Promulgation of Air Quality Implementation Plans; Arizona; Regional Haze and Interstate Visibility Transport Federal Implementation Plan

Federal RegisterSep 3, 2014

Ask Donna

What actually matters in this document.

Text

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 52

[EPA-R09-OAR-2013-0588; FRL-9912-97-OAR]

Promulgation of Air Quality Implementation Plans; Arizona; Regional Haze and Interstate Visibility Transport Federal Implementation Plan

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Final rule.

SUMMARY:

This final action promulgates a Federal Implementation Plan (FIP) addressing the requirements of the Regional Haze Rule (RHR) and interstate visibility transport for the disapproved portions of Arizona's Regional Haze (RH) State Implementation Plan (SIP) as described in a final rule published in the

Federal Register

on July 30, 2013. In that action, we partially approved and partially disapproved the State's plan to implement the regional haze program for the first planning period. This final action includes our responses to comments on our proposed FIP published in the

Federal Register

on February 18, 2014. This final rule, together with a final rule published in the

Federal Register

on December 5, 2012, completes our FIP for the disapproved portions of Arizona's RH SIP. This final rule addresses the RHR's requirements for Best Available Retrofit Technology (BART), Reasonable Progress (RP), and a Long-term Strategy (LTS) as well as the interstate visibility transport requirements of the Clean Air Act (CAA) for pollutants that affect visibility in Arizona's 12 Class I areas and areas in nearby states. The BART sources addressed in this final FIP are Tucson Electric Power (TEP) Sundt Generating Station Unit 4, Lhoist North America (LNA) Nelson Lime Plant Kilns 1 and 2, ASARCO Incorporated Hayden Smelter, and Freeport-McMoRan Incorporated (FMMI) Miami Smelter. The reasonable progress sources addressed in the FIP are Phoenix Cement Company (PCC) Clarkdale Plant Kiln 4 and CalPortland Cement (CPC) Rillito Plant Kiln 4. EPA is prepared to work with the State on a SIP revision that would replace some or all elements of the FIP.

DATES:

Effective Date:

This rule is effective October 3, 2014.

ADDRESSES:

EPA has established docket number EPA-R09-OAR-2013-0588 for this action. Generally, documents in the docket are available electronically at

http://www.regulations.gov

or in hard copy at EPA Region 9, 75 Hawthorne Street, San Francisco, California. Please note that while many of the documents in the docket are listed at

http://www.regulations.gov,

some information may not be specifically listed in the index to the docket and may be publicly available only at the hard copy location (e.g., copyrighted material, large maps, multi-volume reports, or otherwise voluminous materials), and some may not be available at either locations (e.g., confidential business information). To inspect the hard copy materials, please schedule an appointment during normal business hours with the contact listed directly below.

FOR FURTHER INFORMATION CONTACT:

Thomas Webb, U.S. EPA, Region 9, Planning Office, Air Division, Air-2, 75 Hawthorne Street, San Francisco, CA 94105. Thomas Webb may be reached at telephone number (415) 947-4139 and via electronic mail at

r9azreghaze@epa.gov.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Introduction

II. History of State and Federal Plans

A. State Submittals and EPA Actions

B. EPA's Authority To Promulgate a FIP

III. Summary of Proposed Rule

A. Regional Haze

B. Interstate Transport of Pollutants That Affect Visibility

IV. Overview of Final Action

A. BART Determinations

B. Reasonable Progress Determinations

C. Reasonable Progress Goals and Demonstration

D. Long-Term Strategy

E. Interstate Visibility Transport

F. Other Changes From Proposal

V. Responses to General Comments

A. Introduction

B. Comments on State and EPA Actions on Regional Haze

C. Comments on State and Federal Roles in the Regional Haze Program

VI. Responses to Comments on EPA's Proposed BART Determinations

A. Comments on Sundt Generating Station Unit 4

B. Comments on Nelson Lime Plant Kilns 1 and 2

C. Comments on the Hayden Smelter

D. Comments on the Miami Smelter

VII. Responses to Comments on EPA's Proposed Reasonable Progress Determinations

A. Comments on Phoenix Cement Clarkdale Plant

B. Comments on CalPortland Cement Rillito Plant

C. Comments on Other Reasonable Progress NO

Point Sources

D. Comments on Area Sources of NO

and SO

E. Comments on Reasonable Progress Goals and Uniform Rate of Progress

F. Other Comments on Reasonable Progress

VIII. Responses to Comments on Statutory and Executive Order Reviews

IX. Responses to Other Comments

A. Comments on Preamble Language

B. Comments on Rule Language

C. Comments on Other Benefits of the Regional Haze Program

D. Miscellaneous Comments

X. Summary of Final Action

A. Regional Haze

B. Interstate Transport

XI. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review and Executive Order 13563: Improving Regulation and Regulatory Review

B. Paperwork Reduction Act

C. Regulatory Flexibility Act

D. Unfunded Mandates Reform Act

E. Executive Order 13132: Federalism

F. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments

G. Executive Order 13045: Protection of Children From Environmental Health Risks and Safety Risks

H. Executive Order 13211: Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution, or Use

I. National Technology Transfer and Advancement Act

J. Executive Order 12898: Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations

K. Congressional Review Act

L. Petitions for Judicial Review

Definitions

(1) The words or initials

Act

CAA

mean or refer to the Clean Air Act, unless the context indicates otherwise.

(2) The initials

ADEQ

mean or refer to the Arizona Department of Environmental Quality.

(3) The words

Arizona

and

State

mean the State of Arizona.

(4) The initials

BACT

mean or refer to Best Available Control Technology.

(5) The initials

BART

mean or refer to Best Available Retrofit Technology.

(6) The initials

BOD

mean or refer to boiler operating day.

(7) The initials

CAMD

mean or refer to Clean Air Markets Division at EPA.

(8) The initials

CBI

mean or refer to confidential business information.

(9) The term

Class I area

refers to a mandatory Class I Federal area.

(10) The initials

CEMS

refers to continuous emission monitoring system or systems.

(11) The initials

CRP

mean or refer to converter retrofit project.

(12) The initials

mean or refer to deciview, a measure of visual range.

(13) The initials

DOE

mean or refer to United States Department of Energy.

(14) The initials

ESECA

mean or refer to Energy Supply and Environmental Coordination Act of 1974.

(15) The words

EPA, we,

our

mean or refer to the United States Environmental Protection Agency.

(16) The initials

FGD

mean or refer to flue gas desulfurization.

(17) The initials

FIP

mean or refer to Federal Implementation Plan.

(18) The initials

FLM

mean or refer to Federal Land Managers.

(19) The initials

FUA

mean or refer to Fuel Use Act of 1978.

(20) The initials

IMPROVE

mean or refer to Interagency Monitoring of Protected Visual Environments monitoring network.

(21) The initials

IPM

mean or refer to Integrated Planning Model.

(22) The term

lb/MMBtu

means or refers to pounds per one million British thermal units.

(23) The initials

LDSCR

and

HDSCR

mean or refer to low and high dust Selective Catalytic Reduction, respectively.

(24) The initials

LNB

mean or refer to low NO

burners.

(25) The initials

LTS

mean or refer to Long-term Strategy.

(26) The initials

MACT

mean or refer to Maximum Achievable Control Technology.

(27) The initials

mean or refer to megawatts.

(28) The initials

NAAQS

mean or refer to National Ambient Air Quality Standard or Standards.

(29) The initials

NEI

mean or refer to National Emissions Inventory.

(30) The initials

NESCAUM

mean or refer to Northeast States for Coordinated Air Use Management.

(31) The initials

NESHAP

mean or refer to National Emission Standards for Hazardous Air Pollutants.

(32) The initials

mean or refer to nitrogen oxides.

(33) The initials

mean or refer to National Park.

(34) The initials

NPS

mean or refer to the National Park Service.

(35) The initials

NSCR

mean or refer to Non-Selective Catalytic Reduction.

(36) The initials

NSPS

mean or refer to new source performance standards.

(37) The initials

OFA

mean or refer to Over Fire Air.

(38) The initials

mean or refer to particulate matter.

(39) The initials

2.5

mean or refer to fine particulate matter with an aerodynamic diameter of less than 2.5 micrometers.

(40) The initials

mean or refer to particulate matter with an aerodynamic diameter of less than 10 micrometers.

(41) The initials

PSD

mean or refer to Prevention of Significant Deterioration.

(42) The initials

PTE

mean or refer to potential to emit.

(43) The initials

mean or refer to regional haze.

(44) The initials

RHR

mean or refer to the Regional Haze Rule, originally promulgated in 1999 and codified at 40 CFR 51.308-309.

(45) The initials

RMC

mean or refer to Regional Modeling Center.

(46) The initials

mean or refer to Reasonable Progress.

(47) The initials

RPG

RPGs

mean or refer to Reasonable Progress Goal(s).

(48) The initials

SCR

mean or refer to Selective Catalytic Reduction.

(49) The initials

SIP

mean or refer to State Implementation Plan.

(50) The initials

SNCR

mean or refer to Selective Non-catalytic Reduction.

(51) The initials

mean or refer to sulfur dioxide.

(52) The initials

SOFA

mean or refer to Separated Over Fire Air.

(53) The initials

SRP

mean or refer to Salt River Project Agricultural Improvement and Power District.

(54) The initials

tpy

mean tons per year.

(55) The initials

TSD

mean or refer to Technical Support Document.

(56) The initials

TSF

mean or refer to tons of stone feed.

(57) The initials

ULNB

mean or refer to ultra-low NO

burners.

(58) The initials

URP

mean or refer to Uniform Rate of Progress.

(59) The initials

VOC

mean or refer to volatile organic compounds.

(60) The initials

VRP

mean or refer to Visibility Restoration Plan.

(61) The initials

WRAP

mean or refer to the Western Regional Air Partnership.

I. Introduction

The purpose of the Federal and state regional haze plans is to achieve a national goal, declared by Congress, of restoring and protecting visibility at 156 Federal class I areas across the United States, most of which are national parks and wilderness areas with scenic vistas enjoyed by the American public. The national goal as described in CAA Section 169A is “the prevention of any future, and the remedying of any existing, impairment of visibility in mandatory class I Federal areas which impairment results from man-made air pollution.” Arizona has 12 Class I areas, including some of the most magnificent natural areas in our country. Five other Class I areas are close by in neighboring states. Please refer to our previous rulemaking on the Arizona RH SIP for additional background information regarding the CAA, regional haze and EPA's RHR.

77 FR 75704, 75707-75702 (December 21, 2012).

EPA has previously acted to approve a number of elements of the Arizona RH SIP, and to disapprove others. In today's final action, EPA is reducing harmful emissions from six facilities that contribute to visibility impairment in 17 protected national parks and wilderness areas in Arizona and neighboring states. Four of the facilities are subject to Best Available Retrofit Technology (BART) controls for emissions of nitrogen oxides (NO

), sulfur dioxide (SO

), and particulate matter (PM). The other two facilities are subject to limits on their NO

emissions pursuant to the Reasonable Progress (RP) provisions of the Regional Haze Rule (RHR). The BART sources are Sundt Generating Station Unit 4, Nelson Lime Plant Kilns 1 and 2, Hayden Smelter, and Miami Smelter. The RP sources are the Phoenix Cement Clarkdale Plant Kiln 4 and CalPortland Cement Rillito Plant Kiln 4. EPA is promulgating this partial FIP because we found that Arizona had failed to submit a complete RH SIP, and later disapproved portions of Arizona's RH SIP for not meeting all the requirements of the CAA and EPA's RHR.

EPA has worked with the owners and operators of the facilities regulated by today's rule to ensure we have the most up-to-date information for making decisions on BART, RP, and the Long-Term Strategy (LTS), the three major requirements of the RHR. In today's notice, we respond to comments on our proposed rule, present our analysis, and indicate where we have made adjustments based on the comments and additional information. The required emission limits, compliance methods, and deadlines for compliance in our final rule are compatible with each facility's operations, and provide sufficient flexibility for achieving compliance in a reasonable period of time. In several instances we have adjusted the emission limits, averaging times and/or compliance deadlines in response to additional information supplied by the facilities' owners or operators. Further, in the case of TEP Sundt Unit 4, we have included an alternative to BART controls suggested by the facility's owner, which provides better emission reductions to improve visibility.

