


                                       
                                       
                                       
                                       
                                       
National Emission Standards for Hazardous Air Pollutants from Hydrochloric Acid Production (40 CFR part 63, subpart NNNNN)
Residual Risk and Technology Review, Final Amendments


                                       
                                       
Summary of Public Comments and Responses for the Risk and Technology Review for the Hydrochloric Acid Production Source Category 
(84 FR 1570, February 4, 2019)
                                       
                                       
                                       
                                       
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Sector Policies and Programs Division
Research Triangle Park, NC 27711



March 2020


The primary contact regarding questions or comments on this document is:
      Nathan Topham
      U.S. Environmental Protection Agency
      Office of Air Quality Planning and Standards
      Sector Policies and Programs Division
      Metals and Inorganic Chemicals Group
      Mail Code D243-02
      Research Triangle Park, North Carolina 27711
      (919) 541-0483
      Email: topham.nathan@epa.gov


Purpose and Background
On February 4, 2019, the U.S. Environmental Protection Agency (EPA) proposed revisions to the National Emission Standards for Hazardous Air Pollutants from Hydrochloric Acid Production (HCl Production NESHAP, 40 CFR part 63, subpart NNNNN) to address the results of a residual risk and technology review (RTR), amend provisions addressing periods of startup, shutdown, and malfunction (SSM), and require electronic reporting. The EPA received 7 comment letters on the 2019 proposal. The commenter and item number in Docket ID No. EPA-HQ-OAR-2018-0417 are listed in Table 1.
Table 1. List of Commenters, HCl Production NESHAP Residual Risk and Technology Review Proposed Rule (84 FR 1570)



Docket Item No. EPA-HQ-OAR-2018-0417-
Commenter Name
Affiliation
                                     0122
Francis Steitz and Robert Colby
NACAA
                                     0125
Various, see transcript
Public Hearing Speaker
                                     0134
Russell A. Wozniak and Laurie K. Christiano
Dow and Dow Agrosciences
                                     0139
Gregory M. Bowser
Louisiana Chemical Association
                                     0144
Michele Roberts
Environmental Justice Health Alliance National, et al.
                                     0145
Brendan Mascarenhas
American Chemistry Council
                               0147, 0148, 0149 
Emma Cheuse
Earthjustice on behalf of Blue Ridge Environmental Defense League et al.


Risk Assessment Methodology and Results
Comment (0125): EPA states on its website the OAQPS mission is to develop rule-makings to decrease and limit air pollution, and despite this, EPA continues to find cancer and noncancer risks to be acceptable without accounting for the cumulative risk, differences in early life exposure, and the co-location of various source categories at a given facility. EPA operates as if people are exposed to one chemical at a time, when, in reality, people may be exposed to dozens of chemicals with such communities being disproportionately harmed by many pollutants that EPA does have the power and duty to reduce. Communities need relief today. And, in fact, considering that the most current NATA data utilizes air emissions from 2014, communities needed that protection years ago. EPA must act to reduce the facility-wide risks caused by co-located facilities, and by doing so, EPA takes a step closer to addressing cumulative risk that's faced by communities. Moreover, EPA must calculate and reduce the known chronic and acute risk posed by ethylene oxide and other pollutants.
The Clean Air Act provides EPA with authority under Section 112 and through the general duty clause, and commenters urge EPA to use it to protect the health of disproportionately impacted communities. According to the commenters, it is essential that EPA set stronger emission limits for entire industrial sites and facilities that are located within 20 miles of each other, because a lot of people live in areas where there are high concentrations of industrial facilities, and not just in this source-focused way.

By continuing to fail to consider the cumulative impacts of exposures to multiple chemicals from multiple sources, including current exposures from legacy uses, EPA assessments are failing to reflect the toxic reality that families suffer in every state in the country. Allowing toxic emissions and hazardous exposures using risk estimates that consider only one chemical at a time fails to address the real-world impacts of coal plants, refineries, cement kilns, chemical manufacturing plants, metal smelters, incinerators, highways and trucking routes, shipping channels, and Superfund and other hazardous sites.

It's time for EPA to act to protect people from the sources under review in this rule making and the sources that are not under review in this specific rule-making. And it is EPA's job to protect our communities, and they should start with the communities that have the highest emissions and the highest cumulative risk to our health from ethylene oxide and the other chemicals we're breathing.
Response: We disagree with the comment that we find cancer and noncancer risks to be acceptable without considering exposure to a mixture of chemicals and the cumulative risk. The EPA typically assesses the cumulative cancer risks from all emitted carcinogens and the cumulative noncancer hazard indices from all emitted non-carcinogens that are HAP affecting the same target organ system for both the source category emissions and the facility-wide emissions. To address the effect of mixtures of carcinogens, the individual cancer risks for the source categories are aggregated for all carcinogens. Since the hydrochloric acid production source category does not emit carcinogens, our risk assessment was limited to the risk of non-cancer health effects from exposure to the two hazardous air pollutants (HAPs) emitted (chlorine and hydrochloric acid) from this source category.
In assessing noncancer hazard from chronic exposures we assessed the cumulative noncancer hazard indices from all emitted non-carcinogens affecting the same target organ system. In assessing noncancer hazard from chronic exposures for pollutants that have similar modes of action or (where this information is absent) that affect the same target organ, we aggregated the hazard quotients. This process creates, for each target organ, a target-organ-specific hazard index (TOSHI), defined as the sum of hazard quotients for individual HAPs that affect the same organ or organ system. All TOSHI calculations were based exclusively on effects occurring at the "critical dose" (i.e., the lowest dose that produces adverse health effects). 
Further, the risk assessment modeling accounts for the effects of multiple facilities that may be in proximity when estimating concentration and risk impacts at each block centroid. When evaluating the risks associated with a specific source category, we combined the impacts of all facilities within the same source category and assessed chronic exposure and risk for all census blocks with at least one resident (i.e., locations where people may reasonably be assumed to reside rather than receptor points at the fence-line of a facility). The maximum individual risk and TOSHI consider the combined impacts of all sources in the category that may be in proximity.  
Section 112(f) requires the EPA to perform a review of whether there is risk remaining from emissions from a source category after promulgation of the technology-based MACT standards for that source category. To this end, the EPA conducts a comprehensive assessment of the risks associated with exposure to the hazardous air pollutants emitted by the source category and supplements that with additional information that is available about other possible concurrent and relevant risks (i.e., whole-facility). In determining whether the risk presented by the source category is "acceptable" within the meaning of CAA section 112(f)(2), we do not consider the co-location of sources outside the source category. However, the whole-facility risk assessment is generally used to inform the ample margin of safety analysis for the source category in question. The development of whole-facility risk estimates provides additional information about the potential cumulative risks in the vicinity of the RTR sources, as one means of informing potential risk-based decisions about the RTR source category in question. We recognize that, because these risk estimates were derived from facility-wide emissions estimates which have not generally been subjected to the same level of engineering review as the source category emission estimates, they may be less certain than our risk estimates for the source category in question, but they remain important for providing context as long as their uncertainty is taken into consideration in the process.
We disagree with the comment that the risk assessment does not consider early-life exposure. We acknowledge that for a given HAP, population subgroups, including children, may have a potential for risk that is greater than the general population due to greater relative exposure and/or greater susceptibility to the toxicant. The assessments we undertake to estimate risk account for this potential vulnerability; the risk assessments we perform implicitly account for this greater potential for exposure by assuming lifetime exposure, in which populations are conservatively presumed to be exposed to airborne concentrations at their residence continuously, 24 hours per day for a full lifetime, including childhood. With regard to children's potentially greater susceptibility to noncancer toxicants, the assessments rely on EPA (or comparable) hazard identification and dose-response values which have been developed to be protective for all subgroups of the general population, including children. 
For example, a review of the chronic reference value process concluded that the EPA's RfC derivation processes adequately considered potential susceptibility of different subgroups with specific consideration of children, such that the resultant RfC values pertain to the full human population "including sensitive subgroups," a phrase which is inclusive of childhood. With respect to cancer, we use age-dependent adjustment factor for carcinogenic pollutants that are known to act via mutagenic mode of action. In lieu of chemical-specific data on which age or life-stage specific risk estimates or potencies can be determined, default age dependent adjustment factors can be applied when assessing cancer risk for early-life exposures to chemicals which cause cancer through a mutagenic MOA. With regard to other carcinogenic pollutants for which early-life susceptibility data are lacking, it is the Agency's long-standing science policy position that use of the linear low-dose extrapolation approach (without further adjustment) provides adequate public health conservatism in the absence of chemical-specific data indicating differential early-life susceptibility or when the mode of action is not mutagenicity. The basis for this methodology is provided in the 2005 Supplemental Guidance.
To examine the potential for any disproportional harm to the communities that might be associated with the source category, we performed a demographic analysis of the population close to the facilities. In this analysis, we evaluated the distribution of HAP-related non-cancer risks (no carcinogens are emitted from sources in the source category) from hydrochloric acid production across different social, demographic, and economic groups within the populations living near facilities identified as having the highest risks. The methodology and the results of the demographic analyses are included in a technical report, Risk and Technology Review -- Analysis of Demographic Factors for Populations Living Near Hydrochloric Acid Production Source Category Operations, available in the docket for this action (Docket Item Number EPA - HQ - OAR - 2018-0417-0024). These results, for various demographic groups, are based on the estimated risks from actual emissions levels for the population living within 50 kilometers (km) and within 5 km of the facilities. The results of the demographic analysis indicate that emissions from these facilities do not disproportionally impact any demographic group based on race, socioeconomics, education, and linguistic isolation.
In response to the comment on the timeliness of our emissions data, the emission estimates for this source category were obtained from the 2014 National Emissions Inventory (NEI), state databases of Title V permits, and engineering judgement to fill data gaps. We used 2014 data, and augmented it with additional sources of information, because emission inventories from that year were the most complete and up-to-date available at the time of our proposal. The 2017 NEI did not become available until well after the comment period for the HCl Production RTR proposal closed. Working with industries and states, we update our emission inventories every 3 years. The risk estimates assume a lifelong exposure to 2014 levels because calculating exposures based on projections to more recent years would be much more complex and uncertain. With regard to the facility-wide risks due to ethylene oxide and trichloroethylene, which are emitted from emission sources that are not part of the HCl Production source category, as EPA explained in the proposal preamble (84 FR 1570, February 4, 2019), EPA intends to evaluate those facility-wide estimated emissions and risks further and may address these in a separate future action, as appropriate.
Comment (0122 and 0147): In assessing the cancer risks related to the source category, EPA used long-term concentrations affecting the census blocks within 50 kilometers of each facility. One commenter states that this analysis dilutes the effect of sources' emissions by estimating the impact at the centroid of the census block instead of at the property line or wherever the maximum exposed individual is. Census blocks can be large geographically, depending on the population density, so the maximum point of impact can be far from the centroid. It could be elsewhere in the census block, including at or near the property line where people may live or work. EPA itself alludes to this problem in the proposal. Further, even if the area near the property line is not developed, over time homes and businesses could locate closer to the facility. While it is possible that population distribution is homogenous over a census block, this assumption is not necessarily accurate in considering the predicted impacts from the location of a source. The commenter recommends EPA identify and use the truly maximum individual risk, irrespective of its location in the census block, rather than using the predicted chronic exposures at the census block centroid as surrogates for the exposure concentrations for all people living in that block.

Another commenter states that the EPA did not assess chronic or cancer risk at the closest or maximum point of exposure, even in those instances where local residents live nearer to a facility than the geographic centroid of the census block. The commenter explains that this approach underestimates exposure and could easily be remedied because HEMAERMOD allows the system to make such an adjustment. The commenter notes that this conflicts with the recommendation of the SAB, which has urged the EPA to consider "specific locations of residences." The commenter stated that taking geographic variation out of the equation fails to properly account for exposure to the "individual most exposed to emissions" as required by section 112(f)(2)(A), and fails to provide an accurate estimate of risk. The commenter argued that estimating the annual average concentrations at the area-weighted centers of census blocks blatantly and artificially underestimates the risk estimated for people at the fence-line since the center of a census block is almost always further away from the facility than the fence-line. The commenter also noted that census blocks vary greatly in size, yet the EPA provides no evidence that it reviewed census block size or configuration to consider how concentrations of pollutants might vary within these blocks. Therefore, the commenter concluded, area-weighted centers of census blocks may be significantly underestimating exposure in some cases.

The commenter states that the EPA's failure to adjust receptor points for residents living on the fence-line is particularly inexcusable given that the EPA's air dispersion model allows for such an adjustment, and that such an adjustment was appropriately made for the estimation of acute health risks. The commenter concludes that, having recognized that the maximum exposed individual for acute risks is likely present at the fence-line, the EPA cannot justify failing to analyze cancer and other chronic health effects in a similar manner.

