


            


National Emission Standards for Hazardous Air Pollutant Emissions: Group I Polymers and Resins; Marine Tank Vessel Loading Operations; Pharmaceuticals Production; and The Printing and Publishing Industry 



                   Summary of Public Comments and Responses







                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                    (This page is intentionally left blank)

                                       
                               TABLE OF CONTENTS

1	Background	5
2	Summary of Public Comments	9
2.1	Risk Assessment Data and Methodology	9
2.2	Consideration of Risks from Entire Facilities, Industrial Clusters, and Other Sources in Risk Assessment	15
2.3	Consideration of Demographics in Risk Assessment	17
2.4	Additional Standards to Address Risks	19
2.5	Removal of the SSM Exemption	20
2.6	Affirmative Defense Provisions	22
2.7	Source Category Specific Comments	34
2.7.1	Source Category Specific Comment  -  Pharmaceuticals Production	34
2.7.2	Source Category Specific Comment  -  The Printing and Publishing Industry	35
2.7.3	Source Category Specific Comments  -  Group I Polymers and Resins	36
2.7.4	Source Category Specific Comments -- Marine Tank Vessel Loading Operations	38
2.8	Other Comments	48


                                       
                                       
                                       
                                       
                                       
                    (This page is intentionally left blank)


Background
      
      On October 21, 2010, the U.S. Environmental Protection Agency (EPA or Agency) proposed to revise the National Emission Standards for Hazardous Air Pollutants (NESHAP) for National Emission Standards for Hard and Decorative Chromium Electroplating and Chromium Anodizing Tanks; Group I Polymers and Resins; Marine Tank Vessel Loading Operations; Pharmaceuticals Production; The Printing and Publishing Industry; and Steel Pickling -- HCl Process Facilities and Hydrochloric Acid Regeneration Plants, under the authority of section 112 of the Clean Air Act (CAA).  Four of the six MACT included in the proposal are addressed in this final action:  Group I Polymers and Resins; Marine Tank Vessel Loading Operations; Pharmaceuticals Production; and The Printing and Publishing Industry.  We summarized and responded to numerous comments in the preamble to our final action.  This document presents summaries of the additional comments that were submitted and our responses to those comments.  The comments to which we are responding were submitted in 104 letters from industry representatives, governmental entities, environmental advocacy groups, and citizens during the public comment period.  A list of the commenters, their affiliations, and the EPA docket number assigned to their correspondence is presented in Table 1-1.  The comment summaries and responses for the other two MACT rules included in the October 2010 proposal - Hard and Decorative Chromium Electroplating and Chromium Anodizing Tanks and the Steel Pickling -- HCl Process Facilities and Hydrochloric Acid Regeneration Plants - will be included in a separate response to comments document made available at the time we take final action on those two source categories.  
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
TABLE 1-1.  LIST OF COMMENTERS ON PROPOSED NATIONAL EMISSIONS STANDARDS FOR HAZARDOUS AIR POLLUTANTS 
                              Docket Number a b 
Commenter and Affiliation
                                   256, 265
Robert J. Morehouse, ExxonMobil Refining & Supply Company
                                   257, 262
James L. McGraw, Managing Director & CEO, International Institute of Synthetic Rubber Producers, Inc (IISRP)
                                   258, 359 
Matthew Todd, American Petroleum Institute (API)
                                      259
James R. Thorton, Sector Director, Environmental, Health, and Safety, Northrop Grumman Shipbuilding, Inc. (NGSB)
                                      260
Donald R. Schregardus, Deputy Assistant Secretary of the Navy, Department of the Navy
                                      261
Jeffrey K. Francer, Assistant General Counsel, Pharmaceutical Research and Manufacturers of America (PhRMA)
                                      263
Ram Singhal, VP Technology and Environmental Strategy, Flexible Packaging Association (FPA)
                                      264
 Russell Wozniak, EH&S Global Regulatory Affairs, The Dow Chemical Company
                                      272
John Paul, Administrator, Regional Air Pollution Control Agency (RAPCA)
                                      273
Melvin Keener, Executive Director, Coalition for Responsible Waste Incineration (CRWI)
                                      274
David J. Shaw, Director, Division of Air Resources, New York State Department of Environmental Conservation
                                      275
Vinson Hellwig and Robert Colby, Co-Chairs, National Association of Clean Air Agencies (NACAA)
                                      276
Kyle Beall, Kean Miller on behalf of the Louisiana Chemical Association (LCA)
                                      278
Sheila Holman, North Carolina Department of Environment and Natural Resources, Division of Air Quality (NCDAQ)
                                      279
Lorraine Krupa Gershman, Director, American Chemistry Council (ACC)
                                      280
Christopher S. Colman, Counsel, HOVENSA
                                      281
Christina Harris, Kinder Morgan Liquids Terminals LLC
                                    283,301
Jennifer Carpenter, The American Waterways Operators
                                      284
NEDA/CAP
                                      286
Peter Pagano, Vice President, Environment, American Iron and Steel Institute,
                                      287
Matt Todd, Chair, The Residual Risk Coalition
                                      288
John Dege, Leader  -  Air Competency, DuPont
                                      289
SSM Coalition
                                      290
Earthjustice, Sierra Club, California Communities Against Toxics, Natural Resource Defense Council
                                      291
Shannon Broome, Executive Director, Air Permitting Forum
                                      293
Robert Barlow, VP of Manufacturing and Operations, Zeon Chemicals, L.P.
                                      294
Jim Wilson, Vice President and Chief Legal Counsel, Legal and Environmental Services, Roche Colorado Corporation
                                      295
Eric Hiser, Counsel, Nucor Corporation
                                      297
Jesse N. Marquez, Executive Director, Coalition For A Safe Environment
                                      298
Linda Farrington, Eli Lilly
                                      300
Marise Lada Textor, Corporate Director- Regulatory Affairs, TPC Group (formerly Texas Petrochemicals LLC)
                             270, 302-349, 351-358
Comment Campaign Organized by EarthJustice, includes submissions from: Christopher Lish, Alison Leary, Theophilus Griswold, Richard Miller, Laura Jobe, Craig Leman, Kathy Shimata, Virginia Harris, Diane Aikins, Takeshi Imajo, Alyssa Kay, Bob Browning, Gloria Sferra, Katherine Russell, John Gau, N. Gladfelter, Peggy La Point, Chris Dietrich, Elizabeth Zenker, Matthew Reid, Charles Schmalz, Sue Chickman, Salam Tims, Robert Bixby, Charlene Woodcock, Michael Nola, Ralph Philbrick, Lawrence Thompson, Hugh Curtler III, Steve and Deborah Anthes, Kimberly Rowlett, Sarah McKee, David Johnson, C. Morey Jr., George and Frances Alderson, Ze Daluz, Jean Mortensen, Nancy Brothers, Julie Robinson, Duke Briscoe, Ron Rodarte, Jill Harmer, Rebecca Ruggles
                                       

                                       
a Document ID identifies comments found in Docket No. EPA-HQ-OAR-2010-0600 by document number.
b Documents submitted to the docket together appear in the docket as multiple documents with the same 4-digit "root number" followed by a unique extension.  For example, comments received via email are generally assigned two numbers.  First, the email message itself is assigned the root number (e.g., 0019).  Second, the attachment comment document is assigned an extended number based on the root number (e.g., 0019.1).  Additional attachments, if any, would follow the same pattern (e.g., 0019.2, 0019.3, etc.).  For the purposes of this table and comment summary, we listed all related documents under the root number but only included the last three digits (e.g., 0200 would be 200).  
Summary of Public Comments

	A total of 104 written comments were received on the proposed standards for the Group I Polymers and Resins; Marine Tank Vessel Loading Operations; Pharmaceuticals Production; and The Printing and Publishing Industry NESHAP.  Some of these comments were part of a mass comment campaign sponsored by Earthjustice.  The comment reflects some of the same comments as those provided by Earthjustice, commenter 290.  The following comments were part of the mass campaign: 270, 302-349, 351-358.  Although we did not identify the mass campaign commenters in the comment summaries, we responded to each of their comments with the exception of comments specifically regarding the chromium electroplating source categories.  Responses to comments specifically regarding the Hard and Decorative Chromium Electroplating and Chromium Anodizing Tanks NESHAP and the Steel Pickling -- HCl Process Facilities and Hydrochloric Acid Regeneration Plants NESHAP will be presented in a subsequent final rulemaking.

      The following sections provide a summary of these comments, as well as the Agency's responses to these comments.  For the purpose of orderly presentation, the comments have been categorized under the following topics:

 Risk Assessment Data and Methodology
 Consideration of Risks from Entire Facilities, Industrial Clusters, and Other Sources in Risk Assessment
 Consideration of Demographics in Risk Assessment
 Additional Standards to Address Risks
 EPA's Authority Under and Approach to the CAA 112(d)(6) Technology Review
 Removal of the Startup, Shutdown, and Malfunction Exemption
 Affirmative Defense Provisions
 Source Category Specific Comments-
       Pharmaceuticals Production
       Printing and Publishing Industry
       Group I Polymers and Resins
 Marine Tank Vessel Loading Operations (MTVLO) 	
            
Risk Assessment Data and Methodology 
     
     Comment  1:  Commenters 274 and 275 believe the population cancer risk metrics are one of the most transparent ways to make risk management decisions concerning acceptability and ample margin of safety determinations under CAA section 112(f).  In line with EPA's goals, these commenters and commenter 297 would like to see cancer risk from HAPs reduced to below the 1-in-1 million risk level for the majority of the population that resides near a facility if it is technologically feasible. Commenter 297 stated that EJ communities do not accept a risk higher than 1-in-1 million.  This commenter also noted that risk to EJ communities already exceeds 1-in-1 million at the Port of Los Angeles, Port of Long Beach, and at each major petroleum refinery in and around Wilmington.  The commenters believe strengthening the protections for HAP sources is essential to demonstrate EPA's commitment to seeking environmental justice for communities overburdened by toxic pollution.  In contrast,
	Response  1:  As stated in the Benzene NESHAP, in determining the need for residual risk standards, we strive to limit to no higher than approximately 1-in-10 thousand the estimated risk that a person living near a plant would have if he or she were exposed to the maximum pollutant concentrations for 70 years and, in the ample margin of safety decision, to protect the greatest number of persons possible to an individual lifetime risk level of no higher than approximately 1-in-1 million.  However, in determining whether to require additional standards, these levels are not considered rigid lines, and we weigh the cancer risk values with a series of other health measures and factors in both the decision regarding risk acceptability and in the ample margin of safety determination. If we determine that the risk is unacceptable, we must impose additional requirements to reduce emissions without consideration of cost. However, as part of our determination of whether standards protect public health with an ample margin of safety, we do not look solely at whether technology is "feasible," (as urged by the commenter) but also consider cost.  We also consider cost as part of technology review under section 112(d)(6), which we are required to perform periodically. 

	Comment  2:  Commenters 272, 274 and 275 note that EPA has problems with its risk assessments in that it lacks current actual emissions data.  As such, the commenters recommend that periodic emissions testing be included in any MACT standard that requires the installation of control equipment and initial emissions testing for compliance demonstration purposes.  Commenter 272 notes this data would provide greater assurance for the public and regulatory community that sources are complying with emission limits on an on-going basis.  The commenters further note it is often difficult for state or local agencies to impose requirements like this if it is not an explicit requirement of the federal rule.  The commenters say the effect of this is that EPA does not receive information it needs voluntarily from these agencies to conduct risk assessments.
      Commenter 290 states that EPA should not rely solely on industry-reported, unverified emissions data in its residual risk review, which the commenter states, includes some of the data in the NEI.  The commenter states that where there are uncertainties or gaps in the data, EPA needs to complete the section 114 information collection process.  Commenter 297 stated that EPA has not conducted a detailed technical review of HAP emissions data for these source categories and facilities. 
      Commenter 297 stated that EPA has failed to validate the marine terminal self reported annual emissions used to establish the proposed rule.  The commenter requests that independent third-party emissions testing be performed at all U.S. marine terminals to validate annual emissions reporting and to validate that the proposed rule is adequate to protect public health.
      Response  2:  The data for the four MACT rules were derived from the National Emission Inventory (NEI) and directly from industry and states.  The source of the data is documented in the docket.  The EPA has greatly improved the NEI for HAPs since the 1996 NEI.  We have worked diligently with state and local agencies and tribes to improve their emission inventories over the years and we have instituted a number of QA measures to improve data received from state and local agencies since 1996.  The 2005 NEI underwent a number of QA steps that were not used in the 1996 NEI.  Documentation for the 2005 NEI point source emissions inventory is available on EPA's CHIEF website, http://www.epa.gov/ttn/chief/net/2005inventory.html. Quality assurance and augmentation procedures used for the 2005 NEI point sources are contained in the report, EI Quality Assurance and Data Augmentation for Point Sources, posted at: http://www.epa.gov/ttn/chief/net/2002inventory.html. The 2005 NEI data have been reviewed extensively internally and externally during a 9 month effort as part of the development of the 2005 National Air Toxics Assessment (NATA).  As part of the 2005 NATA review, more than one million records for HAPs were revised.
      For RTR categories, EPA's review included an evaluation of facilities contained in the source category, the emissions units expected to be included for the processes in the source category, coordinates and stack parameters and the HAP compounds and emissions levels typically seen. The EPA often had additional information on the source category, and where better data were available, this information was included in the proposal datasets.  Section IV.A.1 of the October 21, 2010 proposal preamble (75FR65068) sets forth a detailed explanation of the review that EPA performed on the emissions data for the source categories subject to this final rule.
	In addition, as  part of the October 21, 2010 proposal, we placed the data sets for review using a RTR data review tool in the docket and solicited comments on the data sets in the preamble to the proposal.  The RTR review of the proposal data sets act as another external third party review mechanism in addition to the external 2005 NATA review of the RTR data sets.  Several opportunities have been provided to review the data used in the current proposal during the last 3 years.  Through enhanced NEI QA procedures, 2005 NATA review, public comment on proposal files, source category-specific engineering reviews, and  modeling, improvements to the data files have been made.  We believe that EPA has provided multiple opportunities for third party review of data
	With regard to the commenters recommendation that periodic emissions testing be included in any MACT standard that requires the installation of control equipment and initial emissions testing for compliance demonstration purposes, we note that we are also requiring the electronic submittal of performance test data to increase the ease and efficiency of data submittal and to improve data accessibility.  Specifically, owners and operators are required to submit electronic copies of applicable reports of performance tests to EPA's WebFIRE database through an electronic emissions test report structure called the Electronic Reporting Tool (ERT).  Having these data, EPA would be able to develop improved emission factors, make fewer information requests, and promulgate better regulations.