Given the combination of State and Federal plans to implement the regional haze program in Arizona, EPA and the Arizona Department of Environmental Quality (ADEQ) must continue to rely on their historically strong partnership under the CAA to protect the environment and human health. We would welcome a State plan to replace some or all of the Federal plan. Moreover, we commit our resources to ensuring a successful regional haze program for Arizona. EPA estimates today's action will result in annual emission reductions of about 2,900 tons/year of NO

and 29,300 tons/year of SO

. These reductions are expected to benefit at least 17 Class I areas in four states, including Arizona.

II. History of State and Federal Plans

A. State Submittals and EPA Actions

EPA made a finding on January 15, 2009, that 37 states, including Arizona, had failed to make all or part of the required SIP submissions to address regional haze.

Specifically, EPA found that Arizona failed to submit the plan elements required by 40 CFR 51.309(d)(4) and (g). In 2011 ADEQ submitted a SIP under section 308 of the

RHR, but did not withdraw its 309 SIP. EPA disapproved Arizona's 309 SIP (with the exception of several smoke management rules) on August 8, 2013.

Both of the Arizona RH SIPs are available to review in the docket for this final rule.

74 FR 2392.

78 FR 48326.

“Arizona State Implementation Plan, Regional Haze under Section 308 of the Federal Regional Haze Rule,” February 28, 2011.

As shown in Table 1, the first phase of EPA's action on the 2011 RH SIP addressed three BART sources. The final rule for the first phase (a partial approval and partial disapproval of the State's plan and a partial FIP) was published in the

Federal Register

on December 5, 2012. The emission limits on the three sources will improve visibility by reducing NO

emissions by about 22,700 tpy. In the second phase of our action, we proposed on December 21, 2012, to approve in part and disapprove in part the remainder of the 2011 RH SIP. Subsequently, ADEQ submitted a supplement to the Arizona RH SIP (“SIP Supplement”) on May 3, 2013, to correct certain deficiencies identified in that proposal. We then proposed on May 20, 2013, to approve in part and disapprove in part the SIP Supplement. Our final rule approving in part and disapproving in part the Arizona RH SIP was published on July 30, 2013. In the third phase of our action, we proposed a FIP on February 18, 2014, to address the remaining disapproved portions of the State's plan, which we are finalizing today.

Table 1—EPA's Actions on the Arizona RH SIP and FIP

EPA actions

Federal Register

Proposed rule

Final rule

Phase 1:

SIP, FIP

BART determinations for Apache, Cholla and Coronado

July 20, 2012 (77 FR 42834)

December 5, 2012 (77 FR 72512).

Phase 2:

SIP

Partial approval and partial disapproval of remaining elements of the SIP, including SIP Supplement

December 21, 2012 (77 FR 75704), May 20, 2013 (78 FR 29292)

July 30, 2013 (78 FR 46142).

Phase 3:

FIP

FIP for remaining disapproved elements of the SIP

February 18, 2014 (79 FR 9318)

Today's Final Action.

B. EPA's Authority To Promulgate a FIP

Under CAA section 110(c), EPA is required to promulgate a FIP at any time within 2 years of the effective date of a finding that a state has failed to make a required SIP submission or has made an incomplete submission, or of the date that EPA disapproves a SIP. The FIP requirement is terminated only if a state submits a SIP, and EPA approves that SIP as meeting applicable CAA requirements before promulgating a FIP. Specifically, CAA section 110(c) provides that EPA “shall promulgate” a FIP “at any time within 2 years” after finding that “a State has failed to make a required submission” or that the SIP or SIP revision submitted by the State does not satisfy the minimum criteria established under CAA section 110(k)(1)(A), or after disapproving a SIP in whole or in part “unless the State corrects the deficiency” EPA approves the plan or plan revision before promulgating a FIP.

Section 302(y) defines the term “Federal implementation plan” in pertinent part, as a plan (or portion thereof) promulgated EPA “to fill all or a portion of a gap or otherwise correct all or a portion of an inadequacy” in a SIP, and which includes enforceable emission limitations or other control measures, means or techniques (including economic incentives, such as marketable permits or auctions or emissions allowances).

In the case of the Arizona RH SIP, two different triggering events have occurred under section 110(c). EPA has made a finding that the State failed to make a required submission,

and we have partially disapproved the submissions that the State subsequently made. Therefore, EPA is required under CAA section 110(c) to promulgate a FIP for the portions of the Arizona RH SIP that we disapproved on July 30, 2013.

74 FR 2392-93 (January 15, 2009).

III. Summary of Proposed Rule

In this section, we provide a summary of the proposed rule that was published in the

Federal Register

on February 18, 2014,

as background for understanding today's final action.

79 FR 9318-9378.

A. Regional Haze

Our proposed rule included proposed BART determinations for four sources and proposed RP determinations for nine sources. These determinations resulted in proposed emission limits, compliance schedules, and other requirements for four BART sources and two of the RP sources. The proposed regulatory language was included under Part 52 at the end of that document. We also addressed the reasonable progress goals (RPGs), as well as the requirements of the LTS. Lastly, we proposed that the approved measures in the Arizona RH SIP, and measures in our previously promulgated FIP and proposed FIP, would adequately address the interstate transport of pollutants that affect visibility.

1. Proposed BART Determinations

Sundt Generating Station Unit 4:

EPA proposed to find that Sundt Unit 4 is BART-eligible and subject to BART for NO

, SO

, and particulate matter of less than 10 micrometers (PM

). For NO

, we proposed an emission limit of 0.36 lb/MMBtu as BART, which is consistent with the use of Selective Non-Catalytic Reduction (SNCR) as a control technology. For SO

, we proposed an emission limit of 0.23 lb/MMBtu as BART on a 30-day boiler operating day (BOD) rolling basis, which is consistent with the use of dry sorbent injection (DSI) as a control technology. For PM

, we proposed a filterable PM

emission limit of 0.030 lb/MMBtu as BART based on the use of the unit's existing fabric filter baghouse. We also proposed a switch to natural gas as a better-than-BART alternative to the proposed BART controls for all three pollutants.

Nelson Lime Plant Kilns 1 and 2:

EPA proposed to find that Nelson Lime Kilns 1 and 2 are subject to BART for NO

, SO

, and PM

. For NO

, we proposed a BART emission limit at Kiln 1 of 3.80

lb/ton of lime and at Kiln 2 of 2.61 lb/ton of lime on a 30-day rolling basis as verified by continuous emission monitoring systems (CEMS). These emission limits are consistent with the use of low-NO

burners (LNB) and SNCR as control technologies. We proposed that BART for SO

is an emission limit of 9.32 lb/ton of lime for Kiln 1 and 9.73 lb/ton of lime for Kiln 2 on a 30-day rolling basis, which is consistent with the use of a lower sulfur fuel blend. For PM

, we proposed a BART emission limit of 0.12 lb/tons of stone feed (TSF) at Kilns 1 and 2 based on the use of the unit's existing fabric filter baghouses.

Hayden Smelter:

EPA proposed that the Hayden Smelter is subject to BART for NO

, and we proposed BART emission limits for NO

and SO

. We previously approved the State's determination that the Hayden Smelter is subject to BART for SO

, but disapproved the State's SO

BART determination. For NO

, we proposed an annual emission limit of 40 tons per year (tpy) of NO

emissions from the BART-eligible units, which is consistent with current emissions from these units. For SO

from the converters, we proposed a BART control efficiency of 99.8 percent on a 30-day rolling basis on all SO

captured by primary and secondary control systems, which can be achieved with a new double contact acid plant. For SO

from the anode furnaces, we proposed a work practice standard requiring that the furnaces be charged only with blister copper or higher purity copper. We previously approved Arizona's determination that BART for PM

at the Hayden Smelter is no additional controls. In order to ensure the enforceability of this determination, we proposed to incorporate the emission limits and associated compliance requirements of the Maximum Achievable Control Technology (MACT),

Subpart QQQ, as part of the LTS.

National Emission Standard for Hazardous Air Pollutants for Primary Copper Smelting at 40 CFR Part 63.

Miami Smelter:

EPA proposed that the Miami Smelter is subject to BART for NO

, and we proposed BART emission limits for NO

and SO

. EPA previously approved the State's determination that the Miami Smelter is subject to BART for SO

, but disapproved the State's SO

BART determination. For NO

, we proposed an annual emission limit of 40 tpy NO

emissions from the BART-eligible units, which is consistent with current emissions. For SO

from the converters, we proposed a BART control efficiency of 99.7 percent on a 30-day rolling basis on all SO

emissions captured by the primary and secondary control systems as verified by CEMS. This control efficiency could be met through improvements to the primary capture system, construction of a secondary capture system, and application of the MACT Subpart QQQ requirements to the capture systems. For SO

emissions from the electric furnace, we proposed as BART a work practice standard to prohibit active aeration. We previously approved Arizona's determination that BART for PM

at the Miami Smelter is the MACT for Primary Copper Smelting. We proposed to find that the federally enforceable provisions of the MACT, which apply to the Miami Smelter and are incorporated into its Title V Permit, are sufficient to ensure the enforceability of this determination.

2. Proposed RP Determinations

Point Sources of NO

EPA conducted source-specific RP analyses of potential NO

controls for non-BART units at nine different sources. Based on these analyses, we proposed to require controls on two cement kilns: PCC Clarkdale Kiln 4 and CPC Rillito Kiln 4. Specifically, EPA proposed an emission limit of 2.12 lb/ton on Kiln 4 of the Clarkdale Plant based on a 30-day rolling average, which is consistent with SNCR as a control technology. We proposed an emission limit of 2.67 lb/ton on Kiln 4 of the Rillito Plant based on a 30-day rolling average, which also is consistent with SNCR as a control technology. We also requested comment on the possibility of requiring a rolling 12-month limit on NO

emissions in lieu of a lb/ton emission limit at these facilities. For the remaining seven sources, as well as other units at CPC, we proposed to find that it was reasonable not to require NO

controls during this planning period. These sources are the CPC Rillito Plant (Kilns 1-3); Arizona Public Service (APS) Cholla (Unit 1); El Paso Natural Gas (EPNG) Tucson, Flagstaff, and Williams Compressor Stations; TEP Sundt (Units 1-3); Ina Road Sewage Plant; and TEP Springerville (Units 1 and 2).

Area Sources of NO

and SO

We proposed to find that it is reasonable not to require additional controls on area sources at this time. Primarily, these area source categories are distillate fuel oil combustion in industrial and commercial boilers and in internal combustion engines, and residential natural gas combustion. While the State's area sources currently contribute a relatively small percentage of the visibility impairment at impacted Class I areas, we recommended better emission inventories and an improved RP analysis in the next planning period for area sources.

Reasonable Progress Goals:

EPA proposed RPGs consistent with a combination of control measures that include those in the approved portion of the Arizona RH SIP and in EPA's finalized and proposed FIPs. While not quantifying a new set of RPGs based on these control measures, we proposed that it is reasonable to assume improved levels of visibility at Arizona's 12 Class I areas by 2018 because the measures in the FIPs produce emissions reductions that are significantly beyond those required by the State.

Demonstration of Reasonable Progress:

EPA proposed to find that it is reasonable not to provide for rates of progress at the 12 Class I areas consistent with the uniform rate of progress (URP) in this planning period.

We also proposed to find that the RP analyses underlying our actions on the Arizona RH SIP

and FIP are sufficient to demonstrate that it is reasonable not to provide for rates of progress in this planning period that would attain natural conditions by 2064.

Lastly, we approved the State's decision not to require additional controls (i.e., controls beyond what the State or we determine to be BART) on point sources of SO

40 CFR 51.308(d)(1)(ii).

See proposed actions at 77 FR 75727-75730, 78 FR 29297-292300 and final action at 78 FR 46172.

40 CFR 51.308(d)(1)(ii).

78 FR 46172.

3. Long-Term Strategy

EPA proposed to find that provisions in the Arizona RH SIP and FIP fulfill the requirements of 40 CFR 51.308(d)(3)(ii), (v)(C) and (v)(F). These requirements are to include in the LTS measures needed to achieve emission reductions for out-of-state Class I areas, emission limitations and schedules for compliance to achieve the RPGs, and enforceability provisions for emission limitations and control measures.

We proposed to promulgate emission limits, compliance schedules, and other requirements for four BART sources and two RP sources to complete this part of the FIP for these requirements.

See 78 FR 46173 (codified at 40 CFR 52.145(e)(ii)).