As part of the residual risk assessment for this source category, EPA unlawfully and arbitrarily fails to assess the health risks for "the individual most exposed to emissions" from HCl facilities as § 7412(f)(2) requires. Instead, EPA has chosen an arbitrary point, a census-block centroid, without demonstrating that this is equivalent to the person "most exposed." 84 Fed. Reg. at 1576. EPA itself implicitly admits that its approach is ineffective because it made adjustments in some instances "to better represent the population in the blocks," and to "add additional receptor locations where the population of a block is not well represented by a single location." Id. at 1580. As EPA's Science Advisory Board explained in a 2018 report, EPA's ad hoc attempt to assess health risks to the most-exposed person is not scientifically justified. EPA's failure to meet this core statutory requirement renders the proposed rule unlawful and arbitrary. 
Even in its current review, the SAB has found that EPA's report "does not provide enough information about the tool, especially regarding criteria that would determine the number and placement of new receptors." Like the SAB panel, Commenters are concerned that EPA's method is not scientific or lawful, would not be reproducible, and that application would vary should a different risk assessor conduct modeling on an individual facility. It is imperative that the EPA has verification metrics in place to ensure that it assesses health risks based on the actual or likely locations of specific receptors representative of those living near the facility and/or who are the people most exposed to a source or source category.
Response: In a national-scale assessment of lifetime inhalation exposures and health risks from facilities in a source category, it is appropriate to identify exposure locations where it may be reasonably expected that an individual will spend a majority of his or her lifetime. In determining chronic risks, it is appropriate to use census block information to locate where people reside and to use those census block locations to estimate exposures and risks to individuals living near such facilities. Census blocks are the finest resolution available as part of the nationwide population data (as developed by the U.S. Census Bureau); on average, a census block is comprised of approximately 50 people and about 20 households. In the EPA RTR risk assessments, the geographic centroid of each census block containing at least one person is used as a default to represent the location where all the people in that census block live. The census block centroid with the highest estimated exposure then becomes the location of maximum exposure, and the entire population of that census block is assumed to experience the maximum individual risk. In its 2010 peer review of the methodologies used to estimate risks as part of the RTR rulemaking efforts, the EPA's SAB endorsed this approach.
However, in some cases, the census block centroid may not accurately indicate the actual locations of residences close to the facilities in question. If, for example, actual residence locations are closer to or farther from facility emission points than the census block centroid, this may result in an overestimate or underestimate of the actual annual concentrations. The larger the census block is, the greater the potential for the census block centroid to not represent actual residential locations. In addition, when a census block centroid is located on the facility property, it is not representative of residential locations, and may result in an overestimate of the actual annual concentrations.
In 2017, we requested SAB's review of updated RTR screening methodologies and several specific enhancements to our chronic inhalation risk assessment as described in the technical report, "Screening Methodologies to Support Risk and Technology Reviews (RTR): A Case Study Analysis." Since the 2010 SAB review, we developed a census block receptor tool that automatically identifies census block centroids near facilities in a source category, including those that might be located on facility property and census blocks that are very large. Specifically, the receptor tool will identify all census block centroids around a facility (e.g., 1,000 meters) and further flag any census block centroid 1) that is within a set distance (e.g., 300 meters) of the facility emission points as being potentially on facility property or 2) as a large block if it has an area greater than 3 km[2]. We asked SAB, "Is the census block receptor check tool an appropriate method for identifying and adjusting model receptors to ensure the receptors are representative of residential locations?"  
SAB provided specific comments on the census block receptor check tool in their September 13, 2018 final report to EPA. The SAB comments are summarized below along with our response. 
 SAB Comment: 
            "Overall, the SAB believes that methods predominantly relying on census block centroid location  -  including cases where the enhancement tool is applied  -  can in some cases be reliable, but additional effort is needed to verify that receptors are representative of residential areas near the facilities. One approach would be to review satellite imagery within 1km of all facilities, not just those in identified large census blocks, and manually add receptors as needed to appropriately represent population centers. However, any manual placement would be subjective and not reproducible between risk assessors."
      EPA Response: We appreciate SAB's acknowledgement that census block centroids can be an appropriate foundation for identifying residential locations near facilities. As SAB recommends, we do review aerial imagery around every facility to identify the residences closest to the facility. We use the tool to highlight large blocks (typically within 1 km), but we review aerial imagery for every facility. User receptors are added if there is not already a census block centroid or other receptor near the closest residences. There is some subjectivity in this review, but it is unlikely to cause significant differences in results.
 SAB Comment: 
            "The SAB recommends the Agency evaluate an alternative approach that uses the same 2011 National Land Cover data (NLCD) used for the Urban/Rural dispersion selection enhancement tool to automate the process of identifying population centers. The NLCD data is available at a high spatial resolution (30 m) and receptors could be placed in areas of developed land use classes 22-24. Aerial photos (e.g., Google Earth) can then be used to check that the land use data population receptor placement is appropriate."
      EPA Response: The developed land use class in NLCD does not specify residential versus commercial or industrial, therefore placing receptors in such areas would not be a good representation of residences. Using aerial imagery to review receptors placed in every area of class 22-24 land use to confirm their validity would be onerous and still be subjective.
 SAB Comment: 
            "If the EPA prefers to continue using census block centroids as nearby exposure receptors, then the SAB suggests additional enhancements to make the tool less ad hoc. Facilities are better represented as polygons than points. Satellite imagery can be used to delineate the facility area and then GIS could easily exclude receptor points that were located within that area."
      EPA Response: We agree that having polygons that represent facility boundaries would allow the use of GIS to exclude block centroids within the polygons. However, we rarely, if ever, have such data for facilities, and while aerial imagery can be used to determine where buildings are, it cannot be used to determine all property owned by a facility. Also, facility boundary data is not useful for determining the location of residences in surrounding blocks.
For this source category, we reviewed aerial images around every facility to determine whether the census block centroids were located on facility property and whether census block centroid locations were representative of residential locations. Census blocks that were determined to be on facility property were relocated offsite to a location central to the residences in the block. If a census block centroid was not representative of the residential locations within that block, we relocated it to better represent them and/or we added receptors for residences nearer to the facility than the centroid. For this source category, we moved 23 census block centroids to better represent residential locations.  In addition, we added 16 additional user receptors to better represent residential locations near the facilities. Appendix 4 of the risk assessment document provides additional information on these changes.  
Comment (0139): One commenter supported EPA's proposed findings under CAA section 112(f) that the existing standards provide an ample margin of safety and under CAA section 112(d)(6) that no developments in technology have occurred. The commenter agrees with EPA's proposed decision that the RTR review does not warrant revisions to the NESHAP. 

Response: We acknowledge the support of our findings under CAA section 112(f) and section 112(d)(6) with respect to the Hydrochloric Acid Production source category.

Comment (0147): One commenter asserts that EPA should find risks at HCl production facilities to be unacceptable because 1) facility-wide cancer risks are 600-in-1 million; 2) facility-wide chronic non-cancer hazard quotient is 6; 3) source category-specific chronic non-cancer HQ is 2; 4) the acute risk could be as high as an HQ of 42 and appears to be well above 1; 5) all of the risk assessment values presented are underestimates because they have gaps that ignore components of real-world health risks. The commenter states that EPA fails to address the source category-specific TOSHI of 2 when determining risk acceptability. 

Response: Regarding the HQ of 42 for acute risk, this value was calculated based on concentrations within a facility's fence-line and thus is not considered a risk to the public. 
The EPA considers both actual and allowable emissions in assessing chronic exposure to determine whether risks are acceptable. We have considered all of the risks to public health remaining after implementation of the HCl Production NESHAP and determined that the very low remaining risks are acceptable. In addition, the development of facility-wide risk estimates provides additional information about the potential cumulative risks in the vicinity of the RTR sources, as one means of informing the ample margin of safety step of our risk analysis for the RTR source categories in question.
We generally draw no bright lines of acceptability regarding cancer or noncancer risks from source category HAP emissions and acknowledge that it is always important to consider the specific uncertainties of the emissions and health effects information regarding the source category in question when deciding exactly what level of cancer and noncancer risk should be considered acceptable. In addition, the source category-specific decision of what constitutes an acceptable level of risk should be a holistic one; that is, it should simultaneously consider all potential health impacts -- chronic and acute, cancer and noncancer, and multipathway -- along with their uncertainties, when determining the acceptable level of source category risk. The Benzene NESHAP (54 FR 38044, September 14, 1989) decision framework acknowledged this; in today's world, such flexibility is even more imperative, because new information relevant to the question of risk acceptability is being developed all the time, and the accuracy and uncertainty of each piece of information must be considered in a weight-of- evidence approach for each decision. This relevant body of information is growing fast (and will continue to grow even faster), necessitating a flexible weight-of-evidence approach that acknowledges both complexity and uncertainty in the simplest and most transparent way possible. While this challenge is formidable, it is nonetheless the goal of the EPA's RTR decision-making, and it is the goal of the risk assessment to provide the information to support the decision-making process.

Comment (0147): One commenter says that the EPA has failed to show that risks are not significantly higher than the values presented in the record based on uncertainties, omissions, and underestimates in the risk assessment. The commenter provides the following examples of uncertainties or omissions: EPA has not appropriately accounted for risks due to early exposure in assessing facility-wide cancer risks; EPA has not conducted a multipathway analysis for any pollutant, including from co-located facilities; EPA has not recognized that the cumulative impacts and multiple source exposure from various sources outside the facility boundaries make the risk to the people most exposed to the facility's pollution even more unacceptable; EPA did not evaluate facility-wide risks based on "allowable" emission rates; and EPA should decide that it is unjust and inconsistent with the Clean Air Act to allow the health risks caused by these sources to fall disproportionately on communities of color and lower income communities who are least well equipped to deal with the resulting health effects without doing anything to reduce those risks. 

Response:  We disagree with the comment that the risk assessment does not consider early-life exposure. We acknowledge that population subgroups, including children, may have a potential for risk that is greater than the general population due to greater relative exposure and/or greater susceptibility to the toxicant. As explained in an earlier response to comment, the assessments we undertake to estimate risk account for this potential vulnerability. With regard to children's potentially greater susceptibility to noncancer toxicants, the assessments rely on EPA (or comparable) hazard identification and dose-response values which have been developed to be protective for all subgroups of the general population, including children. 
As for our multipathway assessment, EPA conducts this assessment using actual source category emissions of persistent and bioaccumulative HAP (PB-HAP). This source category emits only two HAP, chlorine and hydrochloric acid, and neither are PB-HAP. Thus, both actual and allowable emissions of PB-HAP are zero. Thus, we did not perform a multipathway analysis..
As to the cumulative impacts of exposure from various sources outside the facility boundaries, the EPA conducts a comprehensive assessment of the risks associated with exposure to the HAP emitted by the source category and supplements that with additional information that is available about other possible concurrent and relevant risks. While the incorporation of additional background concentrations from the environment in our risk assessments (including those from mobile sources and other industrial and area sources) could be technically challenging, they are neither mandated nor barred from our analysis. In developing the decision framework in the Benzene NESHAP currently used for making residual risk decisions, the EPA rejected approaches that would have mandated consideration of background levels of pollution in assessing the acceptability of risk, concluding that comparison of acceptable risk should not be associated with levels in polluted urban air (54 FR 38044, 38061, September 14, 1989). Background levels (including natural background) are not barred from the EPA's ample margin of safety analysis, and the EPA may consider them, as appropriate and as available, along with other factors, such as cost and technical feasibility, in the second step of its CAA section 112(f) analysis. This assessment excludes background contributions because the available data are of insufficient quality upon which to base a meaningful analysis. Further, our approach here is also consistent with the approach we took regarding this issue in the Hazardous Organic NESHAP (HON) RTR (71 FR 76603, December 21, 2006), which the court upheld in the face of claims that the EPA had not adequately considered background concentrations. (NRDC v. EPA, 529 F.3d 1077 (D.C. Cir. 2008)).

We disagree with the commenter that facility-wide risks should be estimated based on "allowable" emissions. Emissions included in a facility-wide assessment have generally not been subjected to the same level of engineering review as the source category emission estimates and are likely to be less certain than our emissions estimates for the source category. Furthermore, "allowable" must be defined by what is allowable by a regulation. Many of the sources outside of the source category may be unregulated, and thus, not have an "allowable" level. Therefore, estimating risks based on facility-wide allowable emissions would be highly uncertain, limiting their usefulness in risk-based decisions, and not provide additional context to the source category risks beyond what is estimated using "actual" emissions.

To examine the potential for any disproportional harm to the communities that might be associated with the source category, we performed a demographic analysis of the population close to the facilities. In this analysis, we evaluated the distribution of HAP-related non-cancer risks (no carcinogens are emitted) from hydrochloric acid production across different social, demographic, and economic groups within the populations living near facilities identified as having the highest risks. The methodology and the results of the demographic analyses are included in a technical report, Risk and Technology Review -- Analysis of Demographic Factors for Populations Living Near Hydrochloric Acid Production Source Category Operations, available in the docket for this action (Docket Item Number EPA - HQ - OAR - 2018-0417-0024). These results, for various demographic groups, are based on the estimated risks from actual emissions levels for the population living within 50 kilometers (km) and within 5 km of the facilities. The results of the demographic analysis indicate that emissions from these facilities do not disproportionally impact any demographic group based on race, socioeconomics, education, and linguistic isolation.