	Comment  3:  Commenter 287 stated that EPA cannot lawfully use MACT allowable emissions in the proposed risk assessments because the Agency failed to provide adequate explanation for the use of these data.  This commenter and commenters 265 and 359 state that considering the conservative assumptions already included in the risk assessments, adding another layer of conservatism is unneeded, and actual emission should be used.  Commenter 265 notes that potential emissions overstate emissions and risk because facilities do not operate at the level of their potential emissions in order to maintain a compliance margin.  This commenter also states that considering allowable emissions dilutes the more important health measures already examined, which more directly assess risk from the regulated source category.  Commenter 287 states that EPA has also provided no data or analysis to indicate that the use of actual emissions results in negative bias or that affected sources actually do emit at levels significantly higher than the actual data show for a significant period of time.  The commenter states that failure to provide a reasoned explanation for the approach or evidence supporting the use of MACT allowable emissions renders the proposal insupportable under the law.  Commenter 276 also requests that risks assessments be based on source category actual emissions.  
      In contrast, commenter 275 stated, as it has in the past, that the Agency should consider risk based on potential or allowable emissions because emissions and associated impacts could increase over time, and analyses based on actual emissions underestimate residual risk.  This commenter noted that the thresholds for major source status were based on maximum potential-to-emit, and permits are issued based on potential emissions.  Further, the commenter argued that limiting the scope of risk assessments to actual emissions would be inconsistent with the applicability sections of the MACT standards.  While acknowledging that EPA has relied less on actual emissions than it has in the past, the commenter would like to see more reliance on the use of allowable emissions.
	Response  3:  We have discussed the use of both allowable and actual emissions in previous actions, including the final Coke Oven Batteries residual risk rule (70 FR 19998-19999, April 15, 2005) and in the proposed and final Hazardous Organic NESHAP (HON) residual risk rules (71 FR 34428, June 14, 2006, and 71 FR 76603, December 21, 2006, respectively).  In those previous actions, we noted that modeling the allowable levels of emissions (i.e., the highest emission levels that could be emitted while still complying with the NESHAP requirements) is inherently reasonable since they reflect the maximum level sources could emit and still comply with national emission standards.  But we also explained that it is reasonable to consider actual emissions, where such data are available, in both steps of the risk analysis, in accordance with the Benzene NESHAP. We also note that our use of allowable emissions levels in the risk assessments in this rulemaking did not result in revising the previously established standards due to risk concerns.
      	
	Comment  4  Commenter 290 states that EPA should include emissions from malfunctions in its risk assessments for the source categories at issue in this final rule.  The commenter references information that suggests emissions during upsets dwarf emissions from normal operations.  The commenter also states that EPA should include emissions information that occur during periods of violation of the MACT standard in the risk assessment using actual or estimated data and account for variability in actual emission levels.
	Response  4:  We believe SSM events do not contribute significantly to cancer or chronic noncancer risks for the source categories in this rulemaking because SSM events are inherently short-term and infrequent relative to annual operations and emissions.  The commenter did not supply any data to the contrary.  

	Comment  5:  Commenter 297 requests that EPA identify, quantify, assess and mitigate all public health and environmental impacts from HAPs, including impacts from dermal absorption, aerial deposition onto land and water, and climate change.  The commenter also stated that EPA's risk assessment fails to assess bioaccumulative toxins.
	Response  5:  In our risk assessments, dermal absorption and aerial deposition onto land and water are considered for any HAP that are known to be persistent and bio-accumulative in the environment.  More information on the approach used for these source categories can be found in the residual risk documentation in the docket for this action.  
     
	Comment  6:  Commenter 297 stated that the EPA has not considered the incidence (the number of people estimated to suffer cancer or other health effects due to exposure to a pollutant) because it has not conducted a health impact assessment, public health survey, or established a public health baseline, so EPA does not know the number of people impacted.  The commenter requested that EPA conduct these studies to determine the impact of its proposed rulemaking.
	Response  6:  In our risk assessments, we do estimate the incidence of cancer and non-cancer health effects and have considered these factors in our decisions regarding the need for additional standards to address residual risk.  More information about the estimated incidence of health effects can be found in the residual risk documentation in the docket for this action.
     
	Comment  7:  One commenter, 297, stated that the risk factors stated are not accurate because they are based on statewide or national averaging and not the actual impacts.  The commenter noted that risks based on average in an area 50 km from any source is not acceptable when 90%+ of the impacts occur less than 5 km from the source.
	Response  7:  The commenter is misinterpreting the risk results.  In the risk assessments, HAP concentrations and associated risks are estimated at the centroid of every census block from the facility fenceline out to a distance of 50 km.  This includes all areas within 5 km from the facilities.  In addition, the results are not averaged over the area of the nation or a state, but are estimates of risk precisely at the locations of the census block centroids around the facilities.  More information about the method of modeling and estimating risks can be found in the residual risk documentation in the docket for this action.  
     
	Comment  8:  Commenter 290 states that EPA should not consider a non-cancer TOSHI of 1.0 as an acceptable threshold for a single source category.  This approach will not limit the TOSHI to 1.0 or less for a whole facility when a facility has multiple source categories.  The commenter is also unsure why the EPA has chosen to put the threshold at 1.0 when it considered it to be 0.2 for a single source category in the past.
	Response  8:  The EPA has not used a TOSHI value of 0.2 as a default upper limit for acceptable chronic noncancer risk in RTR rule making.  In determining whether to require additional standards, non-cancer risks are considered in addition to cancer risks and other health measures and factors. A TOSHI of 1.0 is not considered a rigid line for acceptability, and we weigh this value with other health measures and factors in both the decision regarding risk acceptability and in the ample margin of safety determination.  In regards to facility-wide risk, if a source category contributes significantly to elevated facility-wide non-cancer risks, we may consider whether additional standards should be required to reduce those risks; however, that situation was not presented for any of the source categories in the October 2010 proposal. 
     
	Comment  9:  Commenter 291 states that EPA should undertake supplemental analyses for all the source categories addressed in the proposal to inform decision makers and the public about the degree of the uncertainties in the risk estimates.  The commenter states it appears EPA performs this type of analysis only when estimated risks for a source category approach the upper range of acceptable risks, for the chromium electroplating source category in this instance.  
	Response  9:  As the commenter notes, we performed a supplemental analysis for the chromium electroplating source categories but not for the source categories subject to this rulemaking. By performing the supplemental analysis for the chromium electroplating source categories, we tried to present a more realistic picture of the risk, to understand the degree to which the risk might be overstated, and how much weight to attach to the conservative analysis in our decisions regarding risk acceptability and ample margin of safety.  Although the type of chromium electroplating plants used in the NEI analysis is based on a variety of reliable sources, including ICR responses for the Plating and Polishing NESHAP, trade association data, data from State agencies, and information from websites, we were unable to identify the specific chromium electroplating plant type for nearly half of the data set.  Such was not the case for the source categories subject to this rulemaking.  Since the emissions data used for these source categories were considered to be of high confidence we did not need to perform a supplemental analysis.  While the data used in the analysis for these source categories were the best actual emissions estimates available, the data used for the chromium categories contained estimates known to be highly conservative and that likely overstate emissions.  

	Comment  10:  We received comments both in favor of and objecting to the use of census block centroids in the analysis of chronic exposure and risk. Two commenters (275 and 290) argued that the use of the census block centroid dilutes the effect of sources' emissions, as the maximum point of impact can be far from the centroid and may be at or near a facility's property line, and suggested that the risks for a source category be based on concentrations at the fenceline and beyond and include risks to the maximally exposed individual. In contrast, other commenters felt the use of the census block centroids was appropriate for these source categories, and one commenter (265) added that using the fenceline as a location to estimate risk is inappropriate in risk assessment because people do not generally live at the fenceline, and this approach would overstate risk.
	Response  10  As we have noted in the development of previous residual risk regulations, such as the HON, EPA contends that, 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. Further, EPA asserts that it is appropriate to use census block information on where people actually reside to locate the estimation of 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 US Census Bureau); each is typically comprised of approximately 40 people or about 10 households.  In EPA risk assessments, the geographic centroid of each census block containing at least one person is used 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 experiences the maximum individual risk. In some cases, because actual residence locations may be closer to or farther from facility emission points, this may result in an overestimate or underestimate of the actual annual concentrations (although there is no systematic bias for average levels). Given the relatively small dimensions of census blocks in densely-populated areas, there is little uncertainty introduced by using the census block centroids. There is more uncertainty when census blocks are larger, although there is still no bias on average.

	Comment  11:  Commenters 265 and 276 state that EPA should include central tendency or most expected probability of risk to better communicate the conservative nature of the risk assessments.
	Response  11:  In most cases, quantitative dose-response estimates representing the upper-bound of risk are the only values available, and thus are the ones used in our assessments of risk.  The upper-bound potency value that is presented in IRIS is routinely characterized using the standards descriptor for the cancer potency ("upper
bound"), by saying that the upper bound is not likely to underestimate risks, that true risks are likely to be less, and that, for some individuals, risk may be zero.  In addition, in our decisions regarding risk acceptability and ample margin of safety, we take into account the distribution of risks across the population and not just the maximally exposed individual.
Consideration of Risks from Entire Facilities, Industrial Clusters, and Other Sources in Risk Assessment
     
     Comment  12:  Commenter 290 states that EPA did not appropriately measure or address aggregate emissions and impacts from clustered or nearby sources in the same source category.  The commenter suggests EPA perform a cluster analysis similar to that of the analysis it did for the synthetic organic chemical manufacturing industry (SOCMI) accounting for all facilities in the source category.  The commenter states this is especially important for the chromium electroplating source categories which likely have several clusters of facilities.
	Response  12:  In our current risk modeling methodology, which has been updated since we performed the analyses for SOCMI, overlapping risks for the same source category are automatically included in determining the MIR, incidence, etc.  Any risks from clusters of facilities in the source categories investigated were included in the reported results of the analysis for the source category. We will respond to rule-specific comments for the Hard and Decorative Chromium Electroplating and Chromium Anodizing Tanks NESHAP in the final rule for that NESHAP.
      
	Comment1 13  Commenter 291states that considering facility-wide risks would constrain whole facilities to the ample margin of safety level established in the Benzene Rule decision framework for individual source categories.  The commenter also states that it will be difficult for the EPA to reconcile aggregated risk results from different sources and that it is unlikely EPA would identify any better emission reduction opportunities than it would by conducting individual source category analyses.  The commenter also states that EPA is correct in concluding that facility-wide and background risk levels are significantly more uncertain because the data relied on for those estimates are more uncertain.  The commenter also states that NATA is too blunt and inexact to be used for this regulatory purpose, citing the NATA website: "The results cannot be used to identify exposures and risks for specific individuals, or even to identify exposure and risks in small geographic regions such as a specific census block, i.e., hotspots."
     	Response  13:  We examined facility-wide risk to provide additional context to the source category risks.  In May 2010, the SAB advised us "* * * that RTR assessments will be most useful to decision makers and communities if results are presented in the broader context of aggregate and cumulative risks, including background concentrations and contributions from other sources in the area."  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 or categories in question.. In developing facility-wide risk estimates, we do not use NATA-derived risk results at the census tract level; rather, we develop census block-level risk estimates using NATA-derived facility-wide emissions estimates.  While we recognize that, because these risk estimates were derived from NATA 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 categories in question, they remain important for providing context, as long as their uncertainty is taken into consideration in the process.  In the case of the source categories in this rulemaking package, the results did not affect the decisions in any way, even taking their uncertainties into consideration.  We will continue to investigate the use of NATA-derived emissions as well as other emission data sources in developing future RTR assessments.
     	
	Comment  14:  Commenters 274, 275, and 290 are pleased to see that EPA has examined facility-wide risks to put the risks posed by the source categories in context.  However, commenters 274 and 275 would like more description provided about the co-located sources contributing to the facility-wide risks.  Commenter 274 adds that a verification of the accuracy of other HAP-emitting operations should have been conducted in the assessments.  Commenter 274 also stated that it is unclear how the NATA point source estimates were used in this portion of the assessment, and asks whether the maximum point of impact from the other HAP emitting sources were added to the results of the portion of the plant in the source category under investigation.
     	Response  14:  When facility-wide risk estimates are significant and significantly larger than the source category risk estimates, we determine the key emission sources associated with that difference and describe them along with our plans to address them via regulation in the preamble.  We may also perform additional QA on the facility-wide emissions in such cases. In the case of this rulemaking package, the results were not significant, and did not change any of the risk-based decisions, so additional follow-up was not needed or performed. When NATA-based emissions are used to develop facility-wide risk estimates, the dispersion modeling reflects the interaction of source category and facility-wide emissions according to the geographical layout of the emission sources and the dispersed plumes as a result of hourly windflow patterns; no artificial superimposition of plume maxima is performed as suggested by the commenter.
Consideration of Demographics in Risk Assessment
      
     Comment  15:  Commenter 291 supports EPA's discussed approach to analyzing potential demographic risks in which the risks would be analyzed only for the population within 5 km expected to have risks greater than 1-in-1 million.  The commenter states that EPA estimates the 5 km radius would include 90 percent of the population estimated to have risks greater than 1-in-1 million, and while EPA apparently plans to extend the radius to 50 km in future assessments, this should only be done if it does not reduce the accuracy of the resulting information, including the link between source category emissions and the population at risk.
	Response  15:  As the commenter notes, we have included the population within a 5-km radius of facilities in the demographic analyses for these source categories.  If we determine in the future to broaden the analysis for future risk reviews, the commenters will have an opportunity to raise any concerns during the comment period on such proposed rules. 
      