B. Interstate Transport of Pollutants That Affect Visibility

We have proposed that a combination of SIP and FIP measures will satisfy the FIP obligation for the visibility requirement of CAA section 110(a)(2)(D)(i)(II) for the 1997 8-hour ozone, 1997 PM

2.5

, and 2006 PM

2.5

NAAQS. CAA section 110(a)(2)(D)(i)(II) requires that all SIPs contain adequate

provisions to prohibit emissions that will interfere with other states' required measures to protect visibility. We refer to this as the interstate transport visibility requirement.

IV. Overview of Final Action

We are promulgating a FIP to address the remaining disapproved portions of the Arizona RH SIP.

We include in Section V below a summary of our responses to comments on our proposed FIP,

and describe where comments resulted in revisions to the proposal. In this section, we provide a summary of the final BART determinations, RP determinations, RPGs and demonstration, LTS provisions, and interstate transport provisions of the FIP. This final FIP also includes emission limits, compliance schedules and requirements for equipment maintenance, monitoring, testing, recordkeeping, and reporting for all affected sources and units. The final regulatory language for the FIP is under Part 52 at the end of this notice.

78 FR 46142 (July 30, 2013).

79 FR 9318 (February 18, 2014).

A. BART Determinations

EPA conducted BART analyses and determinations for four sources: Sundt Generating Station Unit 4, Nelson Lime Plant Kilns 1 and 2, the Hayden Smelter, and the Miami Smelter. The final BART determinations are listed in Table 2, comparing the final limits to the proposed limits with short descriptions of changes in the footnotes. The exact compliance deadlines will be calculated based upon the date that this document is published in the

Federal Register

, which we anticipate will occur sometime in July 2014.

Sundt Generating Station:

In this final rule, we have retained the BART determination and the final BART emission limits as proposed, as well as the option of a better-than-BART alternative that was submitted by TEP. Although the final BART determination and limits are the same, we have made some changes to the better-than-BART alternative based on comments and additional information.

Regarding BART, we are finalizing our determination that Sundt Unit 4 is BART-eligible and subject to BART for SO

, NO

, and PM

. The final BART emission limits are the same as proposed. The NO

emission limit is 0.36 lb/MMBtu, which is equivalent to using SNCR with the existing LNB as control technologies. The SO

emission limit is 0.23 lb/MMBtu on a 30-day BOD rolling basis, which is consistent with using DSI as a control technology. The PM

emission limit is 0.030 lb/MMBtu based on the use of the existing fabric filter baghouse. Compliance is required within three years of the publication of this notice in the

Federal Register

, also as proposed.

Regarding the better-than-BART alternative to switch to natural gas, we are finalizing the proposed emission limit for NO

of 0.25 lb/MMBtu, but revising the SO

and PM

emission limits. The final SO

limit is increased from 0.00064 to 0.054 lb/MMBtu to allow for continued co-firing with landfill gas that has a higher sulfur content than pipeline natural gas. The final PM

limit relies on a performance test due to the uncertainties related to switching from coal to gas, which now includes measuring condensable, in addition to filterable, PM

. Further, we have extended the final compliance deadline by six months to December 31, 2017, consistent with the date that TEP initially included in its better-than-BART proposal. TEP is required to notify EPA regarding its selection of BART or the alternative by March 2017.

Nelson Lime Plant:

EPA is finalizing its determination that Nelson Lime Plant Kilns 1 and 2 are subject to BART for NO

, SO

, and PM

. We have revised the final emission limits for NO

and SO

to account for startup and shutdown emissions, which were not considered in LNA's original BART analysis that was submitted to EPA for consideration. This change to the emission limits for NO

and SO

does not change the corresponding control technologies, which are still SNCR and lower sulfur fuel, respectively. The final BART emission limit for PM

is 0.12 lb/ton for each kiln as proposed, equivalent to using the existing baghouse.

We are making the following revisions to the NO

limits in response to comments received on our proposal. First, we are revising the averaging time for the lb/ton limits to a 12-month rolling average instead of a 30-day rolling average. The longer 12-month averaging time should even out the emission spikes from startup and shutdown events that would more significantly influence a 30-day average. Second, we are requiring an optimization plan to assess the final BART emission limit for NO

based on a 12-month rolling average, which is 3.80 lb/ton for Kiln 1 and 2.61 lb/ton for Kiln 2. Third, we are adding a combined limit for Kilns 1 and 2 of 3.27 tons/day on a 30-day rolling average to ensure short-term visibility protection. Both compliance methods (lb/ton at each kiln as optimized and tons/day for both kilns) are equivalent to using SNCR control technology. The compliance deadline for the final NO

emission limit is three years from the publication date, as proposed.

We are making the following revisions to the SO

limits in response to comments received on our proposal. First, as with the final limit for NO

, we are revising the averaging time for the lb/ton limits to a 12-month rolling average instead of a 30-day rolling average to account for emission spikes from startup and shutdown events that would more significantly influence a 30-day average. The final BART emission limits for SO

are 9.32 lb/ton for Kiln 1 and 9.73 lb/ton for Kiln 2, as proposed. Second, we are adding a combined limit for Kilns 1 and 2 of 10.1 tons/day to ensure short-term visibility protection. Both compliance methods (lb/ton at each kiln and tons/day at both kilns) are equivalent to using lower sulfur fuel, as proposed. Finally, we have extended the compliance deadline for meeting the final limit for SO

from six to 18 months to allow sufficient time for installation of monitoring equipment to demonstrate compliance with the new limits.

Hayden Smelter:

EPA is finalizing its determination that the Hayden Smelter is subject to BART for NO

. We previously approved the State's determination that the Hayden Smelter is subject to BART for SO

and PM

, and the State's determination that BART for PM

is equivalent to existing controls. The final BART emission limit for NO

is 40 tpy and applies to the converters and anode furnaces. The NO

limit is consistent with current emissions and is the same as proposed. The final BART emission limit for SO

from the anode furnaces is equivalent to existing controls, as proposed. For PM

, we are incorporating by reference provisions of the National Emission Standards for Hazardous Air Pollutants (NESHAP) for primary copper smelters

to ensure that Arizona's BART determination is made enforceable, as part of the LTS.

40 CFR part 63 subpart QQQ.

We are making a number of revisions to the proposed SO

emission limits from the converters in response to comments. For SO

emissions from the converters, the final BART emission limits are a 99.8 percent control efficiency on a 365-day rolling average for the primary capture system, and a 98.5 percent control efficiency on a 365-day rolling average for the secondary capture system. The BART limit for the primary capture system corresponds to the existing double contact acid plant, whereas the limit for the secondary capture system is equivalent to a new

amine scrubber as a control technology. We have revised our proposal by applying separate limits to the primary and secondary capture systems in recognition of significant differences in flow volume and SO

concentration between the two systems. We revised the averaging time from 30 to 365 days for the primary capture system in recognition that the control efficiency is based on annual acid production and annual SO

emissions. In addition, we are finalizing a work practice standard requiring that the primary and secondary capture systems be designed and operated to maximize capture of SO

emissions from the converters.

The final compliance deadline for the primary capture and control system to meet the SO

limit is three years from publication, as proposed. The final deadlines for the NO

and PM

limits are also three years from publication. However, we extended the final compliance deadline to meet the SO

limit for the secondary capture and control system from three to four years from publication to provide sufficient time to plan and build a new amine scrubber.

Miami Smelter:

EPA is finalizing its determination that the Miami Smelter is subject to BART for NO

. We previously approved the State's determination that the Miami Smelter is subject to BART for SO

and PM

, and the State's determination that BART for PM

is equivalent to the National Emission Standard for Hazardous Air Pollutants (NESHAP) for primary copper smelters. The final BART emission limit for NO

is 40 tpy that applies to the converters and electric furnace. The NO

limit represents current emissions and is the same as proposed. For SO

from the electric furnace, the final BART emission limit is the existing work practice standard to prohibit active aeration. For PM

, we are incorporating by reference provisions of the NESHAP for primary copper smelters

to ensure that Arizona's BART determination is made enforceable, as part of the LTS.

40 CFR part 63 subpart QQQ.

For SO

from the converters, the final BART emission limit is a control efficiency of 99.7 percent on a 365-day rolling average applied to the combined primary and secondary capture systems on a cumulative mass basis. While the control efficiency of 99.7 percent is the same as proposed, we revised the compliance method from a 30-day average to a 365-day rolling average. We revised the averaging time in response to FMMI's comment that the control efficiency is based on annual acid production and annual SO

emissions. The 99.7 percent control efficiency is equivalent to improvements to the primary control system (existing acid plant with a tailstack scrubber) and construction of new secondary capture and control systems. In addition, we are finalizing a work practice standard requiring that the primary and secondary capture systems be designed and operated to maximize capture of SO

emissions from the converters.

The final compliance deadlines for SO

from the electric furnace as well as the NO

and PM

limits, are two years from the date of the document's publication. However, we extended the final compliance deadline for SO

from the converters to January 1, 2018, to provide sufficient time to plan and build a new secondary capture and control system. We also added a compliance option for the secondary capture system to use either CEMS or to calculate emissions based on the amount of reagent added to the scrubber, because it may be impractical to operate CEMS on the inlet of a new scrubber.

Table 2—Final Emission Limits on BART Sources

Source

Units

Pollutants

Proposed limit

Final

limit

Measure

Corresponding control

technology

Sundt Generating Station

Unit 4

0.36

0.23

0.030

Same

lb/MMBtu

Selective Non-Catalytic Reduction.

Dry Sorbent Injection.

Fabric filter baghouse (existing).

Unit 4 Alternative

0.25

0.00064

0.010

Same

0.054.

Test.

lb/MMBtu

Switch to natural gas.

Nelson Lime Plant

Kiln 1

3.80

Same

3.27

lb/ton

tons/day

Selective Non-Catalytic Reduction.

9.32

Same

10.1

lb/ton.

tons/day.

Lower sulfur fuel.

0.12

Same

lb/ton

Fabric filter baghouse (existing).

Kiln 2

2.61

Same

3.27

lb/ton

tons/day.

Selective Non-Catalytic Reduction.

9.73

Same

10.1

lb/ton

tons/day.

Lower sulfur fuel.

0.12

Same

lb/ton

Fabric filter baghouse (existing).

Hayden Smelter

All BART Units

Same

tpy

None.

Converters 1, 3-5

99.8

Control efficiency

Primary capture: Double contact acid plant (existing).

98.5

Secondary capture: New amine scrubber.

Anode Furnaces 1, 2

None

Same

None

Work practice standard.

Miami Smelter

All BART Units

Same

tpy

None.

Converters 2-5

99.7

Same

Control efficiency

Improve primary and new secondary capture systems, additional controls as needed.

Electric Furnace

None

Same

None

Work practice standard.

Final limit revised to accommodate co-firing with landfill gas that has higher sulfur content.

Final limit is based on result of initial performance test.

Final limit includes a requirement for SNCR optimization plan.

Final limit is based on rolling 12-month average instead of rolling 30-day average.

Final limit is combined for Kilns 1 and 2 with compliance based on a rolling 30-day average.

Final limit is separate for primary and secondary capture systems.

B. Reasonable Progress Determinations

Point Sources of NO

EPA is finalizing its determination that PCC Clarkdale Plant Kiln 4 and CPC Rillito Plant Kiln 4 are subject to NO

emission controls under the RP requirements of the RHR as shown in Table 3. We also are finalizing our determination that it is reasonable not to require controls at this time on NO

emissions from the other seven sources that we evaluated for RP as well as other units at the Rillito Plant. These sources are the CPC Rillito Plant (Kilns 1-3); APS Cholla (Unit 1); El Paso Natural Gas (EPNG) Tucson, Flagstaff, and Williams Compressor Stations; TEP Sundt (Units 1-3); Ina Road Sewage Plant; and TEP Springerville (Units 1 and 2).

Clarkdale Plant Kiln 4:

PCC has two options for meeting the RP requirements. It can choose to meet either a lb/ton limit or tons/year limit for NO

. The final NO

limit for the first option is the proposed 2.12 lb/ton with a requirement for an SNCR optimization plan. The final lb/ton NO

limit is based on a 30-day rolling average consistent with SNCR as a control technology. The second option is an 810 tons/year NO

limit that is achievable by installing SNCR or maintaining clinker production at current levels. The 810 tons/year limit is based on a 12-month rolling average equivalent to a 50 percent reduction in baseline emissions. PCC must notify EPA of the option it has selected no later than July 2018 with a compliance deadline of December 31, 2018.