Comment (0147): One commenter states that EPA has failed to properly apply section 112(f) of the CAA by only considering facility-wide cancer risks at the ample margin of safety stage of the assessment. The commenter argues that while the CAA requires EPA to perform a residual risk rulemaking for each "source category," it also requires EPA to regulate based on the "lifetime excess cancer risks to the individual most exposed to emissions from a source in the category." 42 U.S.C. § 7412(f)(2) (emphasis added). The commenter says that EPA has repeatedly interpreted this as requiring consideration of health risks to that person who is most exposed; not merely to assess emissions or risks from a given source in a category. For example, in the 1999 Residual Risk Report to Congress, EPA stated that its policy was to implement the Commission on Risk Assessment and Risk Management recommendations from 1997 to "examine and choose risk reduction options in light of total facility risks and public health context." See, e.g., EPA, Residual Risk Report to Congress at 22, 129, EPA-453/R-99-001 (Mar. 1999). To properly perform a § 7412(f)(2) assessment, therefore, EPA must evaluate the full scope of cumulative health risks to which the most-exposed individual is exposed. The Act requires EPA to focus on the person "most exposed" and protect public health. EPA must satisfy the Act's directive "to provide an ample margin of safety to protect public health in accordance with this section..." 42 U.S.C. § 7412(f)(2). This provision does not say to protect public health only from the source category. This language governs the risk assessment and the acceptability determination as well as the ample margin determination, as EPA has explained in the preamble to the proposed rule.

Response: We disagree with the comment that we have failed to properly fulfill CAA section 112(f)(2) requirements. The focus of section 112(f)(2) is to ascertain whether the source category at issue is sufficiently regulated through previously promulgated technology-based standards to reduce the risk posed by those sources. Thus, EPA performs a risk review by examining the risk posed by the source category by examining the emissions actually emitted and allowed by sources in the source category.  As part of the risk review for a source category, EPA is not required to address emissions by sources outside the source category. However, in order to fully inform the RTR regulatory decisions in addition to conducting a comprehensive assessment of the risks associated with exposure to the HAP emitted by the source category, EPA supplements that with additional information that is available about other possible concurrent and relevant risks. The references in section 112(f)(2) to examining the risk to the individual cannot be taken out of the broader context of section 112. Section 112 is not a blanket provision requiring EPA to make health-based decisions across sectors and pollutants for specific individuals. Rather, it is focused on addressing source categories  -  first by establishing technology-based standards and then, under second step, reviewing whether emissions from that source category are sufficiently regulated to protect public health.
While the incorporation of additional background concentrations from the environment in our risk assessments (including those from mobile sources and other industrial and area sources of HAP emissions) could be technically challenging, they are neither mandated nor barred from our analysis. In developing the decision framework in the Benzene NESHAP, currently used for making residual risk decisions, the EPA rejected approaches that would have mandated consideration of background levels of pollution in assessing the acceptability of risk, concluding that comparison of acceptable risk "should not be associated with levels in polluted urban air" (54 FR 38044, 38061, September 14, 1989). Background levels (including natural background) are not barred from the EPA's ample margin of safety analysis, but the EPA may consider them, as appropriate and as available, along with other factors, such as cost and technical feasibility, in the second step of its section 112(f) analysis. This assessment excludes background contributions because the available data are of insufficient quality upon which to base a meaningful analysis. Further, our approach is also consistent with the approach we took regarding this issue in the Hazardous Organic NESHAP (HON) RTR (71 FR 76603, December 21, 2006), which the court upheld in the face of claims that the EPA had not adequately considered background concentrations. (NRDC v. EPA, 529 F.3d 1077 (D.C. Cir. 2008)).

The risk assessment modeling accounts for the effects of multiple facilities within the source category that may be in close proximity when estimating concentration and risk impacts at each block centroid. When evaluating the risks associated with a particular source category, we combine the impacts of all facilities within the same source category and assess chronic exposure and risk for all census blocks with at least one resident (i.e., locations where people may reasonably be assumed to reside rather than receptor points at the fence-line of a facility). The maximum individual risk considers the combined impacts of all sources in the category that may be in close proximity.  
The development of facility-wide risk estimates provides additional information about the 
potential cumulative risks in the vicinity of the RTR sources, as one means of informing potential risk-based decisions about the RTR source category in question. Facility-wide estimates can also provide information about other CAA section 112 source categories. We recognize that, because these risk estimates were derived from facility-wide emissions estimates which have not generally been subjected to the same level of engineering review as the source category emission estimates, they may be less certain than our risk estimates for the source category in question, but they remain important for providing context as long as their uncertainty is taken into consideration in the process. 

Comment (0147): One commenter states that EPA's unacceptability benchmark of 100-in-1 million cancer risk is far too high to be considered acceptable and some levels of other kinds of risk should also be recognized as automatically unacceptable. The commenter calls on EPA to recognize that cancer risk from a major industrial source category of toxic air pollution that is 100-in-1 million or less cannot be presumed safe or "acceptable." The commenter states that EPA effectively chose a number out of a hat that it would consider acceptable and has failed to revisit or update this number for the decades since. The commenter asserts that EPA should determine that all chronic non-cancer or acute risk above an HQ of 1 should be presumed to be unacceptable unless EPA can provide a reasoned explanation for finding otherwise. The commenter cites numerous examples of developments in public health policy since the presumptive level of acceptability was set to support their assertion. These developments include many advancements in policies related to children's health and environmental justice that have taken been issued by the executive branch since the 1990s. The commenter also states that EPA should change how the agency evaluates chronic and acute hazard indices, in which a risk number below 1 does not result in policy changes or standards, and where EPA treats risks above 1 as potentially acceptable. The commenter argues that EPA should instead factor in uncertainties and vulnerability factors that adjust the "acceptable level of risk." This is currently done under the Food Quality Protection Act when EPA uses factors to determine a Target Margin of Exposure and risks below this level warrant increased scrutiny and changes to allowable exposures.

Response: We generally draw no bright lines of acceptability regarding cancer or noncancer risks from source category HAP emissions and acknowledge that it is always important to consider the specific uncertainties of the emissions and health effects information regarding the source category in question when deciding exactly what level of cancer and noncancer risk should be considered acceptable. In addition, the source category-specific decision of what constitutes an acceptable level of risk should be a holistic one; that is, it should simultaneously consider all potential health impacts -- chronic and acute, cancer and noncancer, and multipathway -- along with their uncertainties, when determining the acceptable level of source category risk. The Benzene NESHAP (54 FR 38044, September 14, 1989) decision framework acknowledged this; in today's world, such flexibility is even more imperative, because new information relevant to the question of risk acceptability is being developed all the time, and the accuracy and uncertainty of each piece of information must be considered in a weight-of- evidence approach for each decision. This relevant body of information is growing fast (and will continue to grow even faster), necessitating a flexible weight-of-evidence approach that acknowledges both complexity and uncertainty in the simplest and most transparent way possible. While this challenge is formidable, it is nonetheless the goal of the EPA's RTR decision-making, and it is the goal of the risk assessment to provide the information to support the decision-making process.
The EPA disagrees with the commenter's suggestion that EPA should use uncertainties and vulnerability factors to adjust the "acceptable level of risk" similar to the approach used under Food Quality Protection Act (FQPA). In response to the 10X factor enacted by Congress in the FQPA to the EPA non-cancer reference value derivation, the Agency evaluated their methods for considering children's risk in the development of reference values. As part of the response, the U.S. EPA (i.e. the Science Policy Council and Risk Assessment Forum) established the RfD/RfC Technical Panel to develop a strategy for implementing the FQPA and examine the issues relative to protecting children's health and application of the 10X safety factor. One of the outcomes of the Technical Panel's efforts was an in-depth review of a number of issues related to the RfD/RfC process (US EPA, 2002). The most critical aspect in the derivation of a reference value pertaining to the FQPA has to do with variation between individual humans and is accounted for by a default uncertainty factor (UF-H) when no chemical-specific data are available. EPA reviewed the default UF for inter-human variability and found the EPA's default value of 10 adequate for all susceptible populations, including children and infants. The EPA also recommended the use of chemical-specific data in preference to default uncertainty factors when available (US EPA, 1994) and is developing Agency guidance to facilitate consistency in the development and use of data-derived extrapolation factors for RfCs and RfDs (US EPA, 2011). Additionally, the EPA also applies a database uncertainty factor (UF-D) which is intended to account for the potential for deriving an under protective RfD/RfC as a result of an incomplete characterization of the chemical's toxicity. In addition to the identification of toxicity information that is lacking, review of existing data may also suggest that a lower reference value might result if additional data were available. Therefore, an additional uncertainty factor is not needed in the RfC/RfD methodology because the currently available factors are considered sufficient to account for uncertainties in the database from which the reference values are derived (and does not exclude the possibility that these uncertainty factors may be decreased or increased from the default value of 10).
Comment (0147): One commenter recommended that, as part of the risk assessment, the EPA account for community level vulnerability by including factors to account for increased vulnerability based on demographic differences. The commenter says that EPA should revisit or update its presumed acceptable level of cancer risk considering information learned about overburdened community's vulnerability, biomonitoring and other data on adult body burdens of chemicals, and new information on ways to analyze and control the impacts of pollutants on human health. The commenter suggested that EPA look at, consider, and use the data available from the Center for Disease Control's Environmental Public Health Tracking ("EPHT") Program, the U.S. Agency for Toxic Substances and Disease Registry, state and local health agencies in its risk assessment.
The commenter also suggested that the EPA incorporate a more robust analysis of childhood and pre-natal exposure to environmental contaminants into its risk assessment method, citing related recommendations from the EPA's Children's Health Protection Advisory Committee and the EPA's Science Advisory Board (SAB). The commenter also cited methodology changes from California's OEHHA that may affect the estimation of children's risks.
The commenter claimed that the EPA underestimates risks to children and from pre-natal and early-life exposures and suggested that the EPA account for increased early-life susceptibility by using age-dependent adjustment factors for all carcinogens emitted by a source category, not just HAP included in the EPA's list of carcinogens that act by a mutagenic mode of action. 
The commenter stated that most of EPA's IRIS toxicity threshold values (reference concentrations and reference doses) used for chronic noncancer risk assessment do not incorporate the latest science on increased susceptibility of children and that EPA needs to account for early exposure and the greater risk to and susceptibility of children in its risk assessment.
The commenter added that, until the IRIS values fully account for the increased risk caused by early-life exposure to an emitted pollutant, EPA should use the OEHHA child-specific reference doses or benchmarks available to assess chronic non-cancer health risk from ingestion for certain pollutants. EPA should also assess such risk from inhalation by using standard methods to translate these values into child-specific reference concentrations.
The commenter suggested that, where child-specific reference values are unavailable, the EPA use a default or uncertainty factor to account for increased risk from early-life exposures for non-cancer risk, such as the 10X factor enacted by Congress in the Food Quality Protection Act (FQPA) of 1996, to fulfill the CAA's "margin of safety" requirement.
The commenter stated that the EPA must also assess the starting point or baseline overall health status of the affected individuals and communities using the best available data at a local and national level, including the baseline cancer levels, respiratory problems, and health problems associated with the toxic chemicals emitted by a source category. Doing so would be consistent with the 1999 Residual Risk Report. It would also follow EPA's own statements in the 2014 Second Integrated Urban Air Toxics Report that more work is needed to reduce excess cancer risks in urban areas that continue to face elevated risks.
Response: We disagree with the commenter's suggestion that EPA does not adequately consider children and developing fetuses when assessing risk. When the EPA derives exposure reference concentrations and unit risk estimates for hazardous air pollutants, including metal HAP, it also considers the most sensitive populations identified in the available literature, and importantly, these are the values used in our risk assessments. The EPA has an approach for selecting appropriate health benchmark values and in general, this approach places greater weight on the EPA-derived health benchmarks than those from other agencies. Additionally, the approach of favoring EPA benchmarks (when they exist) has been endorsed by the SAB and ensures the use of values most consistent with well-established and scientifically-based EPA science policy. The EPA is continuing to evaluate the most appropriate use for the Cal OEHHA child-specific reference doses. We note that there are currently no such values for HAP inhalation; therefore, their current utility would be limited to persistent and bioaccumulative HAP, which may be associated with non-negligible ingestion exposures. The evaluation of the appropriateness of these values in the context of the RTR Program and EPA science policy is necessary prior to using these child-specific reference doses (RfDs). 
With respect to exposure, we use a health-protective assumption of a 70-year exposure duration, in which populations are conservatively presumed to be exposed to airborne concentrations at their residence continuously, 24 hours per day for a full lifetime, including childhood.  
With respect to cancer, the EPA uses the age-dependent adjustment factor approach referred to by the commenter, but limits the application of age-dependent adjustment factors to carcinogenic pollutants that are known to act via mutagenic mode of action; in contrast, the OEHHA approach is to apply them across the board for all carcinogens, regardless of mode of action (MOA). In lieu of chemical-specific data on which age or life-stage specific risk estimates or potencies can be determined, default age-dependent adjustment factors can be applied when assessing cancer risk for early-life exposures to chemicals which cause cancer through a mutagenic MOA. With regard to other carcinogenic pollutants for which early-life susceptibility data are lacking, it is the Agency's long-standing science policy position that use of the linear low-dose extrapolation approach (without further adjustment) provides adequate public health conservatism in the absence of chemical-specific data indicating differential early-life susceptibility or when the mode of action is not mutagenicity. The basis for this methodology is provided in the 2005 Supplemental Guidance. 
The EPA disagrees with the commenter's suggestion that EPA should use uncertainties and vulnerability factors to adjust the "acceptable level of risk" similar to the approach used under Food Quality Protection Act (FQPA). In response to the 10X factor enacted by Congress in the FQPA to the EPA non-cancer reference value derivation, the Agency evaluated their methods for considering children's risk in the development of reference values. As part of the response, the U.S. EPA (i.e. the Science Policy Council and Risk Assessment Forum) established the RfD/RfC Technical Panel to develop a strategy for implementing the FQPA and examine the issues relative to protecting children's health and application of the 10X safety factor. One of the outcomes of the Technical Panel's efforts was an in-depth review of a number of issues related to the RfD/RfC process (US EPA, 2002). The most critical aspect in the derivation of a reference value pertaining to the FQPA has to do with variation between individual humans and is accounted for by a default uncertainty factor (UF-H) when no chemical-specific data are available. EPA reviewed the default UF for inter-human variability and found the EPA's default value of 10 adequate for all susceptible populations, including children and infants. The EPA also recommended the use of chemical-specific data in preference to default uncertainty factors when available (US EPA, 1994) and is developing Agency guidance to facilitate consistency in the development and use of data-derived extrapolation factors for RfCs and RfDs (US EPA, 2011). Additionally, the EPA also applies a database uncertainty factor (UF-D) which is intended to account for the potential for deriving an under protective RfD/RfC as a result of an incomplete characterization of the chemical's toxicity. In addition to the identification of toxicity information that is lacking, review of existing data may also suggest that a lower reference value might result if additional data were available. Therefore, an additional uncertainty factor is not needed in the RfC/RfD methodology because the currently available factors are considered sufficient to account for uncertainties in the database from which the reference values are derived (and does not exclude the possibility that these uncertainty factors may be decreased or increased from the default value of 10).
In summary, the estimated risks must also be considered in the context of the full set of assumptions used for this risk assessment. Our unit risk estimates for HAP are considered a plausible upper-bound estimate with an appropriate age-dependent adjustment factor; actual potency is likely to be lower and some of which could be as low as zero. Our chronic noncancer reference values have been derived considering the potential susceptibility of different subgroups, with specific consideration of children.  In addition, an extra 10X uncertainty factor is not needed in the RfC/RfD methodology because the currently available factors are considered sufficient to account for uncertainties in the database from which the reference values are derived.  After considering these and other factors, we continue to consider the risks associated with emissions from this source category to provide an ample margin of safety consistent with the Benzene NESHAP framework (54 FR 38044, 38061, September 14, 1989).
      