	Comment  16:  Commenter 270 states that the current methodology that assesses health impacts for a theoretical, healthy adult is insufficient and unrealistic, since the real people exposed to toxic pollutants include children, the elderly, and other vulnerable populations.  The commenter states that EPA has recognized new science that shows the impacts of pollution on these people are different from the assumptions in the conceptual standard EPA has employed in the past and should be considered in the final rule.  The commenter further states that EPA's lifetime exposure metric does not account for greater risk of harm during these times of life.  The commenter also states that EPA must consider whether certain individuals or communities have disproportionate risks in its residual risk determinations.  Commenter 297 noted that the assessment did not appear to take into account all sensitive receptors, such as pregnant women and people with preexisting health conditions, or account for EJ community factors such as availability of public health facilities, paramedic response availability, the population with little or no health care insurance, the population with little disposable income for health care, and the inability due to low income to relocate to cleaner communities.  This commenter added that EJ community residents are predominantly people of color who have more children and infants per population than white communities.
	Response  16:  The health benchmark values used in the acute and chronic non-cancer analysis specifically take into consideration susceptible individuals, and because the benchmark values used in the chronic cancer assessment are upper-bound estimates of increased cancer risk, they effectively consider sensitive subgroups as well. While some people that are exposed to HAP may face factors that make it difficult to obtain treatment for illnesses, the scope of these risk assessments is only to estimate the probability of an adverse effect related to exposure, and possible treatment or relocation difficulty is beyond the scope of these analyses.    
      
	Comment  17:  Commenter 297 stated that the EPA failed to identify and assess the cumulative sources of HAPs and impacts in EJ communities.  The commenter was also unable to find the EPA's cumulative impact data to determine if all sources had been identified and considered, such as co-located facilities, major, and non-major sources, and also noted that EPA has never conducted any long term study of multiple HAP exposure public health impacts.  The commenter requests that EPA determine the actual risks of exposure to HAPs for 70 years or more in EJ communities considering these factors.
	Response  17:  As explained in the risk assessment documents and source category data files in the docket for this action, we do consider many of the things the commenter is concerned about in our assessments.  For one, we estimate risks based on exposure to HAP emitted from sources within the source category under review for 70 years in all areas within 50 km of the facilities, regardless of community type or demographic distribution of the population.  We estimate risks for all the sources in a source category together, including any overlap in exposure to emissions from more than one facility within the source category. We also examined facility-wide risks in the assessments for these source categories.  We appreciate that the commenter would prefer an even more comprehensive analysis of risk; however, such a review goes beyond that mandated under the CAA and we currently lack sufficient data to perform reviews of total HAP exposure and risk to populations.  EPA does not have the resources to perform such a study, as urged by the commenter.    
      
	Comment  18:  Commenter 275 commends EPA for considering environmental justice issues in this proposal but believes improvements in the analysis are needed.  This commenter and commenter 274 are unsure why other factors recommended by the EPA as presented in the Environmental Justice Strategic Enforcement Assessment Tool (EJSEAT) were not considered in the RTR assessment.  The commenters note that the OTIS database appears to do this already at the facility-specific level and could be incorporated into the assessment to more accurately define the number of individuals impacted by the emissions and the demographics of the impacted community. 
	Response  18:  While the EJSEAT and OTIS database are general tools that can be used in considering environmental justice issues, the demographic analyses we performed for the proposal was superior for purposes of this rulemaking because  they were more specific to these particular source categories.  We note that we are working with the Office of Environmental Justice in an ongoing effort to develop new tools for considering environmental justice in rulemakings.
      	
	Comment  19:  Commenter 291 states that EPA should not consider demographics in its regulatory decisions unless there is evidence that a demographic category, or sensitive subpopulation, is more vulnerable to the specific health effects caused by the source category beyond mere proximity to the facility.  This commenter adds that EPA cannot assume that race, income, age, or education is a surrogate for a sensitive subpopulation without supporting health evidence.  The commenter states that because EPA failed to produce such evidence for this proposal, it cannot consider race, income, age, or education in its regulatory decisions.
	Response  19:  As stated in the Benzene NESHAP, in determining the need for residual risk standards, we strive to limit to no higher than approximately 1-in-10 thousand (100 in a million) the estimated cancer risk that a person living near a plant would have if he or she were exposed to the maximum pollutant concentrations for 70 years and, in the ample of safety decision, to protect the greatest number of persons possible to an individual lifetime risk level of no higher than approximately 1-in-1 million.  These considerations are made for all people regardless of racial or socioeconomic status.  However, in determining whether to require additional standards, these levels are not considered rigid lines, and we weigh the cancer risk values with a series of other health measures and factors in both the decision regarding risk acceptability and in the ample margin of safety determination.  We also consider cost of controls in the ample margin of safety determination.  While we analyzed the demographic makeup of the at-risk populations surrounding the facilities with source categories addressed by this rule making and found slightly higher minority populations in some cases, we ultimately reaffirmed our initial findings that the risks from each source category were acceptable and protected public health with an ample margin of safety. Thus, the results of our demographic analyses only served to inform the public about the demographic makeup of the populations exposed to each of these source categories.

Additional Standards to Address Risks
     
   Comment  20:  Commenter 265 states that EPA should include a low-risk exclusion for facilities that are low risk, which do not require additional regulation, when making residual risk rules.  The rules should include a site-specific determination similar to the applicability determinations for other rules.  The facility would determine initial applicability and later review applicability through a management-of-change process related to capital investments and operations changes, and no on-going permitting or reporting requirements would apply if the rule is not applicable.  The commenter suggests the following for this low-risk applicability determination
   * Facilities would use any scientifically accepted, peer-reviewed risk assessment methodology.
   * The applicability threshold would be set at either a risk level that reflects an ample margin of safety or at a cancer risk less than or equal to 1-in-1 million and a hazard index of less than or equal to 1.
   * A notation would be included in the Title V permit or Title V permit application that the residual risk standards are not applicable to the facility if it is low-risk.
   * Applicability determinations would be available for review by regulatory authorities but would not require a regulatory approval process, which would be burdensome and unwarranted. 
   * Facilities would have the on-going obligation to reassess applicability when facility changes could potentially impact rule applicability.
   * Applicability reassessment would not be required for changes outside the facility or from formal rulemaking processes, such as changes in population and census track data, changes in health benchmarks, improvements and changes in dispersion models.
   * The analysis and modeling would be consistent with the health concerns in the source category, for example, if cancer is the concern, modeling for acute effects would not be required.
   * If EPA determines there are no adverse environmental risks for a source category, there would not be a requirement to demonstrate no adverse environmental risks.
   * The applicability determination would be completed before the first major compliance date after the rule.
   Response  20:  No new requirements are included in the final rules for these source categories to address or reduce residual risk, so these comments are not germane. To the extent the commenter has concerns with actions taken in future residual risk reviews, those concerns should be raised at that time.

	Comment  21:  Commenter 265 states that in calculating cost effectiveness, EPA should evaluate costs on a risk reduction basis rather than a cost per ton of HAP reduced basis, which would tie in better with the goals of the residual risk program.
	Response  21:  Our residual risk decisions are based on the approach in the 1989 benzene decision framework.  In that decision, we stated that the level of the MIR, distribution of risk in exposed population, incidence, science policy assumptions, and uncertainties associated with risk measures, and weight of evidence that a pollutant is harmful to health are all important factors which may be considered in the acceptability judgment (first step). In the second step, we again consider all of the health risk and other health information considered in the first step. Beyond that information, additional factors relating to the appropriate level of control will also be considered, including costs and economic impacts of controls, technological feasibility, uncertainties, and any other relevant factors.  As part of the ample margin of safety analysis, we examined HAP emissions control options, the emissions and risk reductions that would be realized with those controls, and the costs associated with the controls.  This information was presented in the October 10, 2008 RTR proposal (73 FR 60432) for the Group I Polymers and Resins, Marine Tank Vessel Loading Operations, Pharmaceuticals Production, and the Printing and Publishing MACT standards.  While the cost effectiveness information was not presented on a cost per risk reduction basis, the capital and annualized costs of the risk reduction associated with reduced emissions due to controls were considered and presented in that proposal.   

Removal of the SSM Exemption

      Comment  22:  Commenters 270 and 275 support the removal of the malfunction exemption.  Commenters 265 and 359 support EPA's edits to render the SSM references in the General Provisions not applicable to the MTVLO MACT, given that the existing rule does not require a SSM plan and the SSM exemption has be vacated by the courts. 
      Response  22:  EPA thanks the commenter for its support.
      Comment  23:  Commenter 265 supports EPA's proposal to address malfunctions as distinct from routine operating conditions but objects to the manner in which EPA proposes to do so.  Commenter 288 states that EPA contends that SSM periods should not be considered distinct operating modes without providing data or soliciting information to support this claim.  Commenter 284 states that EPA has a burden under the CAA to show that compliance is achievable at all times, but it has not demonstrated this for periods of startup and shutdown.  The commenter suggests that in lieu of establishing alternative numerical emissions standards for these events, the Agency should pursue other options.  One suggestion the commenter had would be to not require compliance with the emission limits for a short period of time during these events because the facilities are not operating under normal conditions.  Another suggestion the commenter presents is to establish an affirmative defense work practice standard in each MACT that would be based on minimizing emissions according to a source's SSM plan.  As part of this option could be the requirement to conduct further analyses to determine whether emissions from startups and shutdowns are preventable in the future.
      Response  23:  As explained in more detail in the response to comments section of the preamble to the final rule, EPA's approach to malfunctions is consistent with section 112 and is a reasonable interpretation of the statute.  Based on the information available at the time of proposal, EPA determined that for each of the source categories subject to this final rule, sources could meet the underlying limit during periods of startup and shutdown.  For all source categories, with two exceptions, we did not receive any specific information indicating that sources could not meet the applicable emissions limitations during periods of startup and shutdown that would justify setting alternative limits for such periods.  
      We did receive a comment for the Group I Polymers and Resins MACT standards that compliance with the standards for 98 percent organic HAP destruction during periods of shutdown for front-end vents in the Butyl Rubber and Ethylene Propylene Rubber source categories would not be possible for certain equipment (i.e., flares).  The commenter stated that they may not always be able to route these process vents to a flare during periods of shutdown due to the low pressure or low heating value in the process vent.  EPA agrees with the commenter that it is not possible to comply with the current standard during periods of shutdown.  Therefore, we are promulgating an alternative standard for front-end vents in these categories for these shutdown periods.
      
      Comment  24:  Commenter 297 states that in its experience with and research into petroleum refineries, many past malfunctions were preventable, caused by a failure to conduct more frequent inspections and testing and replacement of systems, equipment and parts.  The commenter also states that not having emergency back-up power also contributed to numerous and regular malfunctions.  The commenter requests that emissions from these malfunctions be included in the development of the MACT standards.
      Response  24:  EPA agrees that some equipment or air pollution control failures are preventable.  Today's final rules remove the SSM exemption and facilities will now be required to comply with the established standards at all times, including periods of malfunction.  As any malfunction resulting in emissions that exceed the established standards will cause a violation of the standard, we expect a significant decrease in the occurrence of preventable malfunctions.  
                                          
Affirmative Defense Provisions
      
      Comment  25:  One commenter (263) supports EPA's proposal to provide an affirmative defense against civil penalties and the proposal to set criteria for establishing the affirmative defense, but has concerns regarding the specific criteria included in the provisions.
      Response  25:  EPA thanks the commenter for its support.  Comments concerning the affirmative defense criteria are addressed below.  

      Comment  26:  Commenter 295 objects to EPA's adoption of the "enforcement discretion" policy as it applies to periods of SSM.  The commenter notes that EPA quotes with approval the D.C. Circuit's decision in Weyerhaeuser v. Costle, 590 F.2d 1011, 1058 (D.C. Cir. 1978) that says the Agency is not bound to anticipate all upset situations and may deal with them by "enforcement discretion."  The EPA should account for why it is entitled to exercise enforcement discretion, which is unbridled discretion in the hands of the government on whether to enforce or not, when this prerogative is denied to every other agency and instrumentality of government in the United States.
      Response  26:  The preamble discussion the commenter appears to be referring to relates to EPA's approach to periods of malfunction, not to periods of startup and shutdown.  As explained in more detail elsewhere in this response to comment document, EPA's approach to malfunctions is consistent with section 112 and is a reasonable interpretation of the statute.  Moreover, EPA disagrees with commenter's statements concerning enforcement discretion.  All government agencies, including EPA, can and must prioritize in determining how to allocate limited agency resources, including enforcement resources.  
      
      Comment 27:  Commenters 265, 286, 287, 288, 289, and 291 state that the affirmative defense provisions do not cure EPA's failure to set emission standards that are achievable during SSM events and are not a substitute for setting emissions standards for these periods.  Commenter 288 states that the affirmative defense provisions are inappropriate for regulations that are required to be achievable, clear, and that a source can comply with as specific in President Clinton's E.O. 12866.  The commenters further state that the proposed standards with the affirmative defense do not represent emission limitations achieved by the best-performing existing sources, nor do they meet the criteria for establishing beyond-the-floor emission standards.  Commenters 287 and  291 state that the affirmative defense is not a permissible substitute for setting standards for periods of malfunction  and is insufficient to meet EPA's obligations under CAA §112.  Commenter 287 requests EPA set aside the proposed affirmative defense for periods of malfunction and establish a work practice standard for these periods.
      Response  27:  The affirmative defense is not intended as a substitute for standards under section 112.  The affirmative defense applies to malfunction events and not to periods of startup and shutdown.  EPA explained the basis for its approach in the preamble and elsewhere in this response to comment document.  Comments concerning the affirmative defense criteria are addressed below.  
	