Rillito Plant Kiln 4:

The final RP emission limit for NO

is 3.46 lb/ton based on a 35 percent control efficiency. We have increased the final limit from the proposed 2.67 lb/ton that was based on a 50 percent control efficiency in response to additional information from CPC regarding constraints on efficiency due to the kiln design. In addition, we are requiring implementation of an SNCR optimization plan to determine if a higher control efficiency is achievable. The final NO

limit is based on a 30-day rolling average and is consistent with SNCR as a control technology. The compliance deadline is December 31, 2018, the same as proposed.

Table 3—Final Emission Limits on RP Sources

Source

Units

Pollutants

Proposed limit

Final limit

Measure

Corresponding control technology

Clarkdale Plant

Kiln 4

2.12

Same

lb/ton

Selective Non-Catalytic Reduction.

810

Same

tons/year

Current Production Levels.

Rillito Plant

Kiln 4

2.67

3.46

lb/ton

Selective Non-Catalytic Reduction.

Final limit includes a requirement for SNCR optimization plan.

Final limit for second option is in tons/year in lieu of lb/ton.

Final limit includes a requirement for SNCR optimization plan.

Area Sources of NO

and SO

EPA is finalizing its determination that it is reasonable not to require additional controls on Arizona's area sources at this time. Area source categories such as distillate fuel oil combustion in boilers and internal combustion engines as well as residential natural gas combustion currently contribute a relatively small percentage of the visibility impairment at Class I areas, but should be considered for controls in future planning periods.

C. Reasonable Progress Goals and Demonstration

Reasonable Progress Goals:

EPA is quantifying our proposed RPGs (in deciviews) for the 20 percent worst days and 20 percent best days in 2018. The RPGs for Arizona's 12 Class I areas account for the emission reductions from BART and RP control measures in the final RH FIP. The RPGs reflect the results of our BART analyses and our RP analysis of point sources of NO

and area sources of NO

and SO

as described in our proposal and in response to comments in today's final rule. The RPGs also include the effects of the three BART determinations finalized in our Phase 1 FIP and the effects of other existing State and Federal controls. Today's final RPGs provide for an improvement in visibility on the worst days and no degradation in visibility on the best days during this planning period.

Demonstration of Reasonable Progress:

EPA's final determination is that it is not reasonable to provide for rates of progress at Arizona's 12 Class I areas that would attain natural visibility conditions by 2064 (i.e., the URP).

Our demonstration that a slower rate of progress is reasonable is based on the RP analyses performed by us and the State that considered the four statutory RP factors. Although progress is slower than the URP, the FIP provides for RPGs that reflect an improved rate of progress and a significantly shorter time period to reach natural visibility conditions at each of Arizona's Class I areas, compared with the RPGs in the Arizona RH SIP.

40 CFR 51.308(d)(1)(ii).

D. Long-Term Strategy

EPA is finalizing its determination that provisions in this final rule in combination with provisions in the approved Arizona RH SIP and the Phase 1 Arizona RH FIP

fulfill the requirements for the LTS.

In this final rule, we are promulgating emission limits, compliance schedules and other requirements for four BART sources and two RP sources. This final action completes the LTS measures needed to achieve emission reductions for out-of-state Class I areas, emission limitations and schedules for compliance to achieve the RPGs, and enforceability of emission limitations and control measures.

In particular, as explained above, we are incorporating by reference provisions of the NESHAP for primary copper smelters to ensure that Arizona's BART determinations for PM

at the Hayden and Miami Smelters are made enforceable and are included in the applicable implementation plan.

77 FR 75512-72580, December 5, 2012.

40 CFR 51.308(d)(3)(ii), (v)(C) and (v)(F).

See 78 FR 46173 (codified at 40 CFR 52.145(e)(ii)).

E. Interstate Visibility Transport

EPA is finalizing its determination that the control measures in the Arizona RH SIP and FIP are adequate to prevent Arizona's emissions from interfering with other states' required measures to protect visibility. Thus, the combined measures from both plans satisfy the interstate transport visibility requirement of CAA section 110(a)(2)(D)(i)(II) for the 1997 8-hour ozone, 1997 PM

2.5

, and 2006 PM

2.5

NAAQS. In our final rule published on July 30, 2013, EPA disapproved these

SIP submittals with respect to the interstate transport visibility requirement for each of these NAAQS, triggering the obligation for EPA to promulgate a FIP.

78 FR 46142, July 30, 2013.

F. Other Changes From Proposal

Our proposed regulatory text incorporated by reference certain provisions of the Arizona Administrative Code that establish an affirmative defense for excess emissions due to malfunctions. We did not receive any comments on this aspect of our proposal. Following the close of the public comment period, the United States Court of Appeals for the D.C. Circuit issued a decision concerning various aspects of the NESHAP for Portland cement plants issued by EPA in 2013, including the affirmative defense provision of that rule.

The court found that EPA lacked authority to establish an affirmative defense for private civil suits and held that under the CAA, the authority to determine civil penalty amounts lies exclusively with the courts, not EPA. The court did not address whether such an affirmative defense provision could be properly included in a SIP. However, the court's holding makes it clear that the CAA does not authorize promulgation of such a provision by EPA. In particular, the court's decision turned on an analysis of CAA sections 113 (“Federal enforcement”) and 304 (“Citizen suits”). These provisions apply with equal force to a civil action brought to enforce the provisions of a FIP. The logic of the court's decision thus applies to the promulgation of a FIP and precludes EPA from including an affirmative defense provision in a FIP. Therefore, we are not including an affirmative defense provision in the final FIP.

NRDC

EPA,

2014 U.S. App. LEXIS 7281 (D.C. Cir.).

We note that, if a source is unable to comply with emission standards as a result of a malfunction, EPA may use case-by-case enforcement discretion, as appropriate. Further, as the D.C. Circuit recognized in an EPA or citizen enforcement action, the court has the discretion to consider any defense raised and determine whether penalties are appropriate.

Id. at 24 (arguments that violations were caused by unavoidable technology failure can be made to the courts in future civil cases when the issue arises).

V. Responses to General Comments

A. Introduction

EPA provided 60 days for the public to submit comments on the proposed rule, with the comment period concluding on March 31, 2014. We held two public hearings in Arizona, one on February 25, 2014, in Phoenix and another on February 26, 2014, in Tucson. The deadline for public comments was March 31, 2014. Certified records of the public hearings, written comments (excluding any confidential business information (CBI) materials), a summary of comments, and a list of commenters are available in the docket for this final action. We received a total of 24 written comments from industry or industrial associations (13), environmental groups (6), citizens (3), a state agency (1), and a federal agency (1). In addition, 14 individuals presented oral testimony at the two hearings. Summaries of significant comments and EPA's responses, organized by subject matter, are provided in the following sections. Because we received no comments regarding the LTS or interstate transport provisions of the FIP, there is no section in this notice addressing comments on these topics.

We are using the following acronyms to refer to representatives of the following entities who submitted comments to us:

• ACCCE—American Coalition for Clean Coal Energy

• ADEQ—Arizona Department of Environmental Quality

• AMA—Arizona Mining Association

• ANGA—America's Natural Gas Alliance

• ASARCO—American Smelting and Refining Company

• CPC—CalPortland Company

• Earthjustice

Comments were provided by Earthjustice on behalf of the National Parks Conservation Association, Sierra Club, San Juan Citizens Alliance, and Arizona Chapter of Physicians for Social Responsibility.

• EPNG—El Paso Natural Gas Company

• FMMI—Freeport-McMoRan Miami, Inc.

• LNA—Lhoist North America of Arizona

• NMA—National Mining Association

• NPS—National Park Service

• PCC—Phoenix Cement Company

• PSR—Physicians for Social Responsibility

• TEP—Tucson Electric Power

• TPMEC—Tucson Pima Metropolitan Energy Commission

B. Comments on State and EPA Actions on Regional Haze

Comment:

One commenter, a former member of the Technical Oversight Committee of the Western Regional Air Partnership (WRAP), recounted the history of the Grand Canyon Visibility Transport Commission and the WRAP, and their efforts under section 309 of the original RHR to develop emission reduction milestones through 2018 for SO

emissions from large industrial sources in the nine-state Commission Transport Region that affects the Colorado Plateau. The commenter noted that Arizona ultimately withdrew from the section 309 process, but asserted that the State's withdrawal should not negate the effort of setting the milestones and the agreements reached during that process. The commenter asserted that by rejecting Arizona's SIP and proposing a FIP, EPA has gone beyond what was agreed on as a reasonable expectation of BART for specific groups of sources, such as smelters, utilities, and cement plants. The commenter added that the new SO

NAAQS will require plants to make changes that go well beyond BART. Therefore, BART should be set at a level no more stringent than what WRAP proposed so as not to interfere with any plans for the nonattainment areas to come into compliance with the new SO

standard.

Response:

These comments largely pertain to EPA's partial disapproval of Arizona's 308 RH SIP and are therefore untimely, as EPA has already taken final action on the SIP.

Furthermore, EPA has already disapproved the majority of Arizona's 309 RH SIP.

As explained further below in response to similar comments regarding the Hayden and Miami Smelters, this FIP will not adversely impact the smelters' ability to come into compliance with the 1-hour SO

NAAQS.

78 FR 46142.

78 FR 48326.

C. Comments on State and Federal Roles in the Regional Haze Program

Comment:

Several commenters (ADEQ, FMMI, AMA, ACCCE and NMA) do not agree with EPA's partial disapproval of Arizona's RH SIP, asserting that EPA has overstepped its boundaries by unnecessarily imposing a FIP. Some of the commenters contend that states are best suited to make BART determinations, not EPA.

ADEQ noted that the RHR is not intended to protect public health, but to address visibility problems. In the commenter's opinion, EPA should have given the State of Arizona the

opportunity to correct specific issues in the SIP, instead of proceeding with a FIP. Citing to CAA section 110(c), ADEQ asserted that EPA should end this rulemaking and allow ADEQ a period of up to two years to correct any deficiencies in its RH SIP. ACCCE discussed the history of the regional haze program and emphasized the discretion provided to states under the CAA and the RHR. FMMI stated that EPA lacks the authority to disapprove a SIP and promulgate the proposed FIP based on its policy disagreements with a state. AMA and NMA asserted that EPA had overstepped its boundaries and should leave the decision of what constitutes BART and reasonable progress to the State of Arizona. NMA proceeded to argue that this is not the first example of EPA going beyond its authority as it relates to regional haze, since it has replaced the regional haze determinations of 14 states with its own federal requirements. NMA went on to say that in the case of the Arizona RH SIP, EPA disapproved parts of the plan due to its own subjective opinion and not because the SIP was inconsistent with the requirements of the CAA.

Response:

To the extent these comments pertain to EPA's partial disapproval of the Arizona RH SIP or other previous SIP actions, they are untimely. To the extent that the comments are relevant to the proposed FIP, we do not agree with their substance. While it is our strong preference that state plans implement CAA requirements, there are circumstances in which a FIP is required by the Act. As explained in response to comments on the Phase 1 Final Rule

and our legal brief responding to petitions for review of that rule,

we are required by the CAA to issue a FIP to meet all requirements of the RHR not addressed by an approved SIP revision. In particular, CAA section 110(c) requires EPA to promulgate a FIP at any time within two years of (1) finding that a State has failed to make a required submission, or (2) disapproving a State submission in whole or in part. This obligation is eliminated only if “the State corrects the deficiency, and the Administrator approves the plan or plan revision, before the Administrator promulgates such Federal Implementation plan.” In this instance, two different triggering events under section 110(c) have occurred: EPA has made a finding that the State failed to make a required submission and has partially disapproved the submissions that the State subsequently made.

77 FR 72568-69 (December 5, 2012).

Brief of Respondent, Arizona v. EPA, No. 13-70366 (9th Cir. Dec. 12, 2013) (EPA Phase 1 Brief) at 66-77.

EPA found that Arizona had failed to submit a comprehensive regional haze SIP in January 2009, which triggered an obligation for EPA to promulgate a FIP within two years, unless the State first submitted and EPA approved a regional haze SIP.

When EPA failed to either approve a SIP or promulgate a FIP by the January 2011 deadline, we were sued by a group of conservation organizations.

In order to resolve this lawsuit, EPA entered into a Consent Decree that established deadlines for action on regional haze plans for various states, including Arizona. This decree was entered and later amended by the United States District Court for the District of Columbia over the opposition of Arizona.

Under the terms of the Consent Decree, as amended, EPA was subject to three sets of deadlines for taking action on the Arizona RH SIP as listed in Table 4. The specific deficiencies that commenters claim to have identified in EPA's proposal are addressed in subsequent responses.