We are not able to determine the baseline health status of individuals or communities in a national rulemaking. Individual privacy issues as they relate to health records and the costs that would be associated with such an analysis make the analysis infeasible. Through the EPA's interim guidance on Environmental Justice and the Action Development Process, the agency is encouraging rule writers and policy makers to look at the whole range of factors that impact communities and population groups when crafting rules. The EPA is continuing to discuss and pilot approaches that are consistent with the agency's responsibilities regarding environmental justice as outlined in EO 12898. In the Urban Air Toxics Strategy Report to Congress we acknowledge that national rules and standards can address part of the risk to communities, but because the assessments did not include background risks or contributions to risk from sources outside the facilities more needs to be done at the community level with other tools available within the CAA and within state, local, and other federal programs. EPA is committed to our efforts to make a difference in communities of concern and developing an integrated strategy focusing work in communities with the most need for EPA's assistance. We have been and will continue to work in thousands of communities across the country.
      
Comment (0147): One commenter states that after assessing each type of risk, the EPA fails to combine all health risks. According to the commenter, leaving these risks segmented in its analysis prevents the EPA and the affected public from being able to perform a full assessment of the total health risk of the most-exposed person. The commenter notes that while the EPA did appropriately combine some risks (e.g., the combined chronic non-cancer risk for different chemicals that affect the same target organ), the EPA needs not only to sum this risk but also ensure that it assesses the synergistic, carcinogenic (and other) effects of different pollutants. The commenter describes specific deficiencies including the fact that the EPA does not create a combined acute risk number that would represent the total acute risk for all pollutants that act in a similar way on the same organ system or systems, similarly to the chronic TOSHI. The commenter is also concerned that the EPA does not create any metric to assess the total, combined health risk that could be used to assess the cumulative impact of all of the types of risks. The commenter states that even though the EPA found a combination of cancer and noncancer risk to create "unacceptable" risk, the EPA also must create a cumulative burden metric to assess and consider the total, cumulative burden of the most-exposed person. The commenter asserts that this is needed to assess the way different individual health risks (even if potentially "acceptable" on their own) combine to cause the full amount of "unacceptable" health risk for this source category. 

The commenter states that to perform a cumulative impact analysis, the EPA should combine current baseline emissions, exposures, and health impacts in addition to those of the specific source category the EPA is reviewing. The commenter states that the EPA should aggregate or add the emissions for the most-exposed communities coming from: (1) the source category (including all individual sources within it); (2) facility-wide risk from collocated sources outside of this category; and (3) all other sources of toxic air pollution in the area. The commenter states that the EPA must also draw on the OEHHA community assessment approach. 
At a minimum, until EPA develops a data-driven approach to comprehensively model cumulative risk or impacts from multiple sources, EPA must not treat multiple source exposure as a missing default or ignored amount of health risk. EPA must incorporate an explicit default or uncertainty factor to adjust the degree to which each individual source category is contributing to the total risk experienced by the most-exposed individuals. For example, wherever there is evidence that the source category is contributing pollutants on top of a history of other exposures or is contributing pollutants in addition to other source categories, the "unacceptable" level of cancer, noncancer chronic, and acute risk from the source category must be adjusted downward based on the number of other facilities contributing HAP exposure risks (such that no single source category could consume all of it, when the most-exposed person is exposed to many other source categories). For a source category in an area with up to 10 other HAP-emitting facilities, this default or uncertainty factor should equal at least 10, consistent with the common scientific use of this factor for other kinds of vulnerability.
Response: We have combined risk assessment results to the extent that it is appropriate to do so and as described in the previous response. We consider the effect of mixtures of carcinogens and use a target organ-specific hazard index (TOSHI) approach for our chronic noncancer assessments. We do not use a TOSHI approach for acute analyses nor do we combine the results of our inhalation and multipathway assessments. 
Regarding the comment that the EPA should assess the synergistic effects of pollutants, we do follow the EPA's guidelines for mixtures because most of the receptors in these assessments receive exposure to multiple pollutants rather than a single pollutant. As EPA typically describes in the risk assessment reports provided in the docket for our risk review rulemakings, we estimate the aggregate health risks associated with all the exposures from a specific source category combined. To combine risks across multiple carcinogens, our assessments use the mixtures guidelines[,]  default assumption of additivity of effects, and combined risks by summing them using the independence formula in the mixture guidelines.  
We do not sum results of the acute noncancer inhalation assessment to create a combined acute risk number that would represent the total acute risk for all pollutants that act in a similar way on the same organ system or systems (similar to the chronic TOSHI). The worst-case acute screen is already a conservative scenario. That is, the acute screening scenario assumes worst-case meteorology, peak emissions for all emission points occurring concurrently and an individual being located at the site of maximum concentration for an hour. Thus, because of the conservative nature of the acute inhalation screening and the variable nature of emissions and potential exposures, acute impacts were screened on an individual pollutant basis, not using the TOSHI approach. 
When evaluating the risks associated with a specific source category, we combine the impacts of all facilities within the same source category that are located in close proximity to each other and assess chronic exposure and risk for all census blocks with at least one resident (i.e., locations where people may reasonably be assumed to reside rather than receptor points at the fenceline of a facility). Thus, the maximum individual risk considers the combined impacts of all sources in the category that may be in proximity. The risk assessment modeling for this source category accounted for the effects of multiple facilities within the source category that may be in proximity when estimating concentration and risk impacts at each block centroid. 
We do not agree with the idea that we should apply an uncertainty factor related to the number of non-HCl facilities near a HCl facility. First and foremost, MIR is highly location-specific and neighboring facilities seldom have any significant contributions to the MIR posed from an individual facility. Second, given the geographic size of the facilities and their emissions, it is non-scientific to suggest neighboring facilities can have such an additive impact on the MIR. When EPA considers the use of an uncertainty factor in a risk analysis, it only includes such factors that have been developed based on scientific data that support their use.
Comment (0147): One commenter states that EPA's ample margin of safety analysis is flawed because it is set at the same level as the level of "acceptability" instead of some level significantly lower than the level of acceptability. As the D.C. Circuit held in interpreting similar language in § 7412(d)(4), "ample margin" means something additional, and EPA cannot simply collapse these stages or say it has evaluated this when plainly it has not. See, e.g., Sierra Club v. EPA, 895 F.3d 1, 13 (D.C. Cir. 2018) (holding that EPA had violated plain language because it has not "met the CAA requirement to include an ample margin of safety"). 

EPA failed to satisfy the Act's requirement because it did not consider or determine what such a margin is at all. Instead, EPA collapsed its acceptability analysis and the technology review  -  a separate rulemaking obligation and test  -  into this second and distinct step, without satisfying the statutory test. See 84 Fed. Reg. at 1583-84 (citing "the results of the technology review, risk assessment and other aspects of our MACT rule review"). Instead of determining what is needed to protect public health with an "ample margin of safety," EPA answers the wrong question: it addresses only "whether there are any cost-effective controls or other measures that would reduce emissions further to provide an ample margin of safety with respect to the risks associated with those emissions." Id. 

Here, EPA did not assess any "margin of safety," much less evaluate or determine what margin would be "ample" to ensure protection of public health. The term "ample" means "more than adequate," according to Webster's Dictionary, and the D.C. Circuit. The focus in section 112 is on the "individual most exposed" to a source category's emissions and requires EPA to provide an ample margin of safety not just for anyone, but for that individual, within the most-exposed communities around these sources.

The preamble illustrates that there is no such determination in the record. Instead, EPA simply repeated the same risk information it had found to be acceptable and stated that it "considers all health factors evaluated in the risk assessment and evaluates the cost and feasibility of available control technologies and other measures." 84 Fed. Reg. at 1583. The agency merely paid lip service to the statutory language, stating without showing, that it "conducted an analysis to determine if the current emissions standards provide an ample margin of safety." Id. As the D.C. Circuit recognized in the Vinyl Chloride decision, at the ample margin stage under § 7412, EPA must at least consider whether uncertainties regarding its health risk assessment require a stronger standard. As the court held recently in its brick kilns decision interpreting the same term ("ample margin of safety") in § 7412(d)(4), EPA must show that it has provided this margin in setting a standard, and it cannot lawfully fail to provide "any margin of safety at all." EPA has failed to consider whether uncertainties in its health risk assessment for this rule, of which there are many, as cited above, require any, much less an "ample margin" of safety than simply leaving the standards at the current level. The additional information EPA states that it evaluated at this stage was "the cost and feasibility of available control technologies and other measures." 84 Fed. Reg. at 1583.

At the ample margin stage, EPA again refuses even to address the fact that the chronic noncancer risk is above the HQ of 1 at which harm is known to occur. EPA has found, for HCl sources, that HCl sources create, at least, a TOSHI of 2, based on allowable emissions. This is two times the harm threshold. Even if EPA could justify finding that level of risk acceptable (which it has not, as it has not even addressed it in the acceptability determination, as discussed above), it still must consider how to assure an ample margin of safety to protect public health from the respiratory system harm this value reflects. Yet EPA does not discuss or find that it is providing any margin, much less an ample one, to protect people from the emissions causing this risk.

Response: We have not established under section 112 of the CAA a numerical range for risk acceptability for noncancer effects, nor have we determined that there is a bright line above which acceptability is denied. We have established that, as exposure increases above a reference level (as indicated by a HQ or TOSHI greater than 1), confidence that the public will not experience adverse health effects decreases and the likelihood that an effect will occur increases. For this source category, for actual emissions from facilities, the maximum TOSHI is well below 1, and, for allowable emissions, the maximum TOSHI is greater than 1 for two facilities (with a TOSHI of 2). We consider our analysis of risk from allowable emissions to be conservative and, as such, to represent an upper bound estimate of risk from emissions allowed under the NESHAP for the source category. The chronic ATSDR MRL, a noncancer dose-response value, is not a safety threshold or "harm threshold," as stated by the commenter. It is a level (or air concentration) below which we do not believe there would be adverse effects given a lifetime of exposure to this concentration. MRLs are set below levels that, based on current information, might cause adverse health effects in the people most sensitive to such substance-induced effects. Thus, exposure to a level above the MRL does not mean that adverse health effects will occur. It means that there is greater potential to see an effect than at a lower concentration. 

We further note that we generally draw no bright lines of acceptability regarding cancer or noncancer risks from source category HAP emissions, and that it is always important to consider the specific uncertainties of the emissions and health effects information regarding the source category in question when deciding exactly what level of cancer and noncancer risk should be considered acceptable. In addition, the source category-specific decision of what constitutes an acceptable level of risk is a holistic one, where all potential health impacts  -  chronic and acute, cancer and noncancer, and multipathway  -  are considered, along with their uncertainties, when determining the acceptable level of source category risk.