      Comment  28:  Commenter 291 states that EPA has previously found a case-by-case basis consideration of potential enforcement to be a less desirable route to address malfunctions. 37 Fed. Reg. 17,214, 17,214 (Aug. 25, 1972).  As the Supreme Court has found, "a reasoned explanation is needed for disregarding facts and circumstances that underlay or were engendered by the prior policy." FCC v. Fox Television Stations, Inc., 129 S. Ct. 1800, 1811 (2009) (citation omitted).  EPA failed to provide any explanation as to why it now believes a case-by-case affirmative defense is a more appropriate route.  Commenter 265 also points out that the determinations in a citizen suit would be made by a lay judge rather than environmental experts at EPA.
      Response  28:  The Federal Register notice cited by commenters is almost four decades old and does not reflect subsequent amendments to the Clean Air Act and court decisions.  In any event, the Federal Register notice merely states EPA's preference at that time for adopting regulations in order to formalize an approach to excess emissions caused by malfunctions.  EPA explained its view that such an approach was preferable to an informal enforcement discretion approach for the following reasons:  "First, the existence of a formal process better informs the public of the policy and factual issues which will underlie enforcement of the standards. Second, affected industries which are making good-faith efforts to meet the standards will on the whole welcome a regularized means of informing the Agency in detail of the circumstances surrounding unavoidable emission excesses. Third, the Agency expects to benefit substantially from the information it will gain about the operation of the processes in question, for both future enforcement and standard setting." 37 Fed. Reg. 17,214, 17,214-15 (Aug. 25, 1972).  The affirmative defense is not an informal enforcement discretion approach of the type that EPA rejected in 1972 and provides the benefits associated with the formalized approach that EPA identified in its 1972 proposal.  
      Further, EPA believes that courts are well equipped and often do evaluate and apply the type of criteria set forth is the affirmative defense.  Many of the conditions were modeled after the conditions of the affirmative defense in EPA's SIP SSM policy, which several states have adopted into their SIPs. (See, e.g., State Implementation Plans: Policy Regarding Excessive Emissions During Malfunctions, Startup, and Shutdown (Sept. 20, 1999); Policy on Excess Emissions During Startup, Shutdown, Maintenance, and Malfunctions (Feb. 15, 1983)). We do not have any indication that parties to enforcement proceedings have had any significant difficulties applying the terms of these SIP affirmative defenses.   
      
      Comment 29:  While commenters 270 and 290 support EPA's removal of the malfunction exemption, the commenters do not support the inclusion of the affirmative defense provisions.  These commenters feel the proposed affirmative defense approach reduces the disincentive to emit pollutants, diminishes the value of the standard, and may affect the ability of citizens to seek civil penalties from violators of standards under the Clean Air Act.  For these reasons, the commenters urge EPA not to establish the affirmative defense provisions and to ensure full public reporting when malfunctions do occur.  Commenter 290 adds that the provisions will reduce the civil enforcement mechanisms available to citizens, which is contrary to Congress's intent when establishing the CAA, and it is outside of EPA's authority to decide when civil penalties will not be allowed.  The commenter states that if EPA did have the authority to promulgate affirmative defense provisions, the following provisions should be added: 
         * A specific amount of compensatory damages that will apply to each reported malfunction, the funds of which would be dedicated to enforcement, inspections, and monitoring in the area around the specific facility to create greater assurance that the malfunction will not recur.
         * A provision that the affirmative defense can only be used once by a specific company or facility within a certain period of time, such as 10 years.
         * Specific public reporting and notification requirements for malfunctions or any emissions exceedance of which the operator is aware, including a requirement that any notification of a malfunction or emissions exceedance provided to EPA should be made public on EPA's website within 14 days.  Also, when these notifications are made, the facility must also provide community notification within 2 business days through an appropriate public forum, such as notices on the facility's website, a written notice to the local municipality and school district, and/or a press release to the local newspaper, radio and TV news station.
      Response 29:  EPA disagrees with the commenters who alleged that the affirmative defense provision for malfunctions is inconsistent with sections 113(e) and  304 of the Act.  Section 304 gives district courts jurisdiction "to apply appropriate civil penalties." Section 113(e)(1) identifies the factors that the Administrator or a court shall take into consideration in determining the amount of a penalty to be assessed, once it has been determined that a penalty is appropriate.  The affirmative defense regulatory provision is not relevant to the amount of any penalty to be assessed.  If a court determines that the affirmative defense elements have been established, then a penalty is not appropriate and penalty assessment pursuant to the section 113(e)(1) factors does not occur.    
      EPA also disagrees that the affirmative defense provision will hamper citizen enforcement.  First, injunctive relief is still available and the threat of penalties would not deter violations in cases where all of the conditions of the affirmative defense have been satisfied because the affirmative defense criteria ensure that all reasonable steps were taken to prevent a malfunction that causes excess emissions.  Further, litigating whether a source has met the affirmative defense will not burden citizen group any more or less than would litigating the appropriate penalty amount in the penalty assessment stage of a citizen suit enforcement action, because the 113(e) penalty assessment criteria and the affirmative defense criteria are similar and in fact overlap.  For example, the requirement that the Administrator or the court consider "good faith efforts to comply" is bound to generate the type of fact-intensive disputes that the commenter complains of.  In addition, several of the affirmative defense criteria are exactly the type of criteria the Administrator or Court might consider in determining whether a source made "good faith efforts to comply."  For example, to take advantage of the affirmative defense, the source must prove by a preponderance of the evidence that, among other things,  the excess emissions "were caused by an unavoidable failure of air pollution control and monitoring equipment, process equipment, or a process to operate in a normal or usual manner" and "could not have been prevented through careful planning, proper design or better operation and maintenance practices" and "did not stem from any activity or event that could have been foreseen and avoided, or planned for." 
      Thus, EPA does not expect the affirmative defense provision to significantly alter the burden of bringing a citizen enforcement action.  For those cases that do proceed to trial, even in the absence of this affirmative defense, sources generally raise equitable arguments to argue for a low penalty and citizens often rebut such arguments.  Therefore, as a practical matter, EPA does not expect the affirmative defense provision to materially affect the practice of Clean Air Act enforcement.
      Additionally, the affirmative defense is reasonable.  EPA's judgment is that the affirmative defense criteria capture the appropriate considerations in determining whether penalties are appropriate when a violation occurs as the result of a malfunction.  As noted above, the affirmative defense criteria overlap to some extent with the penalty assessment criteria set forth in section 113(e), but are not identical.  For example, size of business is one of the factors listed in section 113(e), but is not reflected in EPA's affirmative defense.  This reflects EPA's view that when a violation is caused by a malfunction, the size of the business is not relevant to whether penalties should be excused.  If the violation was unavoidable and could not have been prevented, EPA's view is that it would be unfair to impose a penalty no matter the size of the business.  
      EPA is not adopting commenters' suggestion with respect to compensatory damages or limits on the frequency of use of the affirmative defense.  It is not clear that EPA has authority to require the automatic imposition of compensatory damages and even if such authority exists, EPA does not think automatic imposition of damages is appropriate.  Ensuring that malfunctions do not recur can be handled through imposition of appropriate injunctive relief.  In addition, EPA's view is that it would not be appropriate to limit a source's ability to take advantage of the affirmative defense to one time over a specified period of time such as ten years given that the affirmative defense is only available when the source could not have prevented the excess emissions.  With respect to commenters' suggested reporting requirements, the reporting requirements in the rules promulgated today already require malfunction reporting and the affirmative defense provisions require that parties choosing to assert the affirmative defense meet additional malfunction reporting requirements.  Any such reports submitted to EPA are publicly available pursuant to CAA section 114.  

      Comment 30:  Commenters 263, 265, 284, 286, 289, and 295 stated that EPA should provide that the affirmative defense apply to injunctive relief.  Commenters 262 and 265 state that injunctive relief is not an efficient means to address malfunctions since the affirmative defense is only available to malfunctions that could not have been prevented anyway.  Commenter 262 adds that through injunctive relief, it appears that there is no cost ceiling that would prevent the EPA from demanding that changes be made to a facility to prevent all malfunctions that result in an emissions exceedance.  Commenters 263, 265, 284, 286, and 289 added that it should also apply to civil penalties, civil administrative penalties, noncompliance penalties, and to actions brought by citizens and state and local permit authorities.  Commenter 284 added that it should protect from criminal enforcement under §113(e), (c)(1), and (c)(5)(A).  Commenters 286 and 289 stated that the affirmative defense is inadequate to meet the purported purpose of the provisions, which is to provide relief from emission limitation that cannot be met at times even with equipment that is properly designed and maintained, unless it covers these other types of claims and penalties.
      Response  30:  The affirmative defense applies to civil penalties, including civil administrative penalties and penalties under section 120, but does not apply to injunctive relief or criminal penalties.  The affirmative defense is available in any civil action to enforce the standards set forth in this rule, whether such action is brought by EPA, a state or local authority or a citizen.  EPA agrees that in some cases, injunctive relief may not be appropriate if all the criteria of the affirmative defense have been satisfied.  In such cases, liability for attorney's fees is not a real issue.  However, some form of injunctive relief may be appropriate.  The Sierra Club v. Otter Tail Power Co case cited by commenters is not on point and does not undermine EPA's ability to limit the affirmative defense to penalty claims. The concurrent remedy doctrine provides that where a party's legal remedies are time-barred a party's concurrent equitable claims generally are barred.  The affirmative defense is not a time-bar to civil penalties.  EPA does not intend to provide an affirmative defense to criminal charges as the very nature of criminal charges is that there is alleged to be a knowing violation (CAA Section 113(c)(1-3), or a knowing or negligence release that threatens imminent danger to persons (CAA Section 113(c)(4)and (5)).  In these circumstances, it is likely that the source did not do all that could be done to prevent the excess emissions.  However, if there are circumstances that justify a knowing release as is suggested by the commenter, such circumstances can be raised in any enforcement action and may be the basis for a defense.

      Comment  31:  Commenter 263, 265, 284, 286, 289, 291, and 295 believe the affirmative defense provisions should apply to the potential violation of emission limits during a startup or shutdown event as well as during malfunctions, if EPA is to use such affirmative defense. 
      Response  31:  EPA is promulgating the affirmative defense for malfunctions only, not for periods of startup and shutdown.  EPA believes that malfunction events should be treated differently than periods of startup and shutdown, which are predictable and routine aspects of a source's operations.  In contrast, EPA has determined that CAA section 112 does not require that emissions that occur during malfunctions be factored into development of CAA section 112 standards.  Because startup and shutdown periods are part of a source's normal operations, the same approach to compliance with, and enforcement of, applicable emissions standards during those periods should apply as otherwise applies during a source's operations.  Further, as explained above, periods of startup and shutdown  -  but not malfunctions -- are taken into account when establishing section 112 emissions standards.  For these reasons, EPA does not believe it is appropriate to apply the affirmative defense provision to startup and shutdown periods.