74 FR 2392-93 (January 15, 2009).

National Parks Conservation Association

Jackson

(D.D.C. Case 1:11-cv-01548).

Nat'l Parks Conservation Ass'n

Jackson

(D.D.C. Case 1:11-cv-01548), Memorandum Order and Opinion (May 25, 2012), Minute Order (July 2, 2012), Minute Order (November 13, 2012), Minute Order (February 15, 2013), Order (September 6, 2013), and Stipulation to Amend Consent Decree (November 14, 2013). On appeal, the D.C. Circuit upheld the District Court's finding that it lacked jurisdiction over Arizona's objections.

Nat'l Parks Conservation Ass'n

EPA,

43 ELR 20266 (D.C. Cir. 2013).

Table 4—Consent Decree Deadlines for EPA To Act on the Arizona RH SIP and FIP

EPA actions

Proposed rule

signature date

Final rule signature date

Phase 1—BART determinations for Apache, Cholla and Coronado

July 2, 2012

November 15, 2012.

Phase 2—All remaining elements of the Arizona RH SIP

December 8, 2012

July 15, 2013.

Phase 3—FIP for disapproved elements of the Arizona RH SIP

January 27, 2014

June 27, 2014.

Published in the

Federal Register

on July 20, 2012, 77 FR 42834.

Published in the

Federal Register

on December 5, 2012, 77 FR 72512.

Published in the

Federal Register

on December 21, 2012, 77 FR 75704.

Published in the

Federal Register

on July 30, 2013, 78 FR 46142. Also addresses supplemental proposal published in the

Federal Register

on May 20, 2013, 78 FR 29292.

Published in the

Federal Register

on February 18, 2014.

In Phase 1, EPA approved in part and disapproved in part Arizona's BART determinations for Apache Generating Station, Cholla Power Plant, and Coronado Generating Station, and promulgated a FIP addressing the disapproved portions of the SIP.

In our initial Phase 2 proposal, EPA proposed to approve in part and disapprove in part the remainder of the Arizona RH SIP.

In May 2013, ADEQ submitted a SIP Supplement that addressed some of the elements that EPA had proposed to disapprove. We then proposed to approve in part and disapprove in part the SIP Supplement.

We finalized our partial approval and partial disapproval on July 30, 2013.

We have also disapproved the majority of Arizona's submittal under Section 309 of the RHR.

Given these disapprovals, and our previous finding of failure to submit, EPA is required under CAA section 110(c) to promulgate a FIP for the disapproved portions of the SIP. Indeed, even if we had not previously found that Arizona failed to submit a comprehensive regional haze SIP, we nonetheless would be authorized to promulgate a partial FIP following our partial disapprovals of Arizona's 308 and 309 RH SIPs.

As noted above, however,

EPA remains willing to work with ADEQ on a SIP that would be designed to replace this FIP once such a SIP was submitted and approved by us.

77 FR 72512 (December 5, 2012).

77 FR 75704 (December 21, 2012).

78 FR 29292 (May 20, 2013).

78 FR 46142 (July 30, 2013).

78 FR 48326 (August 8, 2013).

See EPA v. EME Homer City Generation, 134 S. Ct. 1584 (2014), Slip. Op. at 16 (“After EPA has disapproved a SIP, the Agency can wait up to two years to issue a FIP . . . But EPA is not obliged to wait two years or postpone its action even a single day: The Act empowers the Agency to promulgate a FIP `at any time' within the two-year limit.”).

VI. Responses to Comments on EPA's Proposed BART Determinations

A. Comments on Sundt Generating Station Unit 4

1. BART Eligibility

Comment:

Three commenters (ADEQ, TEP, and ACCCE) argued against EPA's proposed finding that Sundt Unit 4 is BART-eligible, and two commenters (Earthjustice and NPS) supported EPA's finding. ADEQ asserted that EPA has no authority to impose BART on Sundt Unit 4 because ADEQ determined that the unit is not BART-eligible. ADEQ noted that under CAA section 169(b)(2)(A), major sources that existed as of August 7, 1962, are considered BART-eligible. However, the statute does not address sources that existed during that time, but were reconstructed after 1977 (Sundt Unit 4 was reconstructed in 1987). According to ADEQ, “EPA filled that gap by adopting regulations treating `reconstructed' units as `new' units.”

ADEQ further noted that the BART Guidelines provide that “any emissions unit for which a reconstruction `commenced' after August 7, 1977, is not BART-eligible” and argued that ADEQ's determination that Sundt Unit 4 is not BART-eligible was consistent with EPA's regulations. ADEQ asserted that EPA rejected the determination on the basis that EPA is not bound by its own guidelines and argued that that it was inappropriate for EPA to fault ADEQ for following guidance that EPA maintains is “persuasive” evidence of the requirements of the CAA. The commenter further argued that the BART Guidelines are clear that any unit that was reconstructed after 1977 is not BART-eligible, but that despite this, EPA has indicated that it does not interpret the BART Guidelines to apply to Sundt Unit 4 because the unit never went through prevention of significant deterioration (PSD) permitting. ADEQ argued that “EPA is not authorized, in the guise of `interpreting' its BART Guidelines, to engage in what amounts to post-hoc rulemaking, by amending its BART Guidelines to make units that are reconstructed after 1977, but which did not obtain PSD permits BART-eligible.”

ADEQ also commented that EPA has ignored the policy reasons that Congress had for excluding reconstructed units such as Sundt Unit 4 from PSD and other requirements. The commenter noted that the Power Plant and Industrial Fuel Use Act of 1978 (FUA), which amended the Energy Supply and Environmental Coordination Act of 1974 (ESECA), authorized the Department of Energy (DOE) to require electric utilities to convert generating stations using oil and natural gas to using coal to reduce the Unites States' dependency on foreign oil and increase its use of indigenous energy resources. ADEQ stated that because Congress wished to ensure the conversion took place, these units were exempted from “environmental requirements.” Therefore, BART should not be required for Sundt Unit 4.

TEP, the owner of the Sundt facility, incorporated by reference the comments it submitted on EPA's proposed partial disapproval of the Arizona RH SIP, in which the commenter opposed EPA's position that Sundt Unit 4 is BART-eligible, and reiterated its position that Sundt Unit 4 is not BART-eligible. Similarly, ACCCE asserted that, “ADEQ's determination that Sundt Unit 4 was reconstructed in the 1980s, and therefore is not BART-eligible was reasonable and should not have been disapproved by EPA.” In contrast, Earthjustice and NPS expressed support for EPA's finding that Sundt Unit 4 is BART-eligible because it did not go through PSD review when it was reconstructed in 1987. Earthjustice asserted that a source reconstructed after 1977 must install either BART controls under the regional haze program or Best Available Control Technology (BACT) controls under the PSD program.

Response:

To the extent that the comments concern EPA's partial disapproval of the Arizona RH SIP, they are untimely, as EPA has already taken final action on the SIP.

Further, we have already addressed many of the commenters' assertions in our proposed and final actions on the SIP and in the Sundt Memo,

all of which are included in the docket for this action. To the extent the comments raise new issues, we address them here.

78 FR 46142.

78 FR 75722 and TEP Sundt Unit I4 BART Eligibility Memo (November 21, 2012) (Sundt Memo).

Contrary to ADEQ's assertion, the RHR does not indicate that “reconstructed” units are to be treated as “new” units for all purposes. In particular, the RHR does not indicate that a source that is reconstructed after 1977 is considered BART-ineligible. Likewise, nothing in the preamble to the 1980 rule regarding Reasonably Attributable Visibility Impairment (RAVI), in which EPA promulgated the definition of “BART-eligible,” or the preamble to the 1999 RHR itself suggests that a post-1977 reconstruction would exempt a source from BART.

The BART Guidelines do state that “any emissions unit for which a reconstruction `commenced' after August 7, 1977, is not BART-eligible.”

However, this statement in the BART Guidelines must be read in the context of the applicable regulatory requirements and associated preambles, none of which even mention such an exemption for post-1977 reconstructions. In particular, the preamble to the BART Guidelines indicates that the post-1977 reconstruction exemption set out in the BART Guidelines is limited to “sources reconstructed after 1977,

which reconstruction had gone through NSR/PSD permitting.”

Although not binding, this statement in the preamble confirms that EPA did not intend to create a blanket exemption for all post-1977 reconstructions in the BART Guidelines. Indeed, it would only make sense to exempt a reconstructed unit from BART if that unit had gone through NSR/PSD permitting to ensure that its emissions were subject to modern-day pollution controls. Sundt Unit 4 never went through such permitting. Thus, we do not agree that we are effectively amending the BART Guidelines or engaging in post hoc rulemaking by applying an interpretation that is consistent not only with the CAA and RHR, but also with the preamble to the BART Guidelines themselves.

See 45 FR 80084, 64 FR 35714.

70 FR 39160.

70 FR 39111.

We also do not agree that Congress intended to provide a general exemption from all “environmental requirements” for units that were converted to coal under the FUA and ESECA. The relevant section of FUA, codified in CAA section 111(a)(8), provides that “[a] conversion to coal . . . by reason of an order under section 2(a) of the [ESECA] or any amendment thereto, or any subsequent enactment which supersedes such Act . . . shall not be deemed to be a modification

for purposes of paragraphs (2) and (4) of [CAA subsection 111(a)].”

Paragraphs (2) and (4), in turn, contain the definitions of “new source” and “modification” that apply to the Act's new source performance standards (NSPS) requirements.

The definition of “modification” in paragraph 111(a)(4) also applies for purposes of the PSD

provisions of the Act.

However, nothing in the Act indicates that Congress intended the exemption in section 111(a)(8) to extend to other provisions of the Act, such as the visibility protection provisions of Section 169A. If Congress had intended to provide such an exemption from BART eligibility for units that were converted to coal under the FUA and ESECA, it could have added such an exemption to section 169A. It did not do so. Thus, for the reasons set out in the Sundt Memo, in our Phase 2 proposed and final rulemakings, and in this response, we are finalizing our proposed determination that Sundt 4 is BART-eligible.

42 U.S.C. 7411(a)(8) (emphasis added).

42 U.S.C. 7411(a)(2) and (4).

42 U.S.C. 7479(2)(C).

2. BART Analysis and Determination for NO

Comment:

ADEQ indicated that it does not support EPA's proposed limit for NO

that is based on SNCR control technology. ADEQ asserted that the significant cost of installing and operating SNCR ($3 million in construction and $1 million in annual operating costs) does not justify the limited visibility improvement that would result from adding this control technology. ADEQ said that EPA's analysis, which ADEQ described as suspect, shows an improvement of only 0.5 dv. ACCCE also objected to EPA's decision to require SNCR, arguing that it is costly and results in no perceptible improvement in visibility. ACCCE discussed the installation costs and the cost-effectiveness of SNCR on Unit 4, and stated that none of the Class I areas affected by Sundt Unit 4 will experience a greater than a 1.0 dv improvement from the installation of SNCR. This “modest” improvement is inconsistent, ACCCE said, with EPA's position that considers 1.0 dv change or more from an individual source as causing visibility impairment and a 0.5 dv change as contributing to impairment.

Response:

We disagree with these comments. Regarding the costs of compliance, although the installation and operation of SNCR will result in TEP incurring certain initial investments and ongoing operational costs, we consider the total annualized cost warranted based on the amount of NO

removed and the expected visibility benefits. As noted in our proposed rule, SNCR at this source has a cost-effectiveness of about $3,200/ton, which we consider very cost-effective. With regard to visibility improvement, we do not agree that only visibility improvements that by themselves result in humanly perceptible changes are relevant. The CAA and RHR require, as part of each BART analysis, consideration of “the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology.”

The Act and RHR do not require that the improvement from a single source be perceptible in order to be meaningful. As EPA explained in the preamble to the BART Guidelines: “Even though the visibility improvement from an individual source may not be perceptible, it should still be considered in setting BART because the contribution to haze may be significant relative to other source contributions in the Class I area.”

Thus, we disagree that the degree of visibility improvement should be contingent upon perceptibility.

CAA section 169A(g)(2), 40 CFR 51.308(e)(1)(ii)(A).