Margin-of-safety determinations are conducted separately, in accord with the two-step framework set forth in the Benzene NESHAP and the en banc opinion in Vinyl Chloride. See not only Vinyl Chloride, 824 F. 2d at 1165, 1166 but NRDC v. EPA, 902 F. 2d  962, 973-74 (D.C. Cir. 1990) (distinguishing the NAAQS process, whereby the margin of safety analysis is incorporated as part of the standard without a two-step analysis, from residual risk determinations). Using that framework, with its consideration of health risks and their associated uncertainties, costs, cost-effectiveness, technological feasibility, and other factors set out in the Benzene NESHAP, we did not identify any additional controls that would need to be implemented to ensure an acceptable level of risk with an ample margin of safety. We have concluded that the conservative estimate of a TOSHI of 2 does not warrant revision to the standard. We do not expect facilities to operate at or near allowable levels, let alone at those levels at all times as was estimated in our assessment. We have determined that the existing requirements provide an ample margin of safety. 
Comment (0147): One commenter claims that EPA attempts to rely on the technology review under section 7412(d)(6) as a justification not to strengthen the standards to protect public health under section7412(f)(2) and that it is unlawful and arbitrary to do so. These are separate statutory tests and obligations that EPA may not lawfully collapse into one. As the D.C. Circuit explained, these are "two distinct, parallel analyses: a recurring `technology review' under section 112(d)(6) and a one-time `risk review' under section 112(f)(2)." Nat'l Ass'n of Surface Finishing v. EPA, 795 F.3d 1, 5 (D.C. Cir. 2015) (explaining that "separately" from the technology review, in the risk review, "EPA addresses . . . lingering public health risk that the initial standard did not eliminate"). EPA's attempt to rely on what it describes as a lack of "developments" under § 7412(d)(6) as a justification for not requiring additional standards under § 7412(f)(2) shows how inadequate and unlawful its "ample margin" determination is. It is not an ample margin of safety determination at all; it is just a recycled "no developments" determination.
Response: The EPA's determination of whether the current standards provide an ample margin of safety was based on a number of factors, which included the health factors evaluated in the risk assessment as well as the cost and feasibility of available control technologies that could be applied to further reduce the risks. As stated in the preamble for the proposal, our risk analysis indicated that the risks from the source category are low for both cancer and noncancer health effects, and, therefore, any risk reductions from further control options, had any been identified, would result in minimal health benefits (84 FR 1584).

Comment (0147): One commenter states that EPA must, at least, acknowledge the facility-wide cancer risk of 600-in-1 million is unacceptable at the ample margin stage, by comparison with its own Benzene Rule-based presumptive benchmark of 100-in-1 million. Failing to do so is ignoring its own longstanding recognition that this evidence is important at least at the ample margin of safety stage, to use the facility-wide risk assessment to protect the most-exposed person. EPA should take public comment on a revised proposal that includes pollution limit reductions for ethylene oxide, trichloroethylene (TCE), Cl, HCl, and all other pollutants contributing to these high risks.

Response: While EPA does consider facility-wide risk as one aspect of its AMOS analysis, we note that the requirements of section 112(f)(2) are linked to promulgation of standards for the category of sources at issue  -  in this case HCl Production. There are only two HAP emitted from HCl Productions source category  -  HCl and chlorine  -  and neither of these are a known or suspected carcinogen and thus neither contribute to the 600-in-1 million facility wide cancer risk.  Thus, there are not limits that EPA can establish for the HCl Production source category under review in this rulemaking that would affect the facility-wide cancer risk.

Comment (0147): EPA has violated notice-and-comment because it has failed to provide essential underlying data it used to reach the facility-wide risk determinations. 42 U.S.C. § 7607(d)(3)-(7), (h). For example, it states that it used NEI data, but it does not provide the emissions estimates it used for any of the facilities. It gives the emissions totals for the HCl sources, but not for the facility-wide risk part of the assessment. It also gives a conclusory chart showing what it describes as the source category's contribution to the facility-wide risk, in the RRA. But it provides insufficient evidence or information showing how it reached those conclusions. Failing to provide the calculations and underlying data used to create the facility-wide risk assessment makes it impossible for the public to meaningfully comment on this. Because these risks are so high, and EPA has cited significant excess cancer cases coming from these facilities, commenters need this information and are prejudiced by EPA's refusal to disclose in the public docket.

Response: We used the best and most robust facility-specific HAP emissions inventory available to us, which was the 2014 National Emissions Inventory and state databases of Title V permits, in performing the risk analysis for the proposed rule. All model input files (containing all the data) used in the risk modeling, for both the source category and the facility-wide analyses, were made available for public comment and review in the docket at proposal (docket ID: EPA-HQ-OAR-2018-0417-0002 through EPA-HQ-OAR-2018-0417-0019). The model input files contained all facilities, all HAP emissions, and all emission locations for the source category and the facility-wide analyses.

Comment (0147): One commenter disagrees with EPA's use of a "low confidence" IRIS value of 0.02mg/m[3] to assess health risk from hydrogen chloride (HCl). Instead, the commenter argues that the 2000 Cal. EPA OEHHA value of 9 μg/m[3] (0.009 mg/m[3]) should be used to assess chronic non-cancer risk. The commenter explains that the IRIS value was one that IRIS had stated it planned to update when additional data became available, but that update has not occurred, and that, in such circumstances, EPA's own prioritization policy directs it to use the best available science, which would include the Cal. EPA OEHHA value.
The commenter states that, by not using the Cal. EPA OEHHA value, EPA underestimates the chronic non-cancer risk from HCl. Additionally, the commenter asserts that EPA did not attempt to evaluate the cancer risk for HCl, and that EPA has not conducted a "complete evaluation and determination under" the "IRIS program for evidence of human carcinogenic potential." The commenter indicated that the D.C. Circuit has held that EPA must analyze the carcinogenic potential of HCl to "base its findings" of no carcinogenic risk "on substantial evidence," and that, therefore, underestimating HCl emissions impacts EPA's findings of chronic non-cancer and cancer risk. The commenter argued that ignoring the potential for carcinogenic risk from HCl is arbitrary.
Response: For the CAA section 112(f)(2) risk reviews, we use dose-response information that has been obtained from various sources. The dose-response information is prioritized according to (1) conceptual consistency with EPA risk assessment guidelines and (2) level of public and peer review received. The prioritization process is aimed at incorporating into our assessments the best available science with respect to dose-response information. The recommendations are based on the following sources: 1) U.S. EPA, 2) ATSDR, and 3) California EPA. In selecting the appropriate chronic noncancer dose-response value for HCl for use in the risk assessment, in this case, the 1995 EPA IRIS RfC, we followed this prioritization approach, and we reviewed newer values as part of that process. The 1995 EPA IRIS RfC for HCl of 2 x 10[-2] mg/m3 was based on the following studies: Sellakumar et al., 1985 and Albert et al., 1982. The ATSDR has not established a chronic noncancer dose-response value for HCl. The 2000 California EPA established a chronic reference exposure level (REL) of 9 μg/m[3] (9 x 10[-3] mg/m[3]) based on Sellakumar et al., 1985. California EPA did not use newer data than the EPA in establishing their chronic REL for HCl. The difference in the dose-response values is due to differences between the EPA and CalEPA methodologies. The EPA chose to rely on its own methodology for the chronic dose-response value for HCl.
In the process of assessing the toxicity of a substance, if enough data from relevant studies and of acceptable quality do not exist, the EPA IRIS program does not establish a dose-response value. In assessments completed prior to the 2000s, the EPA assigned confidence ratings (low, medium, high) to the resultant value (e.g., RfC) and the underlying principal study and full database. The ratings assignment was based generally on the extent and robustness of the database (e.g., number and types of different toxicity test studies, quality of the studies, suitability of the test results for use in dose-response assessment). For HCl, the available data were judged adequate for establishment of an RfC. In recognition of limitations in the overall database and the principal study, the resultant RfC for HCl was given a confidence rating of low.   
The EPA IRIS program has not assigned a carcinogenicity weight of evidence classification to HCl. Little research has been conducted on the carcinogenicity of HCl. (79 FR 75639) There are limited studies on the carcinogenic potential of HCl in humans. Of these, two occupational studies failed to separate potential exposure of HCl from exposure to other substances shown to have carcinogenic activity and are therefore not appropriate to evaluate the carcinogenic potential of HCl (Steenland et al., 1988, Beaumont et al., 1986)[,] Another occupational study failed to show evidence of association between exposure to HCl and lung cancer among chemical manufacturing plant employees (Bond et al., 1991). (80 FR 65488). Consistent with the human data, chronic inhalation studies in animals have reported no carcinogenic responses after chronic exposure to HCl (Albert et al., 1982; Sellakumar et al., 1985).[,](80 FR 65488) Hydrogen chloride has not been demonstrated to be genotoxic. The genotoxicity literature consists of two studies showing false positive results potentially associated with low pH in the test system (Morita et al., 1992; Cifone et al., 1987).[,] (80 FR 65488).

The International Agency for Research on Cancer (IARC) also classifies agents (chemicals and biologics) as to carcinogenicity. The IARC classifies HCl as "not classifiable as to its carcinogenicity to humans." Of the more than 1,000 agents classified by IARC, no agents are classified as "probably not carcinogenic (IARC) to humans." 
The D.C. Circuit decision cited by the commenter was regarding the basis for setting a health-based emission limit for HCl under section 112(d)(4) of the CAA, and not for making a determination about risk acceptability under section 112(d)(f) of the CAA. 
Comment (0122): A commenter is pleased that EPA has recognized the importance of considering the impact of emissions from all HAP-emitting operations in a facility to determine the facility-wide risks, rather than focusing solely on the source category that is the subject of the regulation. In this case, it is especially important that EPA also considered emissions of ethylene oxide (EtO) and trichloroethylene and commenters urge EPA to take additional steps to address those risks in additional actions.

Response: We acknowledge the support of our findings under CAA section 112(f) and section 112(d)(6) with respect to the Hydrochloric Acid Production source category. Regarding EPA's consideration of EtO and trichloroethylene, we note that sources in the HCl Production source category do not emit those HAP although some non-category sources co-located at some of the same facilities do emit those HAP. EPA would consider how and whether to address emissions from non-category sources co-located with HCl Production facilities in future 112(d)(6) or 112(f)(2) actions required for such sources, as appropriate.

Comment (0147): One commenter states that EPA cannot determine that risks from HCl production facilities are acceptable by ignoring the facility-wide cancer risks that exceed EPA's presumptive benchmark for unacceptability by a factor of six. The commenter states that while EPA presents the results of the facility-wide risk assessment, the agency does not discuss these results when determining risk acceptability. 

Response: The development of whole-facility risk estimates provides additional information about the potential cumulative risks in the vicinity of the RTR sources, as one means of informing potential risk-based decisions about the RTR source category in question. We recognize that, because these risk estimates were derived from facility-wide emissions estimates which have not generally been subjected to the same level of engineering review as the source category emission estimates, they may be less certain than our risk estimates for the source category in question, but they remain important for providing context as long as their uncertainty is taken into consideration in the process. Section 112(f) requires the EPA to perform a review of whether there is risk remaining from emissions from a source category after promulgation of the technology-based MACT standards for that source category. To this end, the EPA conducts a comprehensive assessment of the risks associated with exposure to the HAPs emitted by the source category and supplements our ample margin of safety analysis with additional information that is available about other possible concurrent and relevant risks (i.e., whole-facility). 

Comment (0144): A commenter states that the cancer risks due to hydrochloric acid production facilities (including co-located chemical plants and other emission sources under common control) are unacceptable and EPA must address these risks. The commenter also asks EPA to reduce the known acute and chronic noncancer risks posed by these sources (including to the respiratory, immunological, kidney, developmental, neurological, reproductive, and liver organ systems), to assure an ample margin of safety to protect public health. 

Response: We generally draw no bright lines of acceptability regarding cancer or noncancer risks from source category HAP emissions and acknowledge that it is always important to consider the specific uncertainties of the emissions and health effects information regarding the source category in question when deciding exactly what level of cancer and noncancer risk should be considered acceptable. In addition, the source category-specific decision of what constitutes an acceptable level of risk should be a holistic one; that is, it should simultaneously consider all potential health impacts -- chronic and acute, cancer and noncancer, and multipathway -- along with their uncertainties, when determining the acceptable level of source category risk. The Benzene NESHAP (54 FR 38044, September 14, 1989) decision framework acknowledged this; in today's world, such flexibility is even more imperative, because new information relevant to the question of risk acceptability is being developed all the time, and the accuracy and uncertainty of each piece of information must be considered in a weight-of- evidence approach for each decision. This relevant body of information is growing fast (and will continue to grow even faster), necessitating a flexible weight-of-evidence approach that acknowledges both complexity and uncertainty in the simplest and most transparent way possible. While this challenge is formidable, it is nonetheless the goal of the EPA's RTR decision-making, and it is the goal of the risk assessment to provide the information to support the decision-making process. Nevertheless, while EPA does consider facility-wide risk as one aspect of its AMOS analysis, we note that the requirements of section 112(f)(2) are linked to promulgation of standards for the category of sources at issue  -  in this case HCl Production. There are only two HAP emitted from HCl Production source category  -  HCl and chlorine  -  and neither of these are a known or suspected carcinogen and thus neither contribute to the 600-in-1 million facility wide cancer risk. Thus, there are not limits that EPA can establish for the HCl Production source category under review in this rulemaking that would affect the facility-wide cancer risk. Furthermore, we explained in the preamble for the proposed rule why EPA has determined risks due to HAP emissions from the source category are acceptable. 