      Comment  32: Commenter 291 states that it is unclear why EPA has not followed the long-standing upset defense at 40 CFR 122.41(n) promulgated under the Clean Water Act, which provides that an upset constitutes an affirmative defense and establishes criteria for the defense.  These criteria include:
         * An upset occurred and the permittee can identify the cause(s) of the upset;
         * The permitted facility was at the time being properly operated;
         * The permitted facility submitted notice of the upset (required within 24 hours);
         * The permittee took all reasonable steps to minimize or prevent the violation.
The commenter suggests that if an affirmative defense is to be used, EPA should eliminate the criteria in the proposal and require the regulated entity to establish that a malfunction occurred using the definition in 40 CFR 63.2.  Alternatively, the commenter suggests using the Clean Water Act criteria as a model for the affirmative defense requirements, rather than the 1983 and 1999 EPA memoranda on excess emissions, which do not adequately address the treatment of SSM events in technology-based standards. 
      Commenters 262, 264, 265, 284, and 291 noted that the proposed affirmative defense language is not consistent with the definition of malfunction in 40 CFR 63.2 or those specified in specific subparts, and they believe EPA should revise the affirmative defense language to conform with the longstanding definition of "malfunction."  For instance, commenter 264 noted that in the proposed affirmative defense language in §63.1250(g)(4)(i)(A), excess emissions must be caused by an event that, among other things, is short in duration.  The commenter points out that the definition of malfunction in §63.1251 does not specify the duration of the malfunction event.  To conform to the definition of malfunction, commenter 265 suggests the following language for the standards in equivalent paragraphs to §63.480(j)(4)(i)(A):
      The excess emissions were caused by a sudden, infrequent, and not reasonably preventable failure of air pollution control and monitoring equipment, or a process to operate in a normal and usual manner; and that the failure was not caused in part by poor maintenance or careless operation.
      Commenter 262 noted that the affirmative defense provisions essentially require an owner or operator to design, operate and maintain a facility without the benefit of hindsight that comes after a malfunction occurs.  The commenter states that this is a fundamental flaw in the affirmative defense approach.
      Commenters 262, 273, 286 and 289 state that if EPA retains the affirmative defense, the Agency must modify it substantially so that it does not require the source to prove its innocence (before it is even charged).  The commenters are concerned that the affirmative defense provisions imply the facility is guilty until proven innocent and put the burden of proof on the facility rather than the Agency.  Commenter 273 suggests EPA establish a rebuttable presumption rather than an affirmative defense where it is presumed the facility did everything in its power to minimize emissions during SSM events unless the Agency proves certain facts to show that it did not take reasonable actions to minimize emissions. Commenters 286 and 289 state that EPA has not provided any statutory authority or real justification for requiring a source to prove its innocence, producing all documentation within 30 days of the event, without ever being charged.  The commenters state that the ordinary default rule is that plaintiffs bear the risk of failing to prove their claims; however, these provisions require a facility to prove its innocence rather than requiring an enforcement authority to prove a violation of the CAA.  The commenters suggest that if EPA adopts an affirmative defense approach, it should allow that once a source has claimed its excess emissions were related to a malfunction, it will not be considered to be in violation of the standards unless the enforcement authority demonstrates that the source is not entitled to claim the malfunction.  These commenters and commenter 265 suggest that the provisions state that a source "will not be deemed in violation of" the MACT standards for excess emissions associated with SSM, unless the event, and the source's response to the event, do not meet the criteria spelled out in the regulations.  The commenters suggest with these changes, the affirmative defense provisions could then be referred to as an "alternative standard for SSM events."  The commenters also suggest that with this change, it may be unnecessary to state also that the affirmative defense relieves the source from liability for all types of penalties and injunctive relief (save criminal penalties). 
      Commenter 284 states that the affirmative defense provisions should be amended because they do not meet the Agency's burden to demonstrate that compliance with a MACT standard is continuously achievable.  The commenter states that in lieu of being able to establish an emission limitation, EPA should clarify that the affirmative defense provisions are work practices that are part of the standard itself.  Commenters 265 and 359 also suggest that the affirmative defense provisions should be specified as work practice standards.  These commenters also suggest EPA provide an allowance for a reasonable period of routine maintenance for the control device.
      Several commenters (262, 264, 265, 273, 284, 286, 288, 289, 291, and 295) stated that some of the affirmative defense requirements are unreasonable, difficult to prove, and subject to varying interpretation.  Several commenters requested that the term "infrequent" be removed from the regulatory text, as it could be subject to different interpretations by regulatory agencies, and it should be evaluated on a case-by-case basis.  Another suggestion is that the sentence regarding the use of "off-shift and overtime labor" be deleted to allow facilities the flexibility to respond to the matter in the most appropriate manner.  Other suggestions offered were to delete the word "severe" in the descriptions of personal injury and property damage, as the word is subjective and open to a wide range of interpretations; and further define "facility" in the context of the Affirmative Defense requirements.  Further suggestions include the removal of terms such as "careful," "proper," and "better," in §63.342(b)(1)(i)(A), which are not defined and would make it difficult to determine whether the criteria have been met.  Several commenters also suggest removing references to "any" activity and "all" in the affirmative defense language, which would make it difficult to ever satisfy the affirmative defense. Several commenters suggest EPA reword the affirmative defense to state that a source is "not in violation" if it meets the requirements of the affirmative defense.  Commenter 284 suggests that instead of requiring that excess emissions "did not stem from any activity or event that could have been foreseen and avoided" it should read as the definition of malfunction and rely on whether the event was caused by "poor maintenance" or "careless operation."  This and other commenters also suggest the phrase "expeditiously as possible" be replaced by the phrase "expeditiously as practicable" or "in an expeditious manner" and that the phrase "if at all possible" be replaced by "to the maximum extent practicable."  Commenters 286 and 289 requested the removal of the phrase "if at all possible." Commenter 284 also requests the provisions not include the requirements that a bypass was unavoidable to "prevent loss of life, severe personal injury, or severe property damage" as this is an evidentiary consideration for purposes of prosecution and is not proper to include as a recordkeeping and reporting requirement.  However, if this phrase remains in the rule, the commenter requests it also include "severe economic loss."  Commenters 262, 264, 286 and 289 recommended the removal of the term "short."  Commenters 264, 286, and 289 recommended that EPA clarify that the "affected source" rather than the "facility" was operated in a manner consistent with good practices for minimizing emissions.  Commenter 262 objects to using the term "unavoidable" and suggests using "not reasonably preventable."  Commenter 262 adds that these provisions regarding when it is appropriate to bypass emissions inappropriately favors reducing emissions over preventing human injury and suggests that if a limitation on what constitutes an injury is needed, EPA should utilize the OSHA for a "serious" rather than "severe" injury.  Commenter 265, citing issues with the phrasing of the language of §63.480(j)(4)(i)(B) and the same language in the other standards, the commenter recommended this paragraph not be finalized and that the proposed §63.480(j)(4)(i)(C) remain as the basis for demonstrating an appropriate response to the malfunction.  This commenter also recommended that §63.480(j)(4)(i)(E) be removed because it is impossible to demonstrate that "all possible steps" were taken to minimize impacts of emissions.
      Commenters 265 and 284 state that the affirmative defense provision to document excess emissions activities in "properly signed, contemporaneous operating logs," should be deleted.  The commenters state that during a malfunction engineers have little or no time to keep contemporaneous records, a source should be allowed to prove its entitlement to an affirmative defense using any credible evidence which would otherwise be admissible in court, sources should not be required to maintain an operating log which is not required by the underlying NESHAP, and sources should be directing resources to minimize excess emissions and correct malfunctions rather than documenting its actions.  In addition, the commenter states that the term "properly signed" lacks a definition and should be removed if this provision is retained.  Commenter 265 recommends the following language be used in place of this requirement in 63.480(j)(4)(i)(G):
      Records are maintained documenting the event, including actions taken to minimize emissions and the calculation of the amount of excess emissions.
      Commenters 265 and 284 state that the provision, "At all times, the facility was operated in a manner consistent with good practices for minimizing emissions," should be deleted.  The commenters state that the definition of malfunction itself already excludes situations that are caused by careless operation.  Commenter 284 states that, at a minimum, EPA should clarify that this requirement only applies during times of malfunction and that the duty to minimize emissions does not require emissions to be reduced to below the applicable emissions standard.  Commenter 265 states that this requirement is duplicative of the general duty to minimize emissions, which is not included in the individual standards proposed, and deals with the entire facility rather than just the malfunctioning equipment. 
      Commenters 264 and 265 recommend that 2-day initial requirement be deleted and leave only a 180-day written demonstration requirement.  Commenter 265 recommends the following language for the notification requirement (taken from the Polymers and Resins I MACT language):
      (ii) Notification.  The owner or operator seeking to assert an affirmative defense shall submit a written report to the Administrator within 180 days of determining that an exceedance has occurred to demonstrate, with all necessary supporting documentation, that it has met the requirements set forth in paragraph (4)(i) of this subsection.
      Commenter 273 suggested the following language that would apply to all MACT standards subject to this final rule to incorporate its comments regarding a rebuttable presumption, ambiguous terms, root cause analyses, and communicating electronically in § 63.342(b)(1)(i):
      To establish a rebuttable presumption in any action to enforce such a limit, the owners or operators of facilities must timely meet the notification requirements of paragraph (b)(1)(ii) of this section.  The Administrator must prove by a preponderance of evidence that:
      (A) The excess emissions were not caused by a sudden, short; infrequent, and unavoidable failure of air pollution control and monitoring equipment, or of a process to operate in a normal and usual manner; and could not have been reasonably prevented through careful planning, proper design or better operation and maintenance practices; and stemmed from any activity or event that could have been reasonably foreseen and avoided, or planned
for; and were part of a recurring pattern indicative of inadequate design, operation, or maintenance; and 
      (B) If the applicable emission limits were exceeded, repairs were not made as expeditiously as possible, including using off-shift and overtime labor to the extent  practicable; and 
      (C) The frequency, amount and duration of the excess emissions (including any bypass) were not minimized to the maximum extent practicable during periods of such emissions; and
      (D) If the excess emissions resulted from a bypass of control equipment or a process, then the bypass was not needed to prevent loss of life, severe personal injury, or severe property damage; and
      (E) Reasonable steps were not taken to minimize the impact of the excess emissions on ambient air quality, the environment, and human health; and
      (F) Emissions monitoring and control systems were not kept in operation if possible; and
      (G) Your actions in response to the excess emissions were documented by properly signed, operating logs; and
      (H) The facility was not operated in a manner consistent with good practices for minimizing emissions; and
      (l) The owner or operator has prepared a written report to determine and mitigate the primary causes of the malfunction and the excess emissions resulting from the malfunction event at issue.  A root cause analysis may be required only if the cause of the malfunction is difficult to determine or if the serious nature of the event indicates one is needed.  Facility personnel will have the discretion to make this determination.  The analysis shall also specify, using the best monitoring methods and engineering judgment, the amount of excess emissions that were the result of the malfunction.
      (ii) Notification. The owner or operator of the facility experiencing an exceedance
of its emission limit(s) during a malfunction shall notify the Administrator by telephone, facsimile (FAX) transmission, or electronic means as soon as possible, but no later than two business days after the initial occurrence of the malfunction, if it wishes to avail itself of the rebuttable presumption to civil penalties for that malfunction. The owner or operator seeking to assert an affirmative defense shall also submit a written report to the Administrator within 30 days of the initial occurrence of the exceedance of the standard in this subpart to demonstrate, with all necessary supporting documentation, that it has met the requirements set forth in paragraph (b)(1)(i) of this section.  If the report requires a root cause analysis, the report must be submitted within 90 days of the initial occurrence of the event.
      Response  32:  Many of the conditions in the affirmative defense are modeled after the affirmative defense in EPA's SIP SSM policy, See, e.g., State Implementation Plans: Policy Regarding Excessive Emissions  During Malfunctions, Startup, and Shutdown (Sept. 20, 1999); Other conditions are modeled after a Federal Implementation Plan promulgated by EPA.  (40 C.F.R. 50.1312).  EPA's view is that use of consistent terms in establishing affirmative defense regulations and policies across various CAA programs will promote consistent implementation of those rules and policies.  The commenter does not explain its view that the terms of the SIP policy are not appropriate to address malfunctions in the CAA section 112 context.  The affirmative defense does not "require a facility to prove its innocence rather than requiring an enforcement authority to prove a violation of the CAA" or change the burden of proof with respect to establishing a violation.  The affirmative defense applies only where a violation has been established.  The burden of proof that a violation occurred remains with the plaintiff in an enforcement action.  See, e.g., 40 C.F.R. 22.24.  If a violation has been established and a source wishes to assert the affirmative defense with respect to penalties, the source does bear the burden of establishing that the elements of the affirmative defense have been met.  This burden-shifting is appropriate because the source is in a better position to determine the facts required to establish the defense. See, e.g., Arizona Pub. Serv. Co. v. EPA, 562 F.3d 1116, 1120, 1129-30 (10th Cir. 2009) (rejecting industry challenge to EPA's use of an affirmative defense to address excess emissions during malfunction events.).  The notification and reporting requirements are not unreasonable and are consistent with other malfunction reporting and notification requirements that have been applied without significant difficulty.  (See, e.g. 40 CFR 63.6(e)(3)(iv) and 63.10(d)(5)(ii)).  Section 114 of the CAA clearly provides authority to  require such reports. 
      EPA does not agree that the affirmative defense criteria are overly vague and will result in varying interpretations.  Courts are well equipped and often do evaluate and apply criteria that are subject to differing interpretations.  As noted above, many of the conditions were modeled after the conditions of the affirmative defense in EPA's SIP SSM policy, which several states have adopted into their SIPs.  We do not have any indication that parties to enforcement proceedings have had any significant difficulties applying the terms of these SIP affirmative defenses. However, EPA agrees that some of the terms or phrases in the regulatory text establishing the affirmative can be revised or streamlined to some extent.  EPA is revising the affirmative defense language to delete "short" from §63.820(c)(1)(i), because other criteria in the affirmative defense require that the source assure that the duration of the excess emissions "were minimized to the maximum extent practicable." In addition, EPA is revising §63.820(c)(1)(vi) and §63.562(e)(7)(i)(H) to add "consistent with good air pollution control practice for minimizing emissions" and "consistent with safety and good air pollution control practices" as suggested by the commenter.  We are also deleting the term "severe" in the phrase "severe personal injury" in §63.820(c)(1)(iv) because we do not think it is appropriate to make the affirmative defense available only when bypass was unavoidable to prevent severe personal injury.  We have revised sections 63.823, 63.829(g), and 63.829(h)to refer to the "affected source" which is defined in §63.2.  EPA disagrees that other terms and conditions of the affirmative defense are overly subjective or extreme or need further qualification.  EPA acknowledges that some of the criteria may overlap to some extent and under some circumstances, but does not believe it is necessary to change the requirements.  To the extent there is overlap, the same information can be used to meet the provisions in question.  Finally, EPA is not eliminating the word "infrequent" from §63.820(c)(1)(i), because that term appears in the definition of malfunction in 63.2.  
      With respect to the comment that the affirmative defense should be stated instead in terms that, once a source has claimed that its excess emissions were related to a malfunction, it will not be considered to be in violation of the standards unless the enforcement authority demonstrates that the source is not entitled to claim the malfunction, EPA's does not agree that the affirmative defense should operate as the commenter suggests.  The commenter improperly seeks to shift the burden of proof from the source to EPA.  It is the source, not EPA, that has the information relevant to assess whether a particular event qualifies as a malfunction and meets the affirmative defense.  Furthermore, commenter's suggested approach appears to be similar to the scheme that the court vacated in the Sierra Club case, in that there would be no violation unless the enforcing party (EPA or a citizen), established that the event in question was not a malfunction.  EPA also disagrees with the comments suggesting that the affirmative defense provisions should be specified as work practice standards.  As is explained elsewhere is the response to comment document, EPA is not taking malfunctions into account in setting section 112 standards.  
      With respect to the comment concerning operating logs, EPA believes that log entries should be made contemporaneously, which does not mean, at the exact time that the engineers are taking actions to address the malfunction, but rather immediately thereafter.  That way, more accurate records of the malfunction can be expected, as opposed to records reconstructed long after the malfunction.  "Properly signed" means that the records should be signed by an engineer or other employee with knowledge of the situation, in a legible manner so that enforcement officials can tell who made the record and what their responsibilities were.  

      Comment  33:  Commenter 284 requests clarification whether or not a facility would be required to shut down an emission unit until repairs could be made and under what circumstances a source could continue to operate the unit, given the affirmative defense provision that the "frequency, amount and duration of the excess emissions (including any bypass) were minimized to the maximum extent practicable during periods of such emissions."
      Response  33:  EPA is unable to give specific guidance as to how the criteria would apply in the abstract. However, as noted above, we do not believe that courts will have significant difficulty applying these criteria when presented with the specific facts and circumstances of a malfunction event.
      