70 FR 39129.

In our visibility improvement analysis, we have not considered perceptibility as a threshold criterion for considering improvements in visibility. Rather, we have considered visibility improvement in a holistic manner, taking into account all reasonably anticipated improvements in visibility expected to result at all Class I areas within 300 kilometers of each source. Improvements smaller than 0.5 dv may be warranted considering the number of Class I areas involved and the baseline contribution to impairment of the source in question. For example, a source with a 0.5 dv impact at a Class I area “contributes” to visibility impairment and must be analyzed for BART controls. Controlling such a source will not result in perceptible improvement in visibility, but Congress nevertheless determined that such contributing sources should nevertheless be subject to the BART requirement. In the aggregate, small improvements from controls on multiple BART sources and other sources will lead to visibility progress. As a result, although we described the anticipated visibility benefits from the installation of SNCR as “modest,” we still consider those benefits sufficient to justify SNCR as BART in light of the fact that SNCR will be highly cost-effective and has no substantial adverse energy or non-air quality environmental impacts. This has been EPA's consistent interpretation in many regional haze determinations.

Comment:

ADEQ indicated that it supports EPA's rejection of an emission limit equivalent to SCR as BART for NO

at Sundt Unit 4 due to costs. In contrast, Earthjustice asserted that EPA should have set a BART emission limit that reflects the use of SCR at Sundt Unit 4, rather than the less effective SNCR technology. Earthjustice stated that EPA erred when it concluded that the visibility benefits of SCR were not worth the costs after EPA acknowledged that SCR provides substantially greater visibility improvements than SNCR. Earthjustice stressed that EPA's calculated cost-effectiveness value of $5,176 per ton of NO

removed for SCR is within the range of what has been deemed cost-effective in many other instances, based on examples provided in Exhibit 33 submitted with the comments. Earthjustice added that EPA provided no justifiable rationale for rejecting the overall cost-effectiveness value and relying on the incremental cost-effectiveness value for the rejection. Earthjustice also contended that EPA improperly rejected SCR based on numerous erroneous assumptions in its cost analysis that increased the cost-effectiveness values (i.e., $/ton) for SCR. In particular, Earthjustice asserted that EPA used an unreasonably low capacity factor of 0.49, even though a higher and more appropriate capacity factor would have made the SCR controls more cost-effective. Earthjustice also noted that EPA used a retrofit factor for SCR of 1.5, instead of the standard retrofit factor of 1.0, but asserted that EPA did not provide a sufficient reason to enhance the retrofit factor. According to Earthjustice, correcting these two assumptions would make SCR cost-effective to control NO

at Sundt Unit 4 at an emission rate of 0.05 lb/MMBtu.

Response:

We disagree that we improperly rejected SCR. In reaching our BART determination, we have considered both average and incremental costs as well as expected visibility benefits.

In particular, we estimate the average cost-effectiveness of SCR to be $5,176/ton. EPA has not previously required installation of controls with an average cost-effectiveness value this high for purposes of BART.

Similarly, the estimated incremental cost-effectiveness for SCR (compared to SNCR) of $6,174/ton is on the high end of what we have required for purposes of BART.

Such cost values might be warranted if the expected visibility benefits were very high (i.e., over one deciview at a single Class I area or several cumulative deciviews across multiple affected Class

I areas). However, we do not consider this level of cost to be justified here by the expected visibility benefits for SCR of 0.78 dv for the most improved Class I area and 1.6 dv cumulative for all affected Class I areas.

See 79 FR 9329.

See, e.g., BART EGU FIP Summary.

Id. The only example with a higher incremental cost-effectiveness value is Dave Johnston Unit 3 in Wyoming ($7,583/ton based on a remaining useful life of 20 years).

The information provided by Earthjustice regarding the range of $/ton values considered cost-effective is derived from other regulatory programs such as Best Available Control Technology (BACT) determinations for construction of new sources in attainment areas, and Lowest Achievable Emission Rate determinations for construction of new sources in nonattainment areas. The statutory requirements, calculation methodology, and regulatory drivers that may inform a determination of emission reductions appropriate for these programs are not necessarily comparable to those of the Regional Haze program, which is a retrofit program where older sources are required to add pollution controls. We therefore do not consider it appropriate simply to conclude that costs found to be acceptable in other programs are necessarily appropriate in a BART determination.

We also disagree with Earthjustice's assertion that our cost analysis for SCR is based on faulty assumptions. We recognize that a higher capacity factor would result in an increase in the calculated amount of NO

reduced. We also recognize that, historically, Sundt Unit 4 operated at higher capacity factors, ranging from 0.60 to 0.75. However, a review of data from EPA's Clean Air Markets Division (CAMD) Acid Rain Program database indicates that, starting in 2009 and continuing into the present, Sundt Unit 4 has consistently operated at substantially lower capacity factors.

Our use of a 0.49 capacity factor is therefore not based on a single, abnormal year of low capacity, but rather represents an average of multiple, recent years of low capacity at Sundt Unit 4. Given the length of time that Sundt Unit 4 has operated at these low capacity levels, we consider our use of a 0.49 capacity factor in emission calculations to be a “realistic depiction of anticipated annual emissions.”

This emission and generation data was contained in the docket for our proposal, E-45—TEP Sundt4 2001-12 Emission Calcs 2014-01-24.xlsx.

See 70 FR 39167.

Moreover, we disagree with the Earthjustice's assertion that our use of a 1.5 retrofit factor is unsupported in the record. Although the factors contributing to retrofit difficulty were summarized as “certain difficulties” in our TSD, this information is described in detail in the modeling and cost information provided by TEP on May 10, 2013.

Our cost calculations specifically noted the changes we made to account for these factors.

Specifically, a detailed description of these issues is contained on page 6, Attachment C, in TEP's letter dated May 10, 2013. These issues include interference from existing boiler structures and material handling equipment that makes the most common SCR reactor impractical, the need for substantial modifications to the existing air preheater, and site congestion around the boiler that complicates siting of an SCR system. We consider these issues sufficient to warrant a higher retrofit factor.

TEP's May 10, 2013 letter describing this information was contained in the docket for our proposal, C-37 Letter from Erik Bakken, TEP, to Greg Nudd, EPA, re TEP Sundt Modeling & Cost Information.pdf.

Our cost calculations, which note these upward revisions, were contained in the docket for our proposal, E-05 TEP Sundt4 Control Costs (final for NPRM docket).xlsx.

Comment:

In response to EPA's request for comment on whether EPA should use a less stringent SCR emission limit in its NO

BART analysis for Sundt Unit 4, Earthjustice responded in the negative. According to the commenter, EPA's use of a 0.05 lb/MMBtu limit for SCR is consistent with EPA's BART determinations for other coal-fired power plants for which EPA has repeatedly concluded that a 0.05 to 0.055 lb/MMBtu emission limit is BART. In addition, citing reports submitted with the comments, Earthjustice asserted that SCRs often achieve more stringent emission rates and control efficiencies than EPA assumed SCR would achieve at Sundt Unit 4. Earthjustice stated that because a 0.05 lb/MMBtu emission rate is achievable with SCR at Sundt Unit 4, EPA should not use a less stringent emission limit in its BART analysis.

Response:

We agree that our use of a 0.05 lb/MMBtu annual average design value for SCR is consistent with other BART determinations for coal-fired power plants.

Comment:

Earthjustice stated that if EPA does not revise its BART determination to require SCR, it should set a more stringent emission limit that more accurately reflects the emission reductions achievable with SNCR. Earthjustice quoted the BART Guidelines as requiring EPA to “take into account the most stringent emission control level that the technology is capable of achieving,” which Earthjustice said EPA has not done in this case. Earthjustice asserted that EPA should select a level of NO

reduction for SNCR in the range of 50 percent over and above the existing combustion controls, rather than the level of 30 percent above current controls that was selected. As support, Earthjustice noted that SNCR is required by the pending SIP revision (prepared by ADEQ to replace the FIP) for Apache Unit 3 to reduce NO

from 0.43 lb/MMBtu down to 0.23 lb/MMBtu, or roughly 50 percent. Earthjustice recommended that EPA set an emission limit for SNCR in the range of 0.22 lb/MMBtu, reflecting 50 percent reduction from the baseline level of 0.445 lb/MMBtu of NO

in 2011. In addition, Earthjustice disagreed with EPA's inflation of the NO

emission limit by 17 percent to account for variability. According to Earthjustice, EPA assumed without justification that the observed variability without SNCR would be the same as variability with SNCR.

Response:

We disagree with this comment. The Apache Unit 3 example cited by Earthjustice does not support a 50 percent SNCR control efficiency. The 0.43 lb/MMBtu emission rate on Apache Unit 3 noted by Earthjustice reflects the use of over fire air (OFA) only. The 0.23 lb/MMBtu emission rate on Apache Unit 3 noted by Earthjustice reflects the use of LNB with OFA and SNCR. The approximate 50 percent reduction from 0.43 to 0.23 is not solely attributable to SNCR, but rather is the result of the application of LNB and SNCR. Since Sundt Unit 4 already operates with LNB and OFA, we do not consider it appropriate to assume that application of SNCR will result in an additional 50 percent NO

reduction.

With regard to our upward revision to the annual emission rate to develop a rolling 30 day emission limit, we acknowledge that observed variability without SNCR might not be the same as variability with SNCR. We note, however, that even emission units with well-operated controls will experience some degree of emissions variability. As noted in our proposed rule, we developed this upward revision based on site-specific emission data reported to the CAMD for Sundt Unit 4. Given the site-specific basis for our upward revision, we consider it a reasonable estimate of emission variability. We acknowledge that there might be other methods of accounting for this variability. However, we did not receive any comments that described or proposed any such alternate methodology. Accordingly, we are finalizing the emission limit as proposed.

Comment:

NPS indicated that it agrees with the design emission rate of 0.050 lb/MMBtu that EPA used to estimate the control effectiveness of SCR. However, NPS did not agree with the cost of catalyst for SCR of $8,000 per cubic meters (m

), and cited to a recent report indicating the costs are around $5,000/m

. NPS also said that EPA did not consider using regenerated catalyst at a cost of $5,500/m

, which it did in the recent Wyoming RH FIP.

NPS also stated that instead of relying only on the Integrated Planning Model (IPM) to estimate the costs of SCR, NPS used a method similar to what EPA Region 8 used for Colstrip in Montana. In NPS's opinion, using IPM to calculate capital costs and EPA's Control Cost Manual (CCM) to calculate operating costs provides more flexibility, provides greater transparency and is more in line with the BART Guidelines that recommend following EPA's CCM as much as possible.

Response:

We disagree with the NPA's assertion that $8,000/m

is an unreasonable cost estimate for catalyst. Since catalyst prices fluctuate, we recognize that recent prices may be lower than the value used in our cost calculations. However, given that catalyst is an operating cost that will be periodically incurred over the entire useful life of the equipment,

we consider it appropriate to use a catalyst price that reflects more than just recent price levels. The BART Guidelines state, “In order to maintain and improve consistency, cost estimates should be based on the OAQPS Control Cost Manual, where possible” and that “[w]e believe that the Control Cost Manual provides a good reference tool for cost calculations, but if there are elements or sources that are not addressed by the Control Cost Manual or there are additional cost methods that could be used, we believe that these could serve as useful supplemental information.”

As noted in our proposed rule and TSD,

EPA has used IPM in multiple regulatory actions, and considers it an appropriate source of supplemental information.

As opposed to capital costs, which are incurred only once, at the start of the project.

BART Guidelines, 40 CFR Part 51, Appendix Y, section IV.D.4.a.

TSD for the Proposed Phase 3 FIP, January 27, 2013, Page 19 of 233.

3. BART Analysis and Determination for SO

Comment:

ACCCE opposed EPA's proposal to require DSI for the control of SO

emissions at Sundt Unit 4. The ACCCE asserted that this requirement will have no humanly perceptible visibility improvement, so the proposal must be withdrawn. According to ACCCE, the highest visibility improvement expected from this requirement is 0.20 dv at Saguaro National Park. At the other nine affected Class I areas, the visibility improvement is expected to range from only 0.04 to 0.10 dv. ACCCE contended that requiring costly controls with no humanly perceptible visibility improvement is unjustified.

Response:

As noted in our response to a similar comment regarding our NO

BART determination, we have not considered perceptibility as a threshold criterion for considering improvements in visibility. Rather, we have considered visibility improvement in a holistic manner, taking into account all reasonably anticipated improvements in visibility expected to result at all Class I areas within 300 kilometers of each source. Improvements smaller than 0.5 dv may be warranted considering the number of Class I areas involved and the initial contribution to impairment of the source in question. For example, a source with a 0.5 dv impact at a Class I area “contributes” to visibility impairment and must be analyzed for BART controls. While controlling such a source will not result in perceptible improvement in visibility, Congress determined that such contributing sources should nevertheless be subject to the BART requirement. In the aggregate, small improvements from controls on multiple BART sources and other sources will lead to visibility progress. As a result, although the anticipated visibility benefit attributable to DSI is not humanly perceptible, we consider those benefits sufficient to justify DSI as BART in light of the fact that DSI will be highly cost-effective and has no substantial adverse energy or non-air quality environmental impacts.