Comment (0147): One commenter states that EPA is legally required to perform a multipathway (i.e., non-inhalation-based) risk assessment as part of this rulemaking, in addition to assessing inhalation risks. EPA's assessment here is incomplete, unlawful, and arbitrary in the ways detailed below. The net result is a significant underestimate of health risks from these facilities and therefore, flawed decision-making in EPA's proposed decision not to set residual risk standards for these facilities. 

EPA performed no multipathway risk assessment because it proposes to find that no persistent bioaccumulative toxicants ("PBTs") are emitted. However, even if no PBTs are emitted by HCl sources, EPA must evaluate multipathway risk for the facility-wide risk assessment. 

As a result of not completing the multipathway risk assessment at least for the facility-wide emissions, EPA did not come up with a multipathway risk value for cancer or chronic non-cancer risk at all. So affected communities do not have information they need to evaluate that value, or the total. EPA's failure to perform a facility-wide multipathway risk assessment is yet another way in which EPA has underestimated health risk for the most-exposed individual.

Response: EPA conducts multipathway assessments using actual source category emissions of persistent and bioaccumulative HAP (PB-HAP). This source category emits only two HAP, chlorine and hydrochloric acid, and neither are PB-HAP; therefore, a multipathway analysis was not warranted. 
EPA does not conduct multipathway risk assessments for facility-wide emissions for two reasons: the uncertainty of facility-wide emissions and the conservative nature of the multipathway screening assessment. Typically, estimates of facility-wide emissions are more uncertain than estimates of actual emissions. Facility-wide emissions estimates do not undergo the same level of engineering review as the source category emission estimates and carry a higher level of uncertainty. That uncertainty, combined with the conservative nature of the multipathway screening assessment, could lead to results that are too uncertain and conservative to be of practical use. 
Comment (0134): One commenter notes that the maximum whole facility risk due to emissions from outside the source category is listed as 600-in-1 million in the preamble to the proposed rule while the value in the residual risk assessment report is 551-in-1 million. The commenter requests that EPA change the value in the Federal Register to match the value presented in the risk assessment. 

Response: Final risk results provided in the preamble and main body of the risk report are presented with the correct number of significant figures based on standard rules of mathematics. Most of the dose-response values used in our risk assessments are accurate to one significant figure and, therefore, our final risk results are typically also presented to one significant figure. The value the commenter points to ("551-in-1 million") is presented in the Appendix of the risk report for reference and is before final adjustments for significant figures are made.

Comment (0147): One commenter states that EPA has arbitrarily decided to issue the HCl proposal before proposals for other co-located sources that are also subject to court-ordered deadlines for RTR reviews. The commenter suggests that EPA coordinate rulemakings addressing chemical manufacturing to address and reduce cancer and other health risks for people facing the combined or cumulative impact of the pollution from co-located sources. 

Response: The EPA disagrees that the timing of its issuance of the HCl Production RTR proposal is in any way "arbitrary." The commenter correctly notes that the Agency was obligated to issue proposed and final rules addressing a number of source categories subject to court-ordered deadlines at the same time the HCl Production RTR proposal was being developed. Contrary to the commenter's assertion, managing workload to stagger proposed and final rule signature dates is responsible project management, rather than an arbitrary choice. As discussed elsewhere in this document, the EPA evaluated risks from co-located sources of HAP emissions outside the HCl Production source category and presented those analyses in this rulemaking. 

Acute Risk Assessment

Comment (0122, 0147): One commenter notes that they have expressed concerns in the past with EPA's use of Acute Exposure Guideline Levels (AEGLs) or Emergency Response Planning Guidelines (ERPGs) values to address acute exposures in the residual risk assessments and takes issue with EPA using them for those purposes in this proposal. These limits were developed for accident release emergency planning and are not appropriate for assessing daily human exposure scenarios. In the December 2002 EPA document, "A Review of the Reference Dose and Reference Concentration Processes," the agency stated that the primary purpose of the AEGL program is to develop guidelines for once in-a-lifetime short-term exposures to airborne concentrations of acutely toxic chemicals. They are not meant to evaluate the acute impacts from routine emissions that occur over the life of a facility. Unlike the reference concentrations (RfCs) for chronic exposures, the AEGLs and ERPGs do not include adequate safety and uncertainty factors and cannot be relied upon to protect the public from the adverse effects of exposure to toxic air pollutants. The use of AEGLs or ERPGs in residual risk assessments is not appropriate and does not ensure that public health is adequately protected from the acute impacts of HAP exposure. The commenter is gratified to see that EPA has included the use of the California Reference Exposure Levels (RELs) to address acute exposures in the residual risk assessments and we continue to urge EPA to use the RELs for these assessments.

Response: We disagree that AEGLs and ERPGs should not be relied on in our acute screening analyses. The dose-response values EPA considers for acute screening analyses depend on which acute dose-response values are available for the HAP emitted. The EPA generally considers the available acute RELs, AEGLs, and ERPGs to screen for potential acute health hazards. The EPA uses acute RELs when they are available and meet our criteria for acceptability. For two HAP (chlorine and hydrochloric acid) emitted from this source category we used all of the available dose-response values, which included the acute RELs, AEGLs and ERPGs, in our acute screening analyses.  

For the acute screening assessment for this source category, we used the health protective assumptions that every process at the facility releases its peak hourly emissions at the same hour, that the worst-case dispersion conditions occur at that same hour, and that an individual is present at the location of maximum HAP concentration for that hour. The EPA only uses the AEGLs and ERPGs for acute 1-hour inhalation screening assessments. We do not use AEGLs and ERPGs to assess repeat exposures over a lifetime or long-term exposures.

Comment (0147): One commenter says EPA calls its method of calculating acute risk a "worst-case" scenario approach but claims EPA's approach does not come close to modeling the actual "worst case" scenario because each such factor is too low; EPA's method  ignores all malfunctions which exceed the standards. The commenter suggests that  EPA's recognition that it is appropriate to use factors to assess higher emissions shows that the agency could simply use a more accurate factor to account for malfunctions for acute and other types of health risk to close the gap and respond appropriately to SAB's criticism of its current method. EPA regularly uses statistical methods and probability factors, which are readily available tools that EPA can also use to assess health risk due to malfunctions to set clean air standards.

Response: The EPA disagrees with the commenter's assertion that the acute risk scenario is not sufficiently conservative. For the acute screening assessment for this source category, we used the health protective assumptions that every process at the facility releases its peak hourly emissions at the same hour, that the worst-case dispersion conditions occur at that same hour, and that an individual is present at the location of maximum HAP concentration for that hour. We disagree that we should consider malfunction emissions in assessing the acute risk. The purpose of the risk assessment is to ascertain whether the emissions allowed after promulgation of the MACT protect public health with an ample margin of safety. Emissions that exceed the standard are not allowed by that standard and should not be considered in determining whether the standard is protective enough to limit risk. We further note that, as explained in the preamble to the proposed rule, it is not possible to model malfunction emissions in the risk assessment because by nature they are infrequent and unpredictable.

Emissions Estimates

Comment (0147): One commenter states that EPA's risk analysis based on the National Emissions Inventory significantly underestimates emissions. The emission inventory is largely calculated from emission factors and engineering judgment. It is well-documented that emission factors underestimate emissions for a variety of reasons including inherent bias in the factors themselves and the inability to account for equipment malfunctions and environmental conditions. The agency must adjust the emissions inventory using these same conclusions from the technology review and the large body of scientific evidence, which show that emissions factors underestimate emissions, to ensure that the inventory better represents reality and reflects actual emissions. 

The tests used to develop emission factors are intentionally conducted on new equipment operating under normal conditions because emission factors are formulas that attempt to estimate long-term average emissions. EPA itself notes that "[p]arameters that can cause short-term fluctuations in emissions are generally avoided in testing and not taken into account in test evaluation." Further, "[s]ources often are tested more frequently when they are new and when they are believed to be operating properly, and either situation may bias the results." The incorporation of this erroneous assumption in the development of emission factors significantly distorts emissions data in two significant ways  -  neither the emissions that are generated during upset and SSM events, nor the increased emissions that result from poor maintenance of equipment are accounted for in a facility's reported emissions and emissions inventories.

Response: We used the best and most robust facility-specific HAP emissions inventory available to us, which was the 2014 National Emissions Inventory and state databases of Title V permits, in performing the risk analysis for the proposed rule. All model input files (containing all the data) used in the risk modeling, for both the source category and the facility-wide analyses, were made available for public comment and review in the docket at proposal (docket ID: EPA-HQ-OAR-2018-0417-0002 through EPA-HQ-OAR-2018-0417-0019). The model input files contained all facilities, all HAP emissions, and all emission locations for the source category and the facility-wide analyses. The commenter did not provide data corrections to the model input files cited above during the public comment period. In this rule, as in all of our RTR rulemakings, the EPA relied on the best data available to the agency at the time of the assessment.  

Comment (0147): EPA must account for upset emissions in modeling for risk assessments because if EPA has knowledge that a certain amount of excess pollution per year routinely escapes facilities despite existing requirements, it must determine how these reasonably foreseeable "upset emissions" impact public health. EPA must also consider whether changes in the existing standards could mitigate the problem and better protect the public with an ample margin of safety. It is not enough simply to point to the availability of enforcement suits by citizens or government entities.

Malfunction, upset, or so-called "accidental" events increase emissions and thereby pose increased health risks which EPA must consider. Although EPA has used the term "accidental," many such emission spikes occur as a result of events that can be prevented, and thus Commenters disagree that they are actually "accidental." Where control equipment fails for any reason, emissions could be at least 100 times greater than usual (e.g., in the circumstance where a control device has 99% efficiency, such that an uncontrolled release would cause 100 times the usual amount of emissions). 84 FR at 1585. Ignoring such emissions is an unlawful and arbitrary example of the problem of ignoring health risk in the agency's assessment. The higher emissions caused by malfunction or upset emissions can accumulate and combine to increase public health impacts and lead to much higher risks for community members than those they face daily.
      
EPA has at least once assessed chronic and acute risk from "non-routine" or malfunction emissions  -  in the 2015 Petroleum Refinery Sector Rule. Refusing to assess and consider the additional health risks from these incidents here would thus also be arbitrary as there can be no rational basis for recognizing the need to evaluate them in one rule but not another.
      
Response: We disagree with the commenter that emissions that exceed standards should be part of the risk analysis. The main purpose of the risk review is to evaluate whether the emission limits  -  the "standards promulgated pursuant to subsection (d)" not the non-compliance with those standards  -  should be made more stringent to reduce the risk posed after compliance with the underlying MACT standard. To the extent that a source is violating the underlying MACT standard, no tightening of the emission standard as a result of the residual risk review will avoid or mitigate such violations. In other words, a source that is violating the MACT emissions standard promulgated under section 112(d) would not be any more likely to be able to avoid such violations and comply with a different presumably more stringent standard promulgated under section 112(f). Such events are violations and subject to enforcement by the EPA, the States, or citizens, and an action for injunctive relief is the most effective means to address violations whether or not they are caused by malfunctions if an emissions event poses a significant health or environmental risk. Moreover, as explained in the preamble to the proposed rule, it is not possible to model malfunction emissions in the risk assessment because by nature they are infrequent and unpredictable.

Regarding the comment that the EPA assessed chronic and acute risk from "non-routine" or malfunction emissions in the 2015 Petroleum Refinery Sector Rule, the EPA notes that the final Petroleum Refinery Sector Rule set a work practice standard under section 112(d)(2) and (3) for a narrow category of non-routine emissions events from pressure relief devices and flares. A screening analysis was performed using available information collection request data to evaluate the impacts of those events. The HCl Production NESHAP does not have a similar provision addressing non-routine emissions events, so the EPA did not need to consider emissions from those events as part of the risk analysis.

Startup, Shutdown, and Malfunctions (SSM)

Comment (0134, 0145): Two commenters agree with EPA's proposal that facilities should have 180 days after the date of publication of the final rule in the Federal Register to comply with amendments related to startup, shutdown, and malfunction. The commenters support the language in 40 CFR part 63.9005(a)-(c), (d)(4)-(6), and Table 7 to subpart NNNNN concerning implementation of the SSM changes.

Response: The EPA acknowledges the support. 

Comment (0134, 0145): Two commenters recommend that facilities use root cause analysis and corrective actions to address emissions during periods of malfunctions, such as emissions from discharge of a pressure relief device or malfunction of an emission control device. The commenters provided recommended regulatory language to add the root cause analysis and corrective action requirements. The commenters note that some facilities in the source category already use root cause analysis to ensure malfunctions are not repeated. 

The commenters strongly recommend that EPA consider such a work practice for malfunction events in the HCl NESHAP based on this approach. Specifically, a commenter recommends a two-part work practice standard that 1) creates a timeframe for facilities to attempt and finalize a repair and 2) require a root cause analysis to be completed as soon as possible. The commenter recommends a timeframe in which to address malfunction events after they are discovered that would be similar to the leak detection and repair timeframes that require leaks to be repaired within 5 days on a first attempt with a final repair to follow within 15 days. The commenter suggests EPA could require facilities to address potential leaks and other malfunction events, once discovered, in this same timeframe.  For the second part of the work practice, the commenter recommends regulatory text to add to 40 CFR 64.9040 describing the root cause analysis and schedule for completion. 