      Comment  34:  Several commenters (264 and 273) stated that there should be more flexibility in the affirmative defense requirements regarding when a root cause analysis is required.  These commenters both believed a formal root cause analysis for smaller excess emissions events should not be required and that this step should only be required when there is a very significant or repeat event or when other reasonable methods fail to show what caused the malfunction.  Commenters 262, 265, and 273 added that it is reasonable to develop a simple report of the cause of a malfunction and the corrective action taken within 30 days, but it may not be possible to complete a root cause analysis in 30 days, and more time, such as 90 days may be needed or 180 days for more complicated situations.  Commenters 286 and 289 state that the root cause analysis requirement may make the affirmative defense unavailable if the source is not able to determine the primary cause of the malfunction.  The commenters state that it should be expected that in many causes the primary cause will not be ascertainable or that it will not be possible to identify a way to ensure the malfunction will not recur.  Commenter 265 added that it should be clarified that the root cause analysis requirement does not require corrective actions within the demonstration period, because long-term corrective action could require facility modifications that could take years to design and implement.  The commenters also state that requiring the facility to eliminate the primary cause of the malfunction without regard to the costs and impacts of this elimination is unreasonable and inconsistent with the criteria for standards established under CAA section 112(d).Commenters 286 and 289 also stated that 90 days or more should allowed unless EPA substantially streamlines the criteria for the affirmative defense.  Commenter 262 suggested that the root cause analysis be submitted as soon as practical after determining that there was an exceedance due to a malfunction.
      Response  34:  EPA is revising this requirement to clarify that the purpose of the root cause analysis is to determine, correct, and eliminate the primary cause of the malfunction and that the root cause analysis itself does not necessarily require that the cause be determined, corrected or eliminated.  However, in most cases, EPA believes that a properly conducted root cause analysis will have such results.  EPA is also revising this requirement to state that a written report must be submitted within 45 days of the initial occurrence of the malfunction.  We further revised this provision to allow the source to seek an extension of up to an additional 30 days.  

      Comment  35:  Commenters 265, 264, and 273 requested that the affirmative defense provisions allow for actions taken in response to excess emissions be documented and submitted in an electronic format.
      Response  35:  EPA accepts documents in electronic format, as long as the format is compatible with the requirements of the standards.  For the affirmative defense provisions, the owner or operator of a facility experiencing an exceedance of its emission limit(s) during a malfunction must notify the Administrator by telephone or facsimile (FAX) transmission of the exceedance.  However, the written reports required to demonstrate that the affirmative defense provisions have been met and requests for an extension of the deadline for these reports may be submitted electronically.  
      Comment  36:  Commenters 262, 265, and 359 state that the notifications and reports required in the affirmative defense provisions are arbitrary, unrealistic, and unwarranted.  The commenters state that in many cases facilities would not be able to complete the determination of whether a given malfunction event qualifies for the affirmative defense within 2 days following the event and would most likely routinely submit the notification to preserve the potential for an affirmative defense.  Commenters 286 and 289 state that with the short malfunction or deviation report sources currently submit under many air programs, the enforcement authority can determine in many cases whether an exceedance of standards resulted from an unforeseen and unavoidable equipment failure or process upset.  The commenters also state that it is inefficient and burdensome for both sources and regulators to require a complete justification of the affirmative defense before the enforcement authority has indicated need for further investigation.  The commenters say this would create unnecessary burden to the facility and regulatory agency with no associated environmental benefit.  Commenter 264 claims the immediate notifications will also be duplicative of emergency response reporting procedures under SARA/CERCLA/EPCRA and many State regulations for larger emission events and will contradict many existing reporting rules by requiring the reporting of very small excess emissions below those of CERCLA/SARA and State Agencies.  The commenters also say that malfunction reports should be submitted with semi-annual reports rather than within 30 days of the event.  The commenters note that facilities are already required to submit reports of excess emissions and continuous monitoring system performance and summary reports on a semi-annual basis. The commenters further note that EPA promulgated the Recordkeeping and Reporting Burden Reduction Rule expressly to simplify and unify reporting schedules, and a semi-annual reporting frequency was chosen.  Commenter 264 notes that the provisions encourage reporting even if the owner/operator is unsure whether excess emissions occurred for fear of not being able to claim the affirmative defense, and the provisions are unclear as to who should receive the reports.  
     Response  36:  The general provisions regulations specify where reports are to be submitted  (40 C.F.R. 63.14).  Further , 40 C.F.R. 63.9(i) contains provisions for adjustments of reporting deadlines that can be utilized in appropriate circumstances.  The requirement to notify the Administrator of an exceedance during a malfunction is reasonable and consistent with many other upset and malfunction notification requirements that have been applied without significant difficulty.  (See, e.g. 40 CFR 63.6(e)(3)(iv) and 63.10(d)(5)(ii)).  


Source Category Specific Comments
Source Category Specific Comment  -  Pharmaceuticals Production

      Comment  37:  Commenter 264 stated that the proposed rule language related to performance testing is too vague and seemed to indicate that EPA intends to write the test plans for the facilities.  The commenter adds that EPA or a delegated authority is already required to approve the test plans, and another layer of oversight is not needed.
      Response  37:  We have revised the language of §63.1257(a).  Performance tests shall be conducted under conditions representative of performance of the affected source for the period being tested.
      
      Comment  38:  Commenter 262 points out that a malfunction of emissions monitoring equipment could lead to a theoretical exceedance because there would be an absence of monitoring data to demonstrate what the emissions were during that time.  The commenter suggests EPA modify the SSM provisions to explicitly preclude monitoring failures from resulting in assumed emission exceedances.
      Response  38:  EPA recognizes that monitoring system malfunctions do not necessarily result in emission greater than allowed by the standard.  The provisions of §63.505(g) allow that monitoring system breakdowns, repairs, calibration checks, and zero (low-level) and high level adjustments are not required to be included in the calculations to determine whether there has been sufficient monitoring data collected.  
      
      Comment  39:  Commenter 298 concurs with the EPA's determination that the current MACT standards for Pharmaceuticals Production provide an ample margin of safety to protect public health and prevent adverse environmental effects and agrees with the EPA's proposal to re-adopt the existing MACT standards to satisfy Section 112(f) of the CAA.
      Response  39:  We thank the commenter for this feedback.
Source Category Specific Comment  -  The Printing and Publishing Industry

      Comment  40:  Commenter 263 noted that most of the risk discussions in the proposal concern pollutants common to the publishing industry but not the flexible packaging industry, and it might be helpful in the future to discuss the 40 CFR part 63, subpart KK subcategories separately.
      Response  40:  We appreciate this feedback and will consider this suggestion as we move forward with other RTR rulemakings.
      
      Comment  41:  Commenter 263 said that its members reviewed the facility information for flexible package plants and did not find data that were of potential concern.  The commenter also noted that its members agreed that mandating permanent total enclosures would be costly for the industry and not warranted by the identified risks.
      Response  41:  We thank the commenter for carefully reviewing the data and confirming our analysis.
      
      Comment  42:  Commenter 272 recommended that EPA clarify the rule applicability and initial compliance dates in §63.820 to state that area source status must be established prior to the compliance dates specified in §63.826 to avoid "once-in-always-in" policy interpretation conflicts.
      Response  42:  This comment is not germane to this rulemaking.  However, we note that the commenter is correct that area source status is  established prior to the compliance dates specified in §63.826.
      
      Comment  43:  Commenter 278 reviewed the data and the estimated cancer and non-cancer risks for the facilities in its jurisdiction and determined there were no additional risks from the Printing and Publishing Industry for facilities located in North Carolina.
      Response  43:  We thank the commenter for carefully reviewing the data regarding facilities in North Carolina.

Source Category Specific Comments  -  Group I Polymers and Resins
      
      Comment  44:  Commenters 265 and 276 support EPA's conclusion that no further residual risk or technology review is required for the Butyl Rubber or Ethylene Propylene Rubber source categories, noting that EPA's previously finalized analysis reached the appropriate conclusion.
      Response  44:  We appreciate the commenters' support of EPA's risk and technology review conclusions.

      Comment  45:  Commenters 265 and 276 state that since there is no consent decree for the Butyl Rubber and Ethylene Propylene Rubber source categories, EPA should take the appropriate amount of time to thoroughly review the comments received and reassess the cost-effectiveness determination for the proposed controls for these source categories.
      Response  45:  We believe that we have taken the appropriate amount of time to thoroughly review the comments on these source categories and have reassessed our control requirements and cost estimates accordingly.  We do not believe it is necessary to delay final action on the proposed controls, or the amendments thereto that we are promulgating today.

	Comment  46:  Commenter 288 supports EPA's conclusion that the MACT floor should be based on current emission rates and that the beyond-the-floor option for back-end process vents is not cost effective for the neoprene source category.
      Response  46:  We appreciate the commenter's support of EPA's conclusions for the neoprene source category.

      Comment  47:  Commenter 293 agrees with and incorporates the comments by IISRP by reference.
      Response  47:  We have addressed all of the comments from the IISRP.

      Comment  48:  Commenter 265 states that on the 40[th] anniversary of the Clean Air Act, Administrator Lisa Jackson remarked that EPA would set standards that make the most sense, focusing on getting the most meaningful results through the most cost-effective measures.  The commenter states that the proposed controls for the Butyl Rubber and Ethylene Propylene Rubber source categories are inconsistent with this statement, as they are not cost effective, produce no substantive environmental benefit, and reduce the competitiveness of U.S. manufacturing.
      Response  48:  We have reanalyzed the Butyl Rubber and Ethylene Propylene Rubber source categories.  As a result, we are promulgating standards at the MACT floor level of control, which we believe comply with our CAA section 112 authority and are cost-effective, without reducing the competitiveness of U.S. manufacturing.
      
      Comment 49:  Commenter 265 notes that it is unclear what the cost-effectiveness criterion is for decision making to require beyond-the-floor controls but it appears to be $10,000/ton, also noting that for the Butyl Rubber and Ethylene Propylene Rubber source categories there are few if any co-pollutant reduction benefits.
      Response  49:  EPA does not have a set criterion in beyond-the-floor analyses for determining whether or not a control option is cost effective.  Rather, control options are assessed based on a number of factors, including but not limited to cost-per-ton of HAP reduction and co-pollutant reduction, energy impacts, and non-air quality impacts of controls, as appropriate under CAA section 112(d)(2).
      
      Comment  50:  Commenter 265 notes that there is increasing competitiveness in the polymers markets driven by demand for the products in Asia and keeping costs low is important to remaining competitive in a global market.
      Response  50:  We understand that the polymers market is a global business and it is important for United States manufacturers to remain competitive.  We believe that the promulgated standards will result in no major additional burden to the affected facilities, and should have no impact on their ability to remain competitive in a global market.
      
      Comment  51:  Commenter 265 states that there are no data in the record to support the supposition that dioxins and furans are formed when chlorinated hydrocarbons are burned in a flare or that, if this were true, less of these compounds would be formed in a thermal oxidizer/scrubber system, which EPA proposes to require instead of a flare.  The commenter states that the basis for the proposed controls is HCl generation and there is no basis to suggest any benefit related to dioxins and furans.  The commenter states that the preamble statement regarding dioxins and furans should be withdrawn in the final rule preamble.
    Response  51:  Dioxins and furans can be formed as combustion by-products when chlorinated organic materials are burned.  However, the commenter is correct in that we have no data to indicate that these compounds are being formed specifically at the affected facilities in the Butyl Rubber and Ethylene Propylene Rubber source categories.
    
     Comment  52:  Commenter 265 notes that EPA did not indicate the compliance date related to the removal of the SSM provisions in the Group 1 Polymers and Resins MACT standards and requests four years for compliance so that facilities can invest in equipment, develop work practices, test the work practices and train personnel.  The commenter notes that startups and shutdowns occur infrequently, so it will take an extended time to develop and test the work practices.  The commenter further notes that EPA should provide four years to comply with emission limits that require capital investment to install controls due to the current widespread demand for combustion controls, combustion engineers, company resources, materials, and construction resources to meet the compliance deadlines from other EPA rulemakings related to the boiler/process heater MACT, utility air rules, and other air toxics rules.  The commenter notes that the one-year extension of the three-year period allowed for compliance is provided for in CAA section 112(i)(3)(B).  The commenters also requested four years for compliance even for the MACT emission limits that are set at the floor level of control based on current operations. The commenters stated that the additional time would be needed due to additional capital investments they would need to make because the proposed emission limits did not take into account variability in operations, startup and shutdown, and maintenance on the existing RTOs. 
    Response  52:  These comments were made in regard to our proposal to go beyond-the-floor for certain front-end and back-end process vents in the Butyl Rubber and Ethylene Propylene Rubber source categories.  However, we have decided to promulgate standards at the MACT floor level of control for these source categories, which will not require the installation of any new control equipment.  Therefore, we do not believe that it is appropriate to allow four years to comply with a standard that requires no capital expenditure.  Additionally, we have set the first compliance demonstration for these MACT floor levels of control to be two years following the date of the final rule.  As such, we believe that the affected facilities have an ample amount of time to comply with the new standards.  Regarding the compliance time for the removal of the SSM provisions, we have established a separate standard for certain front-end process vents during periods of shutdown for existing sources in the Butyl Rubber and Ethylene Propylene Rubber source categories.  This standard requires no additional control equipment or capital expenditure, so we do not believe that four years is necessary to comply either.  We believe that all other affected emission points are already complying with the current MACT standards during periods of startup and shutdown, and so four years is not needed to comply. Facilities have also had practices in place to minimize emissions during malfunctions at least since the compliance date of the MACT standards, and since the affirmative defense  against civil penalties for exceedances of emission limits that are caused by malfunctions is available immediately, we believe facilities will either be able to meet the emission standards during malfunctions or will be able to use the affirmative defense upon the effective date of this rule
    
Source Category Specific Comments -- Marine Tank Vessel Loading Operations

      Comment  53:  Commenter 297 states that EPA failed to assess simultaneous multiple loading and unloading and other nearby facilities in its risk assessment for marine terminals and requests that EPA do so.
 	Response  53:  In the acute analysis, we include a factor to estimate the approximate maximum short-term emissions.  This factor would account for times of loading and unloading at the facility.  While there may be HAP emitted by other sources at the same time as these peak emissions, we are concerned about the uncertainties of including estimates of cumulative risks in our assessments because we have not conducted a detailed technical review of HAP emissions data for sources contributing to background concentrations or facilities that have not previously or are currently undergoing a RTR.   
     
