel industry to report on the general safety of a product that has been widely produced for several centuries and whose chemical makeup is well known and that poses little risk from exposure. AISI recommends exempting certain steel products from CDR reporting that are shown to pose little to zero risk from exposure. Our members report having to spend approximately 800 man-hours companywide to complete CDR reporting that would be better invested in addressing issues posing a higher risk potential.
r. TSCA Regulation of Asbestos
In December 2016, EPA announced that asbestos would be one of the first ten chemical substances to undergo a risk evaluation under the revised version of TSCA.24 AISI supports regulations based on sound science that prevent harmful exposure to asbestos. Before EPA begins its asbestos risk evaluation, however, the agency must clearly define "asbestos" for the purposes of this risk evaluation and any subsequent risk management rulemaking. That definition must be sufficiently precise to differentiate asbestos from common, rock-forming elongated particles or cleavage fragments, which have not been found to cause health effects like those associated with asbestos. Providing a precise and accurate definition and analytical methods for asbestos will help to focus EPA's risk evaluation on harmful asbestos and asbestiform fibers, which present real risks to human health, and avoid diverting the agency's attention and resources away to non-asbestiform mineral particles that have not been shown to cause asbestos-related diseases.
In a presentation seeking public comment on the scope of risk evaluations, EPA stated it will adhere to the definition of asbestos set out in Title II of TSCA:
23 40 C.F.R. Part 711. 24 81 Fed. Reg. 91927 (December 19, 2016).
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00031
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 32
The term 'asbestos' means asbestiform varieties of--(A) chrysoble (serpentine), (B) crocidolite (riebeckite), (C) amosite (cummingtonite-grunerite), (D) anthophyllite, (E) tremolite, or (F) actinolite. (TSCA 202(3))
AISI supports the continued use of this longstanding definition, additionally, in line with EPA's request for comments, we request the following recommendations to further clarify the scope of the TSCA risk assessment of asbestos:
1. AISI requests EPA to explicitly state in its pending asbestos regulation that "The term 'asbestos' does not include cleavage fragments (sometimes used interchangeably with non-asbestiform elongate mineral particles)." This addition could be bolstered by including a definition of "cleavage fragments." One readily available definition is from Appendix B to OSHA's asbestos regulations (29 C.F.R. 1910.1001), which excludes "cleavage fragments" defined as "mineral particles formed by comminution of minerals, especially those characterized by parallel sides and a moderate aspect ratio (usually less than 20:1)." Alternatively, EPA can define "asbestiform" to clearly exclude cleavage fragments. The National Stone, Sand and Gravel Association (NSSGA), for example, has proposed a workable definition of "asbestiform" as "the mineralogical habit or form of a mineral in which ultra-fine single crystal fibers (fibrils) occur in bundles that can be separated into increasingly finer fiber bundles that typically display curvature." AISI requests EPA to exclude mineral particles with aspect ratios less than 20:1 from the "asbestiform" definition.
2. AISI also requests EPA explicitly state in its pending asbestos regulation that the agency will continue to define "asbestos containing material" to include only "material which contains more than 1 percent asbestos by weight."
s. Modify the Hazardous Waste Generator Improvements Rule
AISI has concerns with the recently issued Hazardous Waste Generator (HWG) Improvement's rule. EPA's rule includes some common-sense updates that will bring grea ter efficiency and clarity to the HWG regulatory program. However, we are concerned by other aspects of the rule. We are especially concerned by the change to the criteria under which a generator of hazardous waste is deemed in
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00032
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 33
violation of the Resource Conservation and Recovery Act (RCRA) permitting program applicable only to facilities that treat, store or dispose of hazardous waste (TSDFs). Under the new rule, failure to meet any one of EPA's long list of 'conditions for exemption' could subject a generator to multiple violations and substantial penalties. Even a minor deviation in compliance would cause a generator to now be considered an illegal TSDF. We oppose this provision and request that the agency promptly change this provision through notice and comment rulemaking.
t. Duplicative Reporting Under Greenhouse Gas Reporting Program
Greenhouse gas (GHG) emissions are based on production and fuel usage data that are also reported in the annual Title V fee emissions reports.25 Completing these reports adds approximately 350 man-hours companywide according to one AISI member with four domestic facilities, while many of AISI's members have upwards of 20 facilities in the U.S. The GHG Reporting Program requires either a mass balance approach or annual emissions testing to calculate emissions from basic oxygen furnaces and electric arc furnace (EAF) steelmaking operations. The mass balance approach is an impractical method for some facilities to use therefore, those members must annually tests their basic oxygen furnaces and EAFs. This testing costs approximately $120,000 companywide each year to complete and is duplicative to similar reporting conducted annually as part of the Title V permitting requirements. AISI recommends that EPA streamline these duplicative reporting programs and use data the agency already collected from facilities for use in the GHG Reporting Program.
25 40 C.F.R. Part 70. Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00033
Docket ID No.: EPA-HQ-OA-2017-0190 May 15, 2017 Page 34
Conclusion
Thank you for your attention on the important matter of addressing unnecessary regulatory burdens facing domestic manufacturing. AISI and its members are committed to working with the Trump Administration and its executive leadership in implementing common sense regulations and policies that are based on sound science and data and consistent with the statutory missions of each agency.
Respectfully submitted,
Ml\ j
\\
s~\
VJ
Thomas J. Gibson President and CEO
Sierra Club v. EPA 18cv3472 NDCA
Tier 7
ED 002061 00119446-00034