Response: The EPA is not finalizing a separate standard for periods of malfunction, although the EPA may establish a standard for periods of malfunction if the available information supports a separate standard in the future. In this case, we did not receive sufficient information on the performance of the best performing sources during periods of malfunction on which to base a standard. Absent sufficient information, it is not reasonable at this time to establish a work practice standard for malfunctions for this source category. We will continue to review this issue to determine if any new data become available in the future. 

Comment (0134): One commenter states that the proposed regulatory text in 40 CFR part 63.9040(e) and Table 7 to subpart NNNNN imply that after the 180 days allowed for facilities to come into compliance, the Administrator will no longer have discretion to determine whether deviations that occur during a period of SSM are violations. The commenter recommends that EPA add language to 40 CFR 63.9040(e) that affirms the Administrator's discretion to determine whether a facility is operating in compliance with regulatory requirements based on information available to the Administrator. 

Response: We have not made the change requested by the commenters to the regulatory text. The EPA has removed the exemptions for periods of SSM, consistent with court decisions. In any enforcement action involving a facility's non-compliance with regulatory requirements during a period of malfunction, the court or administrative decisionmaker can determine whether the facility was or was not in compliance with regulatory requirements. EPA also notes that the language at 63.9(e)(1)(i), which no longer applies, was within the context of operation consistent with good air pollution control practices during periods of SSM ("general duty"). With this final action, EPA is codifying a new general duty requirement at 63.9005(b), which includes language similar to 63.6(e)(1)(i), regarding how the Administrator will determine compliance with the operation and maintenance requirements of that section. 

Comment (0147): One commenter states that although EPA appropriately does not propose any such attempt at a standard here, it still claims discretion "to set standards for malfunctions." 84 Fed. Reg. at 1585. EPA has not cited and can cite no statutory language granting it authority to set such standards, because it has none. Rather, EPA has only the discretion provided by the Act and delegated by Congress. See, e.g., Clean Air Council v. Pruitt, 862 F.3d 1, 9 (D.C. Cir. 2017). The relevant statutory language denies, rather than gives, EPA authority to set malfunction-based standards or exemptions. See 42 U.S.C. §§ 7412(d), (h), 7602(k). Community, environmental, and environmental justice groups petitioned the agency for administrative reconsideration and filed a petition for review of the only such standard it has tried to create (in the Refinery Sector Rule the proposal cites). Their reconsideration petition and comments filed in support of that petition and offered at the November 2016 public hearing have shown that the Refinery Rule malfunction exemption is unlawful and arbitrary and should be removed from the standards. EPA still has not acted on their reconsideration petition. The D.C. Circuit has held the case in abeyance, waiting first to see if EPA corrects its own errors before reviewing the rule. As the malfunction exemption in that rule remains under a cloud of substantial controversy and is unlawful and arbitrary, it provides no support for any other similar proposal. EPA may not contravene core Clean Air Act requirements to assist industry sources that prefer not to take basic steps to reduce toxic air pollution, rather than rewarding industry leaders that do so and reach the lowest achievable limits.

Response: As the commenter notes, EPA did not propose and is not finalizing a standard for any type of malfunction. Thus, this comment is not relevant to the current rulemaking.

 Maintenance Activities

Comment (0147): One commenter argues EPA should not set a work practice standard for emissions that occur during maintenance activities as Dow Chemical has requested. The commenter claims that this would be an "exemption" that allows facilities to evade numerical emission limits during certain time periods. According to the commenter, the EPA is taking comment on Dow's proposal, and the "regulatory text it submitted"; EPA should reject Dow's proposed regulatory changes as unlawful and arbitrary. 84 Fed. Reg. at 1589 (citing Dow correspondence, EPA-HQ-OAR-2018-0417). There are multiple emails between EPA and Dow Chemical in the record, including communication where Dow is seeking a work practice standard instead of an emission limit; however, Commenters were unable to find any actual "regulatory text" that EPA is considering or proposing.

Dow Chemical's request for a "maintenance vent" work practice standard is based solely on the fact that EPA must remove the current exemption in the standards for "startup, shutdown, and malfunction" periods. Id. It is not based on the fact that vents are affiliated with different equipment or any other differences that might justify a different standard. It appears to be for the purpose of avoiding the numerical emission limits. 

EPA may not set a work practice standard without meeting § 7412(h)(1)-(4) of the Act, which requires a determination that it is not feasible or practicable to measure or enforce a numerical limit. EPA cannot meet that test because it has already recognized the need to set numerical limits for these sources and has done so. EPA has found such standards can and should be set, and no data Dow has provided contradicts that. There is no evidence in the record that a facility cannot reduce operations or take other steps needed to avoid any violations or exceedances of the numerical standards during periods of maintenance. Allowing a facility to avoid taking necessary preventive measures, and instead allow unlimited amounts of HAPs to be released under the guise of "maintenance," would violate the Act and be arbitrary. This type of provision is vulnerable to abuse by a facility who may have a malfunction that then requires "maintenance," and attempt to avoid recognizing an exceedance of the numerical limit by labelling the release as through a "maintenance vent."

Further, to the extent there are pressure relief devices or other valves that Dow seeks to bring under a new "maintenance vent" requirement, EPA must make clear that uncontrolled releases of HAPs to the atmosphere are unlawful, and prohibit them as EPA has done in other chemical plants rules.

The Act does not allow any uncontrolled releases of HAPs from a major source, and EPA must ensure that there is a numerical limit that applies (which meets the test pursuant to § 7412(d)(2)-(3), and (f)(2)), or simply prohibit any such uncontrolled releases. There is no reason why a maintenance vent and/or valve cannot be vented to a control device, as EPA is well aware and has required in prior rules for similar sources. Setting standards as the Act requires provides an incentive to perform regular maintenance and take preventative steps needed to ensure all equipment is operating properly, and so there will not be the need for unanticipated, high releases of pollution for which the facility's control devices are not prepared or equipped.

EPA has not provided any proposed regulatory text or rationale of its own or reasoned explanation based on any data in the record for such a maintenance provision, and therefore it could not finalize one without providing notice and comment on such a rationale. 42 U.S.C. § 7607(d)(3)-(7), (h).

EPA may not set an exemption from the emission standards by trying to cloak this as a work practice standard. It would undermine and weaken the numerical emission standards to allow industry to label a vent "maintenance," and avoid the applicability of those numerical limits, during periods of time it chooses. This would also violate the Act's requirement that emission standards must be "continuous" and apply at all times. 42 U.S.C. § 7602(k); Sierra Club v. EPA, 551 F.3d 1028 (D.C. Cir. 2008). As EPA is well aware, the D.C. Circuit has recognized that EPA's attempt to create a new type of malfunction exemption in the guise of or labelled as something else is still unlawful. See, e.g., NRDC. v. EPA, 749 F.3d 1055, 1063-64 (D.C. Cir. 2014) (finding affirmative defense to civil penalties due to malfunctions unlawful).

Response: The regulatory text referenced by the commenter was placed in the docket at proposal and is available at docket ID: EPA-HQ-OAR-2018-0417-0033. We determined that it is impractical to require these narrowly defined intermittent activities to be subject to the process vent standard for the HCl Production NESHAP given the inability to effectively measure emissions from maintenance activities that occur during startup and shutdown. 
As explained in the preamble for this final action, based on consideration of public comments, the EPA is adding work practice standards for maintenance vents to ensure emissions from these activities are subject to standards. As discussed in section IV.D of the preamble, we determined that it is impractical to measure the extremely small amounts of HCl and chlorine (Cl2) that could be emitted after opening these "maintenance vents" to the atmosphere and that these emissions could be adequately addressed through work practice standards. We are requiring facilities to demonstrate that emission points designated as maintenance vents are thoroughly cleaned prior to opening them to the atmosphere. There is very little potential for emissions from these vents, even if all residual HCl and chlorine in the equipment being serviced is emitted.  

Monitoring/Recordkeeping/Reporting

Comment (0134): One commenter submitted a number of requests for EPA to revise the underlying MACT standard: 

 Remove the requirement for facilities to establish a maximum operating pH for water or caustic scrubbers. 
 Provide   an alternative to continuously measuring and recording pH, similar to the provisions of the Miscellaneous Organic NESHAP (MON) or add a "Pre-compliance Plan" provision to request alternative parameters. 
 Remove the pH monitoring requirements for water scrubbers. 
 Add language, similar to language that exists in 40 CFR part 63 subpart SS, that provides flexibility to supplement parameter value measurement with engineering assessments and/or manufacturer's specifications. 
 Eliminate the requirement in 40 CFR 63.9015(a) to conduct performance tests once every 5 years or each time a facility's Title V permit is renewed if there have been no process changes impacting the results of the previous performance test. 
 Use regulatory text from 40 CFR 63.996 (subpart SS) for continuous monitoring systems. 
 Allow the use of EPA Method 26 of 40 CFR part 60 Appendix A in addition to EPA Method 26A to determine HCl and Cl2 concentrations in the vent stream. 
 Amend Table 7 to subpart NNNNN to eliminate the requirement under 40 CFR 63.9(j) to submit a notification of any change to information already provided under 40 CFR 63.9 (which includes the Initial Notification and Notification of Compliance Status among other types of notifications) within 15 calendar days of the change. 
 Amend Table 7 to subpart NNNNN to eliminate the requirements under 40 CFR 63.9(g)(1) and 63.10(e)(2)(i) and make these sections only applicable to continuous emission monitoring systems (CEMS). 
 Amend language in 40 CFR 63.9020(c) to add the words "any other continuous vent stream containing HCl or chlorine." According to the commenter, the language in 63.9020(c) concerning the use of design evaluations in lieu of conducting a performance test for HCl storage tanks and HCl transfer operations is unclear in the reference to "any other continuous vent stream."

Response: EPA is under a court-ordered deadline to take final action on its RTR review and does not have adequate time to consider these requests that are unrelated to the RTR review as part of this rulemaking.

Electronic Reporting

Comment (134): One commenter supports EPA's transition to electronic reporting but recommends limiting the scope of electronic reporting requirements to performance tests only rather than also including performance evaluations, notifications of compliance status, site-specific monitoring plans, and semiannual compliance reports.

Response: The Agency appreciates the commenter's support for a transition to electronic reporting. They Agency finds that electronic reporting eliminates paper-based, manual processes, thereby saving time and resources, simplifying data entry, eliminating redundancies, and minimizing data reporting errors, ultimately reducing the burden on regulated facilities. Moreover, the amendments and clarifications to rule language are expected to result in a reduction of current costs because compliance will be more straightforward. Section IV.D of the proposal preamble (See 84 FR at 1587) also presented the reasons why the electronic reporting required by this rule will provide benefits going forward. Additionally, based on the analysis performed for the Electronic Reporting and Recordkeeping Requirements for the New Source Performance Standards (ERRRNSPS) (80 FR 15100), electronic reporting results in an overall cost savings to industry when annualized over a 20-year period. The cost savings is achieved through means such as standardization of data, embedded quality assurance checks, automatic calculation routines, and reduced data entry through the ability to reuse data in files instead of starting from scratch with each report. As outlined in the ERRRNSPS, there are many benefits to electronic reporting. These benefits span all users of the data  -  EPA, state and local regulators, the regulated entities, and the public. 
However, the Agency disagrees with the commenter's suggestion to limit the scope of electronic reporting to performance tests only and thus is not making the revision requested by the commenter. As mentioned earlier, standardization of the data contained in the performance evaluations, notifications of compliance status, site-specific monitoring plans, and semiannual compliance reports, ensures cost and burden reductions while ensuring rapid recall of compliance activities. 

Comment (0134): One commenter provided feedback on the electronic reporting form templates included in the docket at proposal. The commenter noted some required reporting elements that appeared to be missing from the templates. The commenter also provided suggestions on the structure and functionality of various parts of the templates. The commenter included recommended changes to the draft templates included in the docket at proposal. 

Response: The Agency always appreciates feedback on rule templates, and it will revise the templates to correct misspellings. We note that the templates are not being finalized as part of this rulemaking action.  EPA welcomes continued discussion with stakeholders, should they choose to participate, and will consider revisions to the template as part of that process. 

Comment (0134, 0145): Two commenters request that EPA allow two years from the date of publication of the final rule before requiring electronic submission of the various reports required by the rule. The commenters state that the required reports are complex, and time is needed to train personnel on the use of the new forms and reporting system. The commenters note that EPA has allowed two years for electronic reporting in recent similar rulemakings. 

Response:  Upon consideration of the commenter's suggestion, the EPA agrees that owners and operators should have 1 year after the date of publication of the final rule or 1 year after the reporting template has been on CEDRI, whichever is later, before they are required to submit semiannual compliance reports via CEDRI. This will provide users with 1 year to become familiar with the template and electronic reporting system prior to being required to submit reports electronically, which will provide adequate time for facilities to adjust to electronic reporting, as well as assure that the forms will work properly, prior to the date that owners and operators must start submitting these reports electronically. The EPA encourages users to become familiar with the system well in advance of being required to use it. For previous rulemakings with reports required to be submitted electronically via CEDRI, prior to a compliance reporting deadline, the EPA has provided webinars to our various stakeholders on the access and reporting of the given report in CEDRI. The EPA is planning to provide this same service to the industry trade association and facilities subject to the 40 CFR part 63, subpart NNNNN electronic reporting requirements, if requested to do so. The EPA plans to publish the final template on CEDRI soon after the final rule is signed and published. Although facilities will have up to 1 year after the final template is on CEDRI to begin using the template and submitting reports via CEDRI, facilities may begin submitting reports via CEDRI as soon as the final template is available.