Comment:

Earthjustice stated that EPA should revise its BART analysis for SO

to reflect more stringent emission rates achievable with wet flue gas desulfurization (FGD) and dry FGD because the BART Guidelines require EPA to analyze the most stringent emission control level that the technology is capable of achieving. According to Earthjustice, EPA assumed that wet FGD would achieve a 0.06 lb/MMBtu emission rate (92 percent control efficiency) and dry FGD would achieve a 0.08 lb/MMBtu emission rate (89 percent control efficiency). Earthjustice argued that these figures were cited despite EPA's acknowledgment that both wet FGD and dry FGD are capable of achieving more stringent emission rates. Earthjustice added that reports submitted with its comments show that both wet and dry FGD can achieve emission rates of 0.04 lb/MMBtu or lower along with control efficiencies of 95 to 99 percent.

Response:

We disagree that we underestimated the SO

emission reductions achievable with dry or wet FGD. In our proposed rule, and in the TSD for our proposed rule, we stated that:

[B]oth dry and wet FGD have very high incremental cost-effectiveness values, indicating that while they are more effective than the preceding control, this additional degree of effectiveness comes at a substantial cost.

The incremental cost-effectiveness of dry FGD, in relation to DSI, is approximately $17,000/ton. Assuming a more stringent dry or wet FGD emission rate of 0.04 lb/MMBtu, the incremental cost-effectiveness of FGD, relative to DSI, is approximately $13,000/ton, which is still not within a range that EPA or states have considered cost-effective, especially given that FGD (dry or wet) is expected to result in less visibility improvement than DSI.

As a result, a more stringent FGD emission rate would not alter our SO

BART determination.

See 79 FR 9332-33.

Comment:

Earthjustice asserted that EPA improperly raised the proposed SO

limit (based on use of DSI) from 0.21 to 0.23 lb/MMBtu. Earthjustice said that this increase was inappropriate, as it was based on SO

emission data that did not account for controls. Since proper controls dampen the variability of emissions, Earthjustice said that the emission limit should not be raised to account for variability.

Response:

As noted in a response to a similar comment regarding our NO

BART determination, we acknowledge that observed emissions variability at Sundt Unit 4 without SO

controls may not be the same as its emissions variability when operating with DSI. We note, however, that even emission units with well-operated controls will experience some degree of emissions variability. As noted in our proposed rule, we developed this upward revision based on site-specific emission data reported to EPA's CAMD for Sundt Unit 4. Given the site-specific basis for our upward revision, we do not consider it as an unreasonable estimate of emissions variability. We acknowledge that there might be other methods of accounting for this variability. However, we did not receive any comments that described or proposed any such alternate methodology. Therefore, we are finalizing the SO

emission limit of 0.23 lb/MMBtu as proposed.

4. BART Analysis and Determination for PM

Comment:

ADEQ indicated that it supports EPA's decision to require BART for particulate matter (PM) in terms of a PM

limit of 0.03 lb/MMBtu. While agreeing that fabric filter baghouses are the best technology for PM reductions from Sundt Unit 4, Earthjustice asserted that EPA should set a lower emission limit as BART. According to Earthjustice, stack test results for PM

show that the existing baghouses at Sundt Unit 4 can achieve lower emission rates than the 0.03 lb/MMBtu rate that EPA proposed as BART (citing the TSD at 23). Earthjustice stated that there are hundreds of instances of coal units with baghouses achieving emission rates lower than 0.03 lb/MMBtu, citing the docket for the Mercury Air Toxics Standards (MATS).

Response:

We disagree that the proposed 0.030 lb/MMBtu emission limit for filterable PM

is too high. The 0.022 lb/MMBtu emission rate summarized on page 23 of the TSD is the average of multiple test runs that range from 0.016 lb/MMBtu to 0.039 lb/MMBtu.

Emission limitations under the CAA must be continuous and BART must be an emission limitation that is achievable.

Thus, a BART emission limitation should be one that a facility can continuously achieve. The performance test data indicate that a PM emission limit of 0.030 lb/MMBtu is achievable by the facility, and will also result in actual emission reductions. In addition, the BART limit is substantially lower than the PM limit contained in the facility's current operating permit,

substantially decreasing the PM emissions authorized at the facility.

The original Method 5 test results are included as Docket Item F-28—TEP Sundt4 Test Results.pdf.

42 U.S.C. 7602(k) (definition of “emission limitation”); 40 CFR 51.301 (definition of “BART”).

233 lb/hour, per page 2 of the TSD. The BART limit would be equivalent to approximately 41 lb/hour.

MATS establishes an emission limit of 0.030 lb/MMBtu for filterable PM (as a surrogate for toxic non-mercury metals) as representing MACT for coal-fired electric generating units (EGUs). The BART Guidelines provide that “unless there are new technologies subsequent to the MACT standards which would lead to cost-effective increases in the level of control, you may rely on the MACT standards for purposes of BART.”

We consider baghouses to be the most stringent PM control technology for coal-fired EGUs. Moreover, the commenter has not identified a new or more stringent technology. As a result, we consider 0.030 lb/MMBtu to be an appropriate continuously achievable BART limit for Sundt Unit 4.

BART Guidelines, Section IV.C. “How does a BART review relate to Maximum Achievable Control Technology (MACT) Standards under CAA section 112, or to other emission limitations required under the CAA?”

5. Better-than-BART Alternative

Comment:

Multiple commenters expressed support for the “better-than-BART alternative” for Sundt Unit 4. Sierra Club stated that overall, EPA has done an excellent job in its FIP. However, Sierra Club also asserted that substituting coal with natural gas is not the ultimate solution. The fuel substitution will address the pollution problem associated with coal combustion, but Sierra Club argued that TEP should transition toward renewable energy sources, and be a leader in developing solar, wind, and other renewable sources for the purpose of energy generation.

TEP noted that a fuel change to natural gas meets the RHR's requirements for alternative measures in lieu of BART in that it will achieve greater reasonable progress than the implementation of BART. TEP added that because emissions under BART or the alternative would emanate from the same stack (and therefore the distribution of emissions is not significantly different), the alternative achieves greater reasonable progress simply because it will result in greater emissions reductions. In addition, TEP noted that EPA's finding that “natural gas provides better visibility improvement than the proposed BART determination” is consistent with the results of modeling performed by a contractor (AECOM) for TEP. Several other commenters (ADEQ, ANGA, Earthjustice, NPS, TPMEC, Friends of Saguaro National Park and a private individual) expressed general support for the better-than-BART alternative.

Response:

We acknowledge the commenters' support of the proposed BART alternative. Today's final rule provides TEP with the option to comply either with the BART limits within three years of publication of the final rule or with the requirements of the BART alternative by December 31, 2017. With regard to the comments concerning renewable energy, we note that the BART Guidelines indicate that “[w]e do not consider BART as a requirement to redesign the source when considering available control alternatives.”

We therefore consider a requirement for TEP to transition to renewable energy to be beyond the scope of what the RHR requires.

BART Guidelines, Section IV.D.1.5.

Comment:

ACCCE said that the BART alternative should be rejected because it does not lead to an improvement in humanly perceptible visibility. According to ACCCE, EPA stated that switching from coal to natural gas under the better-than-BART alternative will lead to a higher visibility improvement than the combination of SNCR and DSI together. Yet, with one exception, the areas affected by Sundt Unit 4 will not see a greater than 1.0 dv improvement. Again, ACCCE made the case that it is up to the states to make BART-eligibility determinations, but if it is determined that EPA has correctly classified Sundt Unit 4 as BART-eligible, it is Arizona, not EPA, that must finalize a BART determination for the unit. However, if this does not occur, ACCCE reiterated that it disagrees with EPA's analysis to require BART, since it does not result in humanly perceptible visibility improvement.

Response:

As explained in response to similar comments on our BART analyses above, visibility improvement is not required to be humanly perceptible in order for a control to be required as BART. Arizona did not include a BART analysis and determination for TEP Sundt 4 in any of its RH SIP submittals. If Arizona submits such a determination in the future, we will give it due consideration under the requirements of the CAA and EPA's implementing regulations.

Comment:

TEP stated that the facility has been co-firing landfill gas in the Sundt Unit 4 boiler since 1999, and that this has been an integral part of the company's strategy for complying with Arizona's Renewable Energy Standard and Tariff, as it is among the most cost-effective renewable resources in its portfolio. TEP added that, through the direct displacement of heat input otherwise provided by coal, co-firing landfill gas has resulted in significant avoided emissions of carbon dioxide, SO

, PM, and other pollutants. TEP asserted that it must be allowed to continue an environmentally beneficial program.

TEP further stated that its current tariff agreement with El Paso Natural Gas Company for natural gas deliveries to Sundt Unit 4 does not meet the fuel-sulfur specification in the definition of “pipeline natural gas” in 40 CFR 72.2, but the tariff agreement does meet the sulfur specifications in the definition of “natural gas” in 40 CFR 72.2. TEP indicated that it has no direct control over the sulfur content of the natural gas delivered to Sundt, and limiting the fuel burned at Sundt Unit 4 to “pipeline

natural gas” would prohibit TEP's ability to select the alternative to BART, which TEP and many other stakeholders view as the preferred choice. Accordingly, TEP recommended several revisions to the regulatory language for the better-than-BART alternative that would revise the SO

emission limit and fuel restriction to correspond to the definition of “natural gas” rather than “pipeline natural gas” and provide for co-firing of landfill gas. TEP noted that regardless of the SO

emission limit that EPA selects for the alternative to BART, or the method identified to demonstrate compliance with that limit, SO

emissions from Sundt Unit 4 under the alternative to BART will be orders of magnitude lower than SO

emissions would be through the application of BART.

Response:

We agree that the continued co-firing of landfill gas does not adversely affect whether the fuel switch to natural gas achieves greater emissions reductions than the aggregate BART determinations for Sundt Unit 4. We are therefore revising the regulatory language to provide for the co-firing of landfill gas. In addition, we are revising the SO

emission limit in the better-than-BART alternative (and the emissions value used to evaluate whether the alternative is better-than-BART) to correspond to the definition of “natural gas” per 40 CFR 72.2. These revised emission calculations are contained in our docket, and are summarized in our response to the following comment.

See spreadsheet titled “Revised BART Alternative Emission Calculations.xls.” Specifically, the SO

emission factor for natural gas was revised from 0.00064 lb/MMBtu to 0.057 lb/MMBtu.

Comment:

TEP stated that stack testing to demonstrate compliance with the PM

limit while burning natural gas is unnecessary. According to TEP, the PM

emission limit of 0.010 lb/MMBtu that EPA proposed under the alternative to BART was developed based on a calculation using an AP-42 emission factor, but the proposal requires a compliance demonstration by conducting performance stack testing using EPA Method 201A and Method 202, per 40 CFR part 51, Appendix M. TEP stated that stack testing is a suitable method of determining compliance with an emission limit when either (1) it is necessary to verify that required controls are in place and operating correctly, or (2) to verify that a source is designed and constructed (in the case of a new unit) to meet a particular performance standard. However, according to TEP, neither of those situations applies to implementation of the alternative to BART on Sundt Unit 4, which is essentially a fuel-use limitation. TEP indicated that, while it has no reason to conclude that Sundt Unit 4 could not meet the standard, it has no experience measuring PM

emission levels while burning natural gas. Thus, the inclusion of Method 202 for condensable PM

presents some risk. TEP encouraged EPA to modify the compliance demonstration requirement for PM

to a calculation using AP-42 (as EPA did to set the standard), combined with a demonstration that natural gas is the primary fuel.