      Comment  54:  Commenter 290 states that EPA ignores the unacceptable facility-wide risk at some facilities where MTVLO occur in the current proposal.  The commenter further states that it is necessary for EPA to adopt additional control measures to address the risks of MTVLO or at least commit to reopening and revising the MTVLO standard within a specific timeframe as necessary to address facility-wide risks.
      Response  54:  We performed the facility-wide analysis to put the source category risks in context and to provide us with information to help inform our decision on whether it is necessary to revise the MACT standards under CAA section 112(f)(2).  At proposal, we identified two facilities with facility-wide MIRs greater than 100 - in-1 million (each with values of 200).  Using new data obtained since proposal, we corrected the errors noted by the commenters for both MTVLO emission sources and other emission sources at the facilities.  We found incorrect latitudes and longitudes for some emission sources, incorrect emissions reported for some sources, or incorrectly identified HAP.  We updated the 2005 NEI data sets for each facility with corrected data and conducted a reanalysis of the risk using the corrected data set.  The revised risk assessment results show no facilities with MTVLO have a facility-wide risk of greater than 100-in-1 million
      
      Comment  55:  Commenter 290 states that EPA must address risk at the facility-level for MTVLO at refineries, and it is unlawful for EPA to defer its residual risk making duties for this source category by dividing the source category into different segments.  The commenter states that EPA may address MTVLO at refineries in the petroleum refinery rulemaking also since the refineries standards incorporate the MTVLO standards by reference, but not as a replacement for satisfying its legal duties regarding MTVLO now.
      Response  55:  We perform and make our decisions from the risk and technology reviews on a source category or subcategory basis.  The MTVLO that occur at petroleum refineries are not included in the MTVLO source category but are included in the Petroleum Refinery source category and are regulated under 40 CFR 63 subpart CC.  These MTVLO will be evaluated along with the other HAP emissions sources at petroleum refineries when that source category is reviewed.   

      Comment  56:  Commenter 297 requests that a specific amount of HAP that represents a low risk level be established in a format such as parts per million per cubic meter, rather than a percentage by which HAP emissions must be reduced.  The commenter notes that there is a large difference between being exposed to 3% of 1 ton of emissions and 3% of 200 tons of emissions.  The commenter also states that the 97% control requirement also does not identify and prioritize HAP that are more toxic or carcinogenic compared with those that are only irritants and notes that the cumulative impact of exposure to multiple marine terminals releasing the same HAP should be considered.   
      Response  56:  In our risk analyses, risks are estimated by analyzing the amount of each HAP people are exposed to, the toxicity of each HAP at these exposure levels, and the aggregate risk from exposure to all of these HAP emitted by facilities in the source category.  By including toxicity information for each HAP, the analysis does consider the relative toxicity or carcinogenicity of the HAP.  Also, exposure to all HAP from all facilities in the source category are considered in estimating risk, which includes possible overlap in exposure to HAP from multiple facilities in the source category.  These estimated risks represent conditions after implementation of the MACT standards.  If the risks are low, then the control requirements of the MACT are effective and do not need to be revised to reduce risks, regardless of the form of the standard or the amount of emissions that are still released under that standard.  If the risks are not low after implementation of the MACT, risks under additional regulatory requirements are estimated.  In these additional analyses, risks remaining after the additional control are estimated including all the emissions remaining after additional control is instituted.  Additional requirements are selected to reduce risk to acceptable levels and to provide an ample margin of safety to protect public health.  The format of the standard is not relevant as long as it provides that these goals are met.  However, priority is placed on the format for which it is easiest and most cost effective to ensure and demonstrate compliance.  For these reasons, a percentage by which HAP emissions must be reduced is often preferable to a format such as HAP parts per million allowed per cubic meter.
      
       Comment 57:  One commenter (280) stated their petroleum refinery operates an "Offshore Loading Terminal" and it is not clear if the proposed changes to subpart Y for offshore loading terminals are required for refineries.  The commenter recommended rule text changes for §63.560(d)(6). 
      Response  57:  The current rule excludes offshore loading facilities from controls in §63.560(d)(6).  This final rule will add a requirement for submerged fill for the offshore loading subcategory and for the existing sources with MTVLO emissions less than 10/25 tpy subcategory.  The current rule in §63.561 defines Offshore loading terminal as "a location that has at least one loading berth that is 0.81 km (0.5 miles) or more from the shore that is used for mooring a marine tank vessel and loading liquids from shore."  The definition does not specify the connected facility.  Additionally, the petroleum refinery MACT exclusion under §63.560(d)(3) specifies the exclusion from §63.562(b)(2), (3), and (4), but not the existing source offshore exclusion from §63.560(d)(6).  Thus, there is no apparent connection of refineries and the definition of offshore facilities.  In the proposal preamble, we stated that petroleum refineries would be excluded from the additional controls (75 FR 65112/c2).  Thus, we are adding to the final rule in §63.560(a)(4) that submerged fill is not required at petroleum refineries. 
	
Comment  58:  Since every petroleum-based product has HAP, one commenter (297) requested that this rulemaking include public health protection from exposure to cover all liquid fuels (gasoline blends, diesel fuel blends, jet fuel blends, bunker/marine fuel blends, and partially refined fuels), all gas fuels (LNG, methane, propane, butane, and ethane), solvents (kerosene and turpentine), crude oil and petroleum based chemicals and plastics. 
Response  58:  We used the 2005 NEI/NATA database to identify emissions information for marine loading of all products in the inventory and conducted a risk analysis.  Using that database, we found that the current MACT-based standards provide an ample margin of safety to protect public health and the environment, and therefore did not make any changes to the existing standards due to the residual risk.  However, we proposed and are making final a rule requirement for all products that are subject to the current rule to be loaded by submerged fill.  We proposed vapor controls for MTVLO for gasoline based on the fact that it is the highest-quantity commodity loaded in MTVLO in the U.S.  In addition, gasoline has a relatively high vapor pressure compared to other fuels and chemicals, higher emissions, and therefore would provide the best case for cost effectiveness when applying add-on control devices.  However, as discussed in the soon to be published FR preamble, we are not requiring additional vapor recovery. 

      Comment  59:  Two commenters (264, 301) provided input on the reference to Coast Guard regulations 46 CFR 153.282.  
      One commenter (0301) noted that splash loading is used for certain cargoes, particularly cargoes that require heating to ensure liquid flow.  The commenter believed the language in 40 CFR 63.560, that exempts commodities with vapor pressures less than 10.3 kilopascals (kPa) (1.5 psia) from the submerged fill requirement, is appropriate.
      One commenter (301) noted that the EPA and Coast Guard regulations governing tank vessel loading should be consistent with one another and supported cross-referencing to Coast Guard regulations.  The commenter encouraged EPA to make this point explicit in the final rule or the preamble to the final rule. This consistency should continue for any future possible changes to EPA or Coast Guard regulations.
      One commenter (264) suggested that instead of referencing 46 CFR §153.282 in §§63.560(d)(6)(i) and 63.560(f)(1), EPA should directly cite the actual specification for the submerged fill line.  The commenter noted that this is necessary because 46 CFR §153.282 is strictly a regulation governing tank ships and is not applicable to tank barges and because referencing the Coast Guard rules could cause issues if that particular section (§153.282) is later revised.
      Response  59:  We appreciate the input on use of cross-referencing to the Coast Guard regulations. The Coast Guard regulations ensure safety for MTVLO, and as one commenter notes, submerged fill is a safety consideration from the Coast Guard perspective, to reduce static charge, and submerged fill also provides emission reductions.  We believe that cross-referencing to submerged fill requirements is the appropriate approach, since we want to be consistent with Coast Guard requirements and their safety considerations.
      The applicability provisions are provided in subpart Y rule text (e.g., for tank barges and tank ships, for commodities with vapor pressure greater than 1.5 psia) and the requirements for submerged fill are contained within 46 CFR §153.282.  In discussions with a Coast Guard representative, we understand that similar submerged fill requirements are included under 46 CFR parts 151 (Barges Carrying Bulk Liquid Hazardous Material Cargoes) and 153 (Ships Carrying Bulk Liquid, Liquefied Gas, or Compressed Gas Hazardous Materials) for hazardous cargo rules and under 46 CFR part 32 (Pumps, Piping, and Hoses for Cargo Handling, under subchapter D for Tank Vessels) for less combustible cargo rules.  While parts 151 and 153 refer to barges and ships, respectively, the Coast Guard representative indicated that the requirements for submerged fill are essentially the same, and that the cited section (§153.282) is more explicit.  As a practical matter the Coast Guard representative believes that all barges and ships are conducting submerged fill that meet the §153.282 requirements, and no additional retrofit for fill pipe in these vessels would be needed.  We have therefore retained the cite to 46 CFR 153.282 to be required for loading into barges and ships.

      Comment  60:  Commenter 280 stated that EPA concluded that there were no technology changes that would affect the 1995 MTVLO MACT determinations for MTVLO or offshore loading. Commenter 291 stated that EPA found that there were no developments in practices, processes, and control technologies.
      Response  60:  In the 2008 proposal, under section 112(d)(6), we had not identified any developments in practices, processes and control technologies (the so-called technology review) for MTVLO.  For the 2010 proposal, we conducted a further technology review and identified the possibility of cost effective vapor recovery for gasoline loading at lower thresholds than the current standard.  In addition, as discussed in the soon to be published Federal Register preamble to the final rule, we are not requiring these additional vapor recovery controls proposed in 2010.  However, as presented in the soon to be published Federal Register preamble, the final rule does require submerged fill for the two subcategories of offshore loading and existing sources with MTVLO emissions less than 10/25 tpy HAP, as this is the MACT floor.  

      Comment  61:  One commenter (280) incorporated by reference their 1995 MTVLO NESHAP comments as they regard to the applicability of sections 183 and 112 to the refinery.  The commenter cited from their previous public comments on the 1995 MTVLO NESHAP the following reasons EPA should refrain from requiring the controls:  EPA lacks authority under 183(f) to apply controls in attainment areas; EPA must make a source-specific assessment; a source-specific assessment of their facility  would show that features of the facility combine to reduce exposure to emissions and ozone formation (i.e., not reasonably anticipated to endanger public health and welfare); compliance costs would exceed EPA's cost-effectiveness cutoff due to the additional costs and lower emissions for ships and barges; EPA lacks authority to regulate under section 112 since authority is under section 183(f) only; EPA should create a separate source category for offshore facilities because of unique circumstances.
      Response  61:  We responded to similar public comments on MACT and RACT in the 1995 MTVLO NESHAP in the Technical Support Document for Final Standards:  Summary of Public Comments and Responses.  In summary, we concluded that the Agency has authority to regulate MTVLO under section 183(f) and section 112; both VOC and HAP emissions from MTVLO may be reasonably anticipated to endanger human health and welfare; the Agency does not need to show that emissions from each facility causes or contributes to pollution; and the title of an Act (Ozone Nonattainment Areas) does not demonstrate that section 183(f) applies only in nonattainment areas (section 2.1.3).  We established a separate subcategory for offshore loading (section 2.3.2).  We believe the 1995 complete responses are appropriate.
      
      Comment  62:  Commenter 359 noted that application of beyond-the-floor controls to offshore facilities is not justified and EPA did not consider offshore cost impacts.  Commenter 359 stated that offshore facilities are large facilities, typically distribute product by ships rather than barges, and have loading berths on very long piers farther from shore.  Commenter 359 noted issues with the cost evaluation, including that it used a fixed assumption for installation costs and is based on a technology not in use at offshore facilities.  Commenter 359 noted that capital costs are higher for offshore loading due to need for additional platform structure and for piping to carry vapors. 
        One commenter 280 stated that capital costs for island facilities are 1.3 to 2.3 times higher on average.  One commenter 280 stated that island facilities transport almost all products by marine vessels, have large MTVLO, must build and operate their own utilities, do not have access to natural gas, and secondary emissions are higher.
      Commenter 280 noted that EPA created the offshore loading terminal subcategory in 1995 because of these differences and the extreme costs.  The commenter reiterated their comments submitted for the 1995 MACT and noted the comments are accurate today.[2]  Commenters 280, 359 stated that nothing has changed since 1995 to justify the changed conclusion on costs and cost effectiveness.
      Response  62:  As discussed in the soon to be published Federal Register notice preamble, the final rule does not include the proposed vapor recovery requirements.  We have reviewed the comments on the beyond-the-floor costs for offshore loading, and we agree that capital costs for offshore loading are likely higher than those for onshore MTVLO terminals, but we do not have any other information than that supplied by the commenters.  We agree that offshore loading should be considered separately based on higher costs and have costed them separately.  We included emission levels at offshore loading facilities based on the emission factor for ship loading, because MTVLO at offshore facilities are more likely to load ships than barges, and the emission rate would be lower.  See additional cost analysis discussion for offshore loading in the cost memorandum.  
      In regards to island facilities, we did not analyze this. The commenter expressed concerns that island facilities have higher costs.  Island or remote facilities likely have similar, but possibly somewhat less, logistic and cost issues as offshore facilities that we considered above.
      