Comment (0134): One commenter requests clarification of how the requirement affects sources in states, such as Texas, that do not require copies of reports to be submitted to EPA, i.e., where there has been delegation to the state under 40 CFR 63 Subpart NNNNN. 

Response: Enforcing and adhering to electronic reporting requirements by delegated authorities is no different than enforcing and adhering to other rule requirements by delegated authorities.  A delegated authority must agree to act in a certain manner in order to obtain delegation for a specific rule; should the applicant not encompass the total rule, delegation would not occur. 
The electronic reporting requirements apply to all affected sources. The EPA's electronic reporting system provides quick and easy access to submitted data for state, local, and tribal agencies in two convenient locations: CEDRI and WebFIRE. The EPA works with state, local, and tribal agencies to encourage and implement the use of CEDRI for electronic delivery of reports within their jurisdiction. As more information is provided through EPA's electronic reporting system, EPA expects more state, local, and tribal agencies to adopt the use of the EPA's electronic reporting system. Electronic reporting minimizes submission of unnecessary or duplicative reports in cases where facilities report to multiple government agencies and the agencies opt to rely on EPA's electronic reporting system to view report submissions. Where air agencies continue to require a paper copy of these reports and will accept a hard copy of the electronic report, facilities will continue to have the option to print paper copies of the electronic report to submit to the air agencies, and, thus, minimize the time spent reporting to multiple agencies. Owners or operators should contact their EPA Regional representatives to determine the status of delegation for state, local, or tribal entities. Note that without delegation to a state, local, or tribal entity, the EPA Regional office with jurisdiction over the source is the delegated authority.
Comment (0134): Two commenters request that EPA continue to work with State agencies to develop an efficient and one stop location for submittal of information that is now required to be reported electronically to EPA or provide states access to EPA's Central Data Exchange (CDX) to eliminate duplicative reporting obligations. The commenters support electronic reporting conceptually but argue that the current system includes duplicative requirements such as the reporting of deviations in the semi-annual compliance report as well as the Title V deviation report. The commenters state that companies will have to send in 1) an electronic HCl NESHAP semi-annual compliance report to EPA, 2) a semi-annual HCl NESHAP compliance report to the delegated authority (if different than EPA), and 3) the Title V semi-annual deviation report. The commenters state that most state agencies will continue to request submittal of performance test reports, performance evaluations, notifications of compliance status, site-specific monitoring plans, and semi-annual compliance reports directly to them. Therefore, populating EPA's CDX using CEDRI results is duplicative and inefficient work for companies. 

Response: The Agency appreciates the commenters' recognition of the EPA's continuing work with State agencies on electronic reporting. On an ongoing basis, the EPA works with state, local, and tribal agencies to encourage and implement the use of its Compliance and Emissions Data Reporting Interface ("CEDRI"), which is a node on the CDX, for electronic delivery of reports within their jurisdiction. All state, local, and tribal agencies have access to CEDRI now. As more information is provided through EPA's electronic reporting system, EPA expects more state, local, and tribal agencies to adopt the use of the EPA's electronic reporting system.
The EPA finds that electronic reporting minimizes submission of unnecessary or duplicative reports in cases where facilities report to multiple government agencies and the agencies opt to rely on EPA's electronic reporting system to view report submissions. As mentioned in 40 CFR 63.9050(e), which was not changed by this rulemaking, a single report that contains all of the relevant information can be submitted once to fulfill the compliance (and deviation) report requirement. Of course, ensuring the requisite information is included in the submission remains the responsibility of the submitters.

Comment (0134): One commenter requests clarification that approved alternative reporting schedules are not impacted by the proposed amendments requiring electronic reporting. The commenter notes that Table 7 to subpart NNNNN (the General Provisions applicability table) allows for alternative reporting schedules if agreed to by the delegated authority. The commenter requests that language reaffirming this provision be added to 40 CFR 63.9050(l). 

Response: The EPA agrees with the commenter that the general provisions of 40 CFR 63.10(a)(5) allow alternative reporting schedules for this rule if agreed to by the delegated authority but sees no need to add redundant language to 40 CFR 63.9050(l). Source owners or operators may continue to avail themselves of that ability with no need for a rule change. 

Comment (0145): One commenter suggested that EPA should provide additional instructions for completing each cell with details about the type of information requested for each response in the electronic reporting forms. Alternatively, the commenter states that EPA could provide a pull-down options list for each cell of potential responses from which the user could select a pre-loaded response. One commenter also noted that some required information does not fit well within a single cell. According to the commenter, an affected source may consist of multiple HCl production facilities at a site and each HCl production facility may have multiple emission limitations, more than one control device, and multiple parameter monitoring systems. Attempting to report information pertaining to control devices, parameter monitoring systems, CMS downtime, etc., where required is difficult to do in one cell. The commenter suggests that EPA provide multiple cells to accommodate this information. 

Response: The EPA thanks the commenter for its suggestions to improve the draft templates. We note that the templates are not being finalized as part of this rulemaking action.  EPA welcomes continued discussion with stakeholders, should they choose to participate, and will consider revisions to the template as part of that process. While not all suggestions may be incorporated, as individual commenters desires may conflict with rule requirements or with other stakeholder views, such suggestions will be considered for inclusion.  

Comment (0145): One commenter recommends that EPA provide some capacity to electronically sign required forms to fulfill the requirements in 40 CFR 63.9(h)(2)(i) and 40 CFR 63.9055(c)(2). 

Response: The EPA notes that the existing template already incorporates electronic signature requirements for the notification of compliance status reporting requirements in 40 CFR 63.9(h)(2)(i) and that no reporting requirement exists for the recordkeeping requirements associated with the deviation information contained in 40 CFR 63.9055(c)(2). Therefore, no changes to the template or rule are planned.

Comment (multiple commenters): The EPA requested public comment on the use of the IRIS value for ethylene oxide for regulatory purposes in the proposed rule. A number of comments were received supporting and opposing the use of the IRIS value. Other commenters provided comments regarding emissions of ethylene oxide from specific facilities that are not a part of the HCl Production source category. 

Response: The EPA is not taking any action regarding our request for public comment on the use of the ethylene oxide IRIS value in this rulemaking. The Agency intends to respond to the comments received on the use of the ethylene oxide IRIS value in the final rule addressing the residual risk and technology review for the Miscellaneous Organic NESHAP (MON). The HCl Production source category does not emit ethylene oxide and the EPA is not regulating ethylene oxide from other source categories as a part of this final rule. 

Technology Review 

Comment (0147): One commenter claims that EPA provides no evidence of the technology review on which it is relying  -  it refers to the "results" of the technology review, but it cites no document in the record providing these results. It discusses the fact that sources use scrubbers, but it does not provide information on the kinds of scrubbers or any other details. EPA also does not discuss any other acid gas controls, or types of VOC controls or leak detection and repair, or any additional methods that it considered for these sources. EPA states that it reviewed Title V permits for these sources. None of the documents that EPA has described here appear to be in the docket. Therefore, EPA has failed to provide any support for its determination, and it is unlawful and arbitrary.

One commenter disagrees with EPA's proposal and believes that there are developments in VOC and acid gas controls, leak detection and repair, and monitoring that EPA must consider under the technology review  for this source category. The commenter states that commenters do not have access to the range of information EPA has, and that is partly why it is so prejudicial that EPA has not provided the information it used (and cited) for its technology review. However, since EPA finalized the original standards, EPA has recognized such developments in other contexts, and Commenters provide information on developments of which they are aware.

A commenter points to EPA-required monitoring of HCl in multiple national standards in recent years as a development for monitoring of acid gases and claims EPA should strengthen monitoring here due to these demonstrated developments.170 

One commenter states that EPA must also account for developments in addressing and reducing equipment leaks, using the most up-to-date leak detection and repair methods. Currently the rules just require a non-specified "LDAR plan," without any hard limit or specific requirements of any kind to prevent or reduce equipment leaks, or corrective action required within any specific timeframe. See 40 C.F.R. Subpart NNNNN tbl. 1.
 
The commenter claims that EPA failed to review  use of low-emission valves, and optical gas imaging, both of which can increase the effectiveness of leak monitoring and ensure faster repair. The commenter also claims that additional deployment of infrared ("IR") or open-path FTIR systems can supplement other monitoring. In addition, the commenter suggests that EPA should take information from enforcement of existing LDAR programs and use that to strengthen the regulations. The commenter points to several provisions in EPA consent decrees that should be considered as developments: 
 
 Lowering the leak definition, to capture leaks at lower and still potentially harmful levels, such as 250 ppmv or lower; 
 Requiring "low-leak" technology and other materials for repairs; 
 Periodic monitoring of devices used for closure of open-ended lines; 
 Requiring more stringent or faster repair when leaks are found, which will reduce the amount of HAPs allowed to leak into the air; 
 Require monitoring audits using third parties for leak-prone equipment, closure devices, valves, pumps, connectors, agitators, etc.; 
 Require more frequent regular monitoring of similar leak-prone equipment.

Response: The EPA has determined that the technologies cited by the commenter are not warranted for the HCl production source category at this time. The EPA does not have data illustrating that these technologies are applicable to the HCl Production source category. The fact that technologies for acid gas controls other than those used in the HCl Production source category exist and may have been used in other industries with vastly different emissions than the minimal amounts of HCl and Cl2 emitted from the source category in question in this rulemaking does not justify applying those technologies here. The HCl Production NESHAP requires facilities to develop and operate according to a plan to identify and address leaks in a timely manner. HCl is highly corrosive to equipment and it is in a facility's best interest to identify and correct any leaks of HCl as expeditiously as possible. We have concluded we do not have sufficient data to codify any more elaborate leak detection and repair requirements at this time. We will evaluate whether such requirements are appropriate during our next technology review, as appropriate.   

The EPA disagrees that evidence for the proposed technology review was not available to the public. Copies of permits are available from the permitting authority, and in most cases are posted on the websites for those agencies. As stated at proposal, our search of the RBLC and other peer-reviewed literature databases did not yield any information regarding technologies related to HCl production. The EPA is entitled to rely on a summary of publicly available information without including all the underlying information in the rulemaking docket.

Comment (0147): One commenter asserts that EPA is required to "review, and revise as necessary" the emission standards for HCl production sources. 42 U.S.C. § 7412(d)(6). EPA is blatantly ignoring this obligation in regard to the "no control" standards for wastewater. EPA neither acknowledges its statutory obligation to set limits on such uncontrolled HAPs, nor provides any justification for how it could refuse to find that it is "necessary" to revise the existing standards to set a limit. 

Further, EPA has failed to complete a technology review of the current standards' treatment of the HAPs emitted by wastewater  -  i.e., the no-control allowance for them. EPA cannot determine whether developments in pollution control make it "necessary" to revise the emission standards without first determining what developments, if any, have occurred for all HAPs emitted by HCl sources.
Response: CAA section 112(d)(6) requires the EPA to review and revise, as necessary (taking into account developments in practices, processes, and control technologies), emission standards promulgated under this section. We do not agree with the commenter's assertion that the EPA must establish new standards for unregulated emission points or pollutants as part of a technology review of the existing standards. The EPA reads CAA section 112(d)(6) as a limited provision requiring the Agency to, at least every 8 years, review the emission standards already promulgated in the NESHAP and to revise those standards as necessary taking into account developments in practices, processes, and control technologies. Nothing in CAA section 112(d)(6) directs the Agency, as part of or in conjunction with the mandatory 8-year technology review, to develop new emission standards to address HAP or emission points for which standards were not previously promulgated. As shown by the statutory text and the structure of CAA section 112, CAA section 112(d)(6) does not impose upon the Agency any obligation to promulgate emission standards for previously unregulated emissions.
When the EPA establishes standards for previously unregulated emissions, we would not establish those initial standards pursuant to CAA section 112(d)(6) but instead would establish the standards under one of the provisions that govern initial standard setting--CAA sections 112(d)(2) and (3) or, if the prerequisites are met, CAA section 112(d)(4) or CAA section 112(h). Establishing emissions standards under these provisions of the CAA involves a different analytical approach from reviewing emissions standards under CAA section 112(d)(6).
Though the EPA has discretion to develop standards under CAA section 112(d)(2) through (4) and CAA section 112(h) for previously unregulated pollutants at the same time as the Agency completes the CAA section 112(d)(6) review, any such action is not part of the CAA section 112(d)(6) review, and there is no obligation to undertake such actions at the same time as the CAA section 112(d)(6) review. Therefore, we are not establishing new standards for previously unregulated emissions as part of this rulemaking. We do not have data to establish standards for previously unregulated wastewater sources at HCl Production facilities at this time. We will evaluate whether it is appropriate to do so at a later time, as appropriate, if we have data to do so. 