      Comment  63:  Commenter 359 noted that EPA did not demonstrate any benefit from applying control requirements to offshore facilities, and EPA did not assess the nature and fate of the emission to determine if emissions would reasonably be anticipated to endanger public health or contribute to ozone non-attainment status.
      Response  63:  The final submerged fill requirements are for previously unregulated emission sources and thus we are required to set technology standards under CAA section 112(d) (3) standards.  The CAA section 112(d)(3) requires that the MACT floor for existing sources not be less stringent than "the average emission limitation achieved by the best performing 12 percent of the existing sources."  Submerged fill is the MACT floor for two subcategories, as discussed in the October 21, 2010 proposal preamble.
      
      Comment  64:  One commenter 280 provided updated cost information for a vapor combustion unit [for 54 million bbl/yr island loading facility] of $67 million in TCI and $15 million in TAC for a MTVLO with an annual gasoline throughput of 54 million bbl/yr.  The commenter estimated the cost effectiveness of HAP reductions as $106,000/ton HAP and VOC reductions as $7,700 per ton VOC, and noted the cost effectiveness could be higher.  Commenter 280 noted the beyond-the-floor option for offshore loading is not cost effective.  The commenter 280 does not agree that MACT beyond-the-floor controls should be judged by a criteria pollutant, and noted that secondary emission impacts from operation of the control device and associated utilities should also be considered.
      The commenter states that if additional controls were cost justified for offshore loading terminals based on emissions reductions, the $67 million capital cost is economically infeasible and thus qualifies for treatment under section 112(h). This approach is similar to that included in the rule preamble for Gas 1 Boiler MACT units.
      Commenter 280 noted that a vapor recovery system would be 2.5 to 3 times more costly to build [than a combustion system].  A vapor recovery system would have a VOC cost effectiveness over $13,000/ton VOC and a HAP cost effectiveness of over $177,000/ton HAP.  
      Response  64:  We appreciate the input on capital and annualized costs for combustion units for offshore loading.  We do not require that a particular control technology be installed, so a facility would install any control technology that meets a MACT requirement.  We costed a lean oil absorption system as an appropriate, applicable control technology (see Technical Issues below) and that is lower cost.  As discussed in the soon to be published Federal Register preamble, we are not requiring this proposed additional vapor recovery option.

Comment  65:  Commenter 274 noted that many Ozone Transport Commission (OTC) states have lowered the threshold for control of emissions from gasoline loading and have adopted VOC reduction requirements, and the commenter provided a summary of regulations in an attachment to their comment letter.  The thresholds in OTC rules varied from 15,000 gal/day to 6,000,000 gal/day loading (one was 60,000 gal/day in ozone season), control efficiencies varied from 90 to 98 percent VOC reductions, and some rule limited emissions to 2 lb VOC/1000 bbl loaded.  The commenter recommended EPA conduct a review of other state regulations in consideration of strategies to reduce ozone levels.  The commenter further suggested a lowering of the gasoline loading threshold below the proposed 1 million bbl/yr threshold.
      One commenter 297 stated that the proposed rule of 97 percent control is not the best MACT standard that is currently feasible, cost-effective, and available.  The commenter stated that EPA failed to require technologies that eliminate 100 percent of emissions, such as several off-the-shelf Zero Emission Technologies, and failed to require technologies that are 99+ percent effective in capturing emissions and fugitives for ship and land storage tanks, conveyance systems, and connection nozzles.  The commenter requested that EPA research, identify, ascertain and establish a list of available Zero Emission Technologies and MACT Vapor Recovery Systems that could be used for compliance to this proposed ruling.
      The commenter noted that EPA failed to require a secondary enclosure or encasement device to prevent fugitive emissions from escaping from leaking valve connections.
      One commenter 297 recommended that EPA adopt work practices and procedures that would identify, track and require mandatory periodic replacement of frequent equipment, systems, and parts failures, because it is common petroleum industry practice to wait for something to break.  Equipment is subject to metal fatigue, embrittlement, and breakdown.
      Response  65:  We appreciate the information regarding additional rules that support our determination of controlled HAP emissions from gasoline loading at 1 million bbl/yr or less.  The commenter did not provide the data to support those vapor recovery rules.  We believed they are based on VOC reductions and cost effectiveness.  From our analysis, we do not believe additional vapor controls are cost effective for HAP reductions beyond the current standard.  
      While we have conducted searches to identify zero emission technologies for control of HAP emissions from MTVLO, the searches revealed mostly zero emission technologies related to combustion emission sources, and not for MTVLO.  Thus, we do not have, nor did the commenter provide, any specific technologies and environmental benefits and costs to assess these technologies for MTVLO.  While one commenter recommended work practices that require mandatory replacement of equipment, the commenter did not provide data on failure frequencies or rates, costs, or environmental impacts on which to evaluate the commenter's suggestion.  Additionally, current Coast Guard rules include safe handling practices and operations that may address these issues.  See 92 FR 25005; May 13, 1994, Background and History for Proposal of the Coast Guard and EPA Standards.
      
      Comment  66:  Two commenters (359, 264) indicated that §63.560(f) needs corrections related to submerged fill requirements.
      One commenter (359) noted that EPA indicated that the submerged fill requirement would be applied to existing facilities below the 10/25 tpy threshold by a new paragraph §63.560(f), however, in the actual rule language submerged fill is proposed only for offshore facilities.  Commenter (359) noted that proposed edits to §63.560(a)(1), (2), and (3) refer to a paragraph (f) "of this section" but paragraph (f) does not exist.
      Commenter (264) stated that a correction is needed in the proposed language of subpart Y to add a new paragraph §63.560(f) concerning the submerged loading requirement for 2 subcategories (i.e., those facilities emitting less than 10/25 TPY of HAP and those facilities located more than 0.5 miles from shore) and to reference the new provisions in §63.562(f)(1) or (f)(2) for marine terminals that load more than 1 million bbl/yr of gasoline.
	Response  66:  We appreciate the commenters pointing out the omission of §63.560(f) in the proposed rule text.  We mistakenly omitted this paragraph but did include the text in the docket (see Docket item EPA-HQ-OAR-2010-0600-0253).  In the final rule, we are adding the submerged fill requirement to a new subparagraph §63.560(a)(4) for sources with MTVLO emissions less than 10/25 tpy HAP.  (The same submerged fill requirements are also included in the final rule for offshore loading in §63.560(d)(6).)  No new paragraph §63.560(f) will be added to the rule.
	
	Comment  67:  One commenter (359) noted that the proposed edits to §63.562(b)(1)(i) should read "1 million bbl/yr of gasoline" rather than "1 M barrels of gasoline," so as to read correctly and to be consistent with proposed edits to §63.562(b)(1)(ii) and (iii).
One commenter (359) stated that while proposed edits to §63.562(b)(1) would apply the vapor collection and vapor-tightness requirements to facilities that load at least 1 M bbl/yr of gasoline, the control standard for these facilities is specified separately in a proposed new paragraph §63.562(f). The commenter explained that the manner in which this is done results in the following provisions not being applicable to the new subcategory:	(1)  §63.562(b)(5) Prevention of carbon adsorber emissions during regeneration.	(2)  §63.562(b)(6) Maintenance allowed for loading berths. 	(3)  §63.563(b) Compliance determination for affected sources (initial performance test).  	The commenter explained that it is not apparent whether EPA intended for these provisions to not apply to the new subcategory of facilities that load at least 1 M bbl/yr of gasoline.
      Response  67:  We appreciate the commenter's questions on certain rule provisions being applicable to existing sources with MTVLO emissions less than 10/25 tpy and loading 1 million bbl/yr of gasoline, however, the final rule does not include the vapor recovery requirements.  The proposed vapor recovery requirements paragraph in §63.562(f), and the proposed changes to §63.562(b)(1) are not in the final rule. 
      
      Comment  68:  One commenter (359) noted that the proposed edits to Table 1, at §63.10(c)(10) and (11), include a reference to §63.657(m)(1) for reporting malfunctions but no paragraph §63.657(m) exists.
      Response  68:  We appreciate the commenters pointing out the omission of §63.567(m) in the proposed rule text.  We mistakenly omitted this paragraph.  In the final rule, we are adding a new paragraph in §63.567(m) for reporting malfunctions.  This paragraph is similar to malfunction paragraphs in other rules included in the proposal (e.g., in subpart KK, §63.829(b)(6)(v); and in subpart CCC, §63.1164(c)).  
      
      Comment  69:  Commenter (264) recommended that EPA clarify that §63.562(e) does not apply to existing sources with emissions less than 10 and 25 tons (since the emission limits do not apply) and that only §§63.560(f)(1) [new section], 63.567(j)(4) and 63.565(l) apply as per §63.560(a)(3).  As proposed, §63.56[0](a)(2) states that existing sources with emissions less than 10 and 25 tons are not subject to the emission standards in §63.562(b) and (d), except as specified in paragraph (f) [submerged fill].  The commenter noted §63.56[0](a)(3) implies that only the recordkeeping requirements in §63.567(j)(4) and the emission estimation requirements in §63.565(l) apply to these sources, except as specified in paragraph (f) [submerged fill].  Commenter (0264) noted that the way §63.560(a)(2) is worded, one might conclude that §63.562(e) applies to existing sources with emissions less than 10 and 25 tons. However, §63.560(a)(3) could again lead one to conclude that only those citations mentioned are applicable to sources with emissions less than 10 and 25 tons.
      Because the emission standards do not apply to existing sources with emissions less than 10 and 25 tons, the commenter suggested that one could conclude the operation and maintenance requirements in §63.562(e) for air pollution control equipment and monitoring equipment do not apply either.  
      Commenter (264) noted that the proposed changes to the Marine MACT include adding paragraph (7) to §63.562(e) to address recordkeeping and reporting requirements during periods of malfunction.  The rule also is proposing an affirmative defense against civil penalties for exceedances of emission standards caused by malfunctions, as well as the criteria for establishing the affirmative defense.  Because §63.560(a)(2) specifically states that the emission standards in §63.562(b) and (d) do not apply to existing sources with emissions less than 10 and 25 tons, then the affirmative defense requirements proposed in §63.562(e)(7) for exceedances of the emission standards caused by malfunctions should not apply either.  The commenter recommended that the language in §63.560(a)(3) be revised:  
      (a)(3) Only the recordkeeping requirements in §63.567(j)(4) and the emission estimation requirements of §63.565(l) apply to existing sources with emissions less than 10 and 25 tons, except as specified in paragraph (f) of this section. No other requirements in this subpart apply to existing sources with emissions less than 10 and 25 tons.
      Response 69:  We have reviewed the comments and we believe that the current and final §63.562(e) "operation and maintenance requirements for air pollution control equipment and monitoring equipment for affected sources" are applicable to existing sources with MTVLO emissions less than 10/25 tpy and are now subject to submerged fill requirements.  The proposed and the current language of §63.562(e) provides "owner and operators of affected sources shall operate and maintain a source, including associated air pollution control equipment, in a manner consistent with safety and good air pollution control practices for minimizing emissions".  Additionally, submerged fill reduces the generation of vapor emissions, so it is considered air pollution control equipment as well as process equipment. Thus, it applies to both "control equipment" and the "source" and is not limited to, as the commenter says, to "emission limits" and "air pollution control equipment."  Additionally, we believe the added subparagraph (§63.562(e)(7)) for malfunctions is applicable for the same reasons stated above.
      We do not believe §63.560(a)(2) and (3) specify requirements for emission estimation and recordkeeping as the "only" requirements since the current rule text does not use "only."  However, for clarity, instead of revising the subparagraphs to add the submerged fill requirement as we did in the proposal rule text, we are adding the submerged fill requirement as a new subparagraph §63.560(a)(4) in the final rule.
      
      Comment  70:  One commenter (264) requested that in the emission estimation procedures under §63.565(l), EPA clarify that the exclusion for "Commodities Exempted by §63.560(d)" was intended to mean that annual HAP emissions do not need to be estimated from marine tank vessel loading operations that meet any of the exemptions provided in §63.560(d).  The commenter believes that, in addition to excluding "commodities," other exemptions in §63.560(d) that should be excluded from the annual emission estimation requirement in §63.565(l) include §63.560(d)(2) thru (7).
   The commenter stated EPA could improve the language in §63.565(l) by rewording it as follows:
      (l) Emission estimation procedures. For sources with emissions less than 10 or 25 tons and sources with emissions of 10 or 25 tons, the owner or operator shall calculate an annual estimate of HAP emissions, excluding commodities emissions from marine tank vessel loading operations exempted by §63.560(d), from marine tank vessel loading operations.
      Response  70:  We appreciate the commenter requesting clarification to §63.565(l); however, the commenter's proposed changes to the rule text would not change this paragraph, and the impacts are unknown.  Therefore, we are not adopting the suggested language in this final action. 
      
Other Comments

      Comment  71:  Commenter 297 stated that EPA failed to provide adequate public notice and meetings to the public and Environmental Justice Communities that would be most impacted by this rulemaking.  The commenter noted that EPA provided the minimum public notice required by law, but made no effort to identify EJ advocacy organizations that traditionally participate in EPA rulemakings or those located in US port areas where Marine Tank Vessel Loading Operations are located, noting that most of these groups do not subscribe to the Federal Register.  The commenter stated that EPA could have worked with their Regional Offices to identify EJ advocacy organizations. The commenter requested that EPA conduct public hearings in Environmental Justice Communities that would be most impacted by this new rulemaking, including areas where Ports and Marine Fuel Terminals are located.
      Response  71:  EPA followed the Clean Air Act provisions of section 307(d) in providing public notice of its actions and generally does not provide special notice to any individual parties. These EJ organizations may also use the free website http://www.federalregister.gov/environment to help identify Federal Register notices related to environmental topics that may be of interest.  
      
      Comment  72:  Commenter 265 supports EPA's efforts to conduct the residual risk and technology reviews concurrently, which provides for efficiency, avoids potential inconsistency between standards, and allows for the results of the residual risk assessment to inform the technology review determination.
      Response  72  We agree with the commenter about the benefits of conducting these reviews concurrently, where the statutory residual risk and technology review deadlines make this practicable.   
      
      