Response:

We partially agree with this comment. The BART alternative PM

emission limit in the proposed rule (0.01 lb/MMBtu) is based on AP-42 emissions factors for natural gas usage. This factor is based on information that might not represent the emission characteristics of Sundt Unit 4 (i.e., a coal-burning unit that is converted to natural gas). We do not agree, however, that it is appropriate to eliminate entirely the performance test requirement, but recognize that there is a lack of experience and history regarding condensable PM

test results at the Unit. As a result, we are revising the PM

compliance determination to a “test and set” approach. An initial performance test for PM

, based on the results of Method 202 plus either Method 5 or Method 201A, is still required along with subsequent performance tests if requested by the Regional Administrator. The results of the initial performance test will establish the PM

limit with which subsequent performance tests must demonstrate compliance. For purposes of evaluating the better-than-BART alternative, our estimate of PM

emissions is based upon this 0.30 lb/ton PM

BART limit. Although this results in PM

emissions equivalent to BART, the natural gas fuel switch still results in a net decrease in both NO

and SO

relative to the respective BART determinations. As a result, this approach does not alter our determination that the natural gas fuel switch is better-than-BART. A comparison of emissions between the BART determination and the revised better-than-BART alternative is summarized in Table 5.

Table 5—Comparison of BART Determination to Better Than BART Alternative

Parameters

Units

BART determination

BART alternative

(natural gas fuel switch)

Emission

reduction

(tpy)

Heat Duty

MMBtu/hour

1,371

1,820

Capacity Factor

Percentage

0.49

0.37

Control Technology

SNCR+LNB+OFA

LNB+OFA

lb/MMBtu

0.31

0.25

TPY

912

737

175

Control Technology

Dry Sorbent Injection

None

lb/MMBtu

0.18

0.057

TPY

530

169

361

Control Technology

Fabric Filter

None

lb/MMBtu

0.03

TPY

6. Other Comments on Sundt Unit 4

Comment:

TEP stated that it generally supports EPA's BART determinations for Sundt Unit 4 because the control technologies selected as BART are available and technically feasible for the control of the respective pollutants. Furthermore, while TEP asserts that the level of visibility improvement achieved by application of these technologies is marginal, they conclude that the identified controls can be installed and operated at Sundt Unit 4 without a significant impact on reliability or customer rates.

Response:

We acknowledge TEP's support.

Comment:

TEP agreed with EPA's selection of 2011 as the baseline year for Sundt Unit 4's emissions and operating characteristics. In contrast, Earthjustice stated that EPA's BART analyses are flawed due to errors in EPA's emissions baseline and baseline capacity factor. Earthjustice noted that EPA considered Sundt Unit 4's historical emissions from 2008 to 2012, and selected 2011 as the baseline because Sundt Unit 4 predominantly burned coal that year. However, according to Earthjustice, Sundt Unit 4 also burned large amounts of coal in 2008, making it unclear why EPA did not use 2008 instead of, or in addition to, 2011 when determining the baseline (e.g., by creating a baseline averaging 2008 and 2011 emissions).

Response:

We disagree with Earthjustice's comment. In 2008, Sundt Unit 4 operated at a much higher capacity factor than in subsequent years. As discussed in a response to a previous comment, we do not consider the higher capacity factors observed during the pre-2009 period to be a realistic depiction of anticipated annual emissions. As a result, we do not consider it appropriate to incorporate 2008 annual emissions into the development of baseline emissions.

Comment:

Earthjustice stated that EPA should set a one-year compliance deadline to install BART controls, rather than the proposed three-year deadline. Earthjustice noted that the CAA requires sources to install BART controls as “expeditiously as practicable,” and judicial opinions interpreting similar compliance deadlines in the CAA read this language to require compliance as soon as possible. According to Earthjustice, EPA set a three-year compliance deadline to install both DSI and SNCR based on EPA's conclusion that it will take three years to install DSI. The commenter asserted that DSI can be installed in just one year based on the record established for the MATS rulemaking and the rulemaking docket for this action. Earthjustice also noted that EPA has recognized that typical SNCR retrofits take ten to 13 months. Earthjustice stated that it is not aware of any circumstances at Sundt that would require additional time to install DSI and SNCR. Accordingly, the commenter suggested that because the CAA requires BART to be installed as quickly as possible and the record shows that both DSI and SNCR can be installed in one year, EPA should set a one-year compliance deadline for both controls.

Response:

We disagree with this comment. Although we agree that either control technology can be installed in as little as one year, we do not consider it reasonable to require installation of both technologies, in parallel, within a single year. The CAA and the RHR require compliance with the BART emission limit as expeditiously as possible, but in no event later than five years after promulgation of the FIP.

The three-year time frame in our proposed rule is consistent with this requirement.

CAA section 169A(g)(4), 42 U.S.C. 7491(g)(4), 40 CFR 51.308(e)(1)(iv).

Comment:

A private citizen indicated support for the proposal to end coal burning at the Sundt facility by the end of 2017 and requested that Sundt implement the requirement sooner. Specifically, the commenter recommended that TEP, the owner of the Sundt facility, use up the existing supply of coal and not purchase any additional coal. TPMEC similarly asked that TEP use up the coal it has on site and not buy any more, but proceed with the conversion. In contrast, TEP stressed that the timing of the elimination of coal is an integral part of the alternative to BART and should not be adjusted. TEP stated that because EPA may not consider a fuel switch as a control option for determining BART for a source (citing section IV.D.1.5 of the BART Guidelines), the decision whether to implement the alternative to BART is at the sole discretion of TEP. TEP added that because (1) the alternative was originally developed by TEP and (2) it clearly meets the requirements for “better than BART,” EPA is limited in its ability to make changes to certain aspects of TEP's approach.

TEP asserted that it will need until December 31, 2017, to burn the existing fuel on site, ensure an adequate natural gas supply, and make the operational and mechanical changes necessary to achieve the proposed NO

emission rate. According to TEP, since the alternative to BART results in lower emissions on an annual basis, the timing for implementation is inconsequential relative to the long-term visibility goals of the RHR and should remain as originally outlined by TEP. TEP added that EPA has no obligation or authority to arbitrarily make a better-than-BART alternative even better by adjusting the timing for implementation, and therefore the timing for implementation of the alternative should not be adjusted.

Response:

We have considered TEP's request to revise the compliance deadline to December 31, 2017. We agree with TEP that this deadline is reasonable, given that the alternative results in greater emission reductions than BART on a lb/MMBtu basis for NO

, SO

, and PM and meets the other requirements for a better-than-BART alternative under 40 CFR 51.308(e)(2) and (3). Therefore, we are setting a compliance deadline of December 31, 2017.

Comment:

TEP asserted that EPA underestimates the costs of controlling NO

and SO

emissions from Sundt Unit 4. TEP indicated that it hired a professional engineering and construction firm, Burns and MacDonnell (BMD), to review the cost estimates developed by EPA as part of its five-factor BART analysis and to provide new cost estimates for the installation and operation of various control technologies on Sundt Unit 4. The results of BMD's analysis are in Table 6. TEP further noted that the BART Guidelines provide for incorporation of site-specific factors or “elements . . . that are not addressed by the Cost Control Manual,” and stated that the most significant site-specific factors for Sundt Unit 4 have been identified by BMD in the report attached to the comments. TEP asserted that these factors should be incorporated into the final BART determination for the facility.

Table 6—Comparison of EPA's and BMD's Bart Analysis Results

[All values are in $/ton of pollutant removed]

Control technology

EPA

(proposed)

TEP

Difference (percent)

Control Technology

Selective Non-Catalytic Reduction

$3,222

$3,637

Selective Catalytic Reduction

5,176

7,874

Control Technology

Dry Sorbent Injection

1,857

3,088

Dry Flue Gas Desulfurization

5,090

9,359

Wet Flue Gas Desulfurization

5,505

8,229

Response:

As noted in our proposed rule and TSD, we revised upwards our contractor's original control cost estimates based on certain site-specific factors noted by TEP in its letter dated May 10, 2013. We incorporated many, but not all, of the factors raised in that letter. In its comment letter on our proposed rule, TEP raised additional factors and asserted that the cost estimates for each of the control options is underestimated. In the case of SCR, dry FGD, and wet FGD, we stated in our proposed rule that we consider these control options to not be cost-effective, either in general or in relation to their anticipated visibility benefits. In the case of SNCR and DSI, even if we were to accept all of TEP's revisions included in the comment letter, we would still consider these options to be cost-effective generally and to be BART based on our consideration of costs and visibility benefits.

Comment:

NPS commented that that although EPA has not stated the reasonable level of cost-effectiveness, it assumes that the Agency typically uses $5,000/ton and 0.5 deciviews (dv) as thresholds. Yet, NPS has seen higher cost-effectiveness thresholds from EPA and other states. While NPS commends EPA for its presentation of cumulative visibility impacts and cumulative visibility benefits of reducing emissions, NPS also requested that EPA work with NPS to develop a consistent and transparent method to relate cost to visibility improvement.

Response:

As noted in responses to other comments, we have not established specific thresholds for the cost and visibility factors for BART. NPS is therefore correct to note that BART determinations made by EPA may not precisely align along a specific set of $/ton or deciview improvement values. Further, even where the costs of compliance and expected degree of visibility improvement are similar at two different sources, consideration of other statutory factors may result in different outcomes.

With regard to determinations made by state agencies, we note that the RHR provides states with significant discretion in considering and weighing the five BART factors, so long as the factors are appropriately evaluated and the state's determination is supported by reasoned explanations for adopting the technology-based limits selected as BART. As a result, while a direct comparison of $/ton and deciview improvement values associated with BART determinations from multiple state agencies and EPA is informative and should carry weight in the ultimate decision, such comparisons are not outcome determinative.

We also note that it is unusual for controls at two different sources to have similar visibility benefits across all affected Class I areas.

Comment:

NPS indicated that it has collected and reviewed close to 100 BART determinations for EGUs and has found that the average cost per deciview for NO

reductions at EGUs is $14 million and the maximum cost per deciview is $34 million based on the Class I area with highest visibility improvement. NPS asserted that the $14 million figure is a good indication of the value states have placed upon reducing NO

for visibility purposes.

Response:

We agree with NPS that cost per deciview improvement is informative as a cost-effectiveness metric, including comparing the effect of controls on sources located in different parts of the country. We provided calculations of this metric in our proposal for this action. However, consistent with the BART Guidelines,

we have relied more heavily on cost-effectiveness calculated as cost per pollutant ton reduced and related visibility improvements in deciviews (both at individual areas and as a cumulative sum over all affected areas) as opposed to the cost per deciview metric.

See e.g. 70 FR 39167 (“For purposes of air pollutant analysis, `effectiveness' is measured in terms of tons of pollutant emissions removed, and `cost' is measured in terms of annualized control costs.”)

Comment:

NPS expressed support for EPA's inclusion of the cumulative visibility impacts and improvements associated with the control scenarios that were considered, noting that the EGUs evaluated are unusual because they impact from ten to 15 Class I areas within 300 kilometers (km).

Response:

We agree with NPS that it is important to account for visibility impacts at multiple Class I areas, given that the goal of the visibility program is to remedy visibility impairment at all Class I areas.

The cumulative sum, while not the only means of analyzing benefits across multiple Class I areas, is an easily understood and objective method of weighing cumulative visibility improvement, and is useful as part of the overall BART determination.

CAA section 169A(a)(1).

Comment:

TEP stated that EPA should adopt version 6.42 of CALPUFF as the approved regulatory version for modeling regional haze, since this version corrects deficiencies in the chemistry and the dispersion functions of CALPUFF version 5.8. TEP indicated that several studies conducted over the last few years demonstrate that the deficiencies in version 5.8 result in over-estimation of the visibility impacts of NO

emissions in Class I areas. This causes erroneous over-estimation of the visibility improvements from proposed BART controls leading to biased cost-benefit values.

Response:

We disagree with TEP for two reasons. First, CALPUFF 5.8 is approved as a regulatory model for use by EPA in regional haze determinations. CALPUFF version 5.8 has been thoroughly tested and evaluated, and has been shown to perform consistently with the initial 2003 version in the analytical situations for which CALPUFF has been approved. CALPUFF 6.42 is not an approved regulatory model because CALPUFF 6.42 has not yet undergone adequate review. We relied on version 5.8 of

CALPUFF because it is the EPA-approved version in accordance with the Guideline on Air Quality Models (“GAQM”, 40 CFR part 51, Appendix W, section 6.2.1.e). We updated the specific version to be used for regulatory purposes on June 29, 2007, including minor revisions as of that date. Second, EPA took into account limitations with Version 5.8 when it suggested use of the 98th percentile day versus the maximum day.

Memorandum in docket, “Full Technical Response to Modeling Comments for June 2014 Final Arizona Regional Haze FIP (Phase III),” Colleen McKaughan and Scott Bohning, EPA, June 16, 2014.

Comment:

TEP commented that the background ammonia concentration used in visibility modeling is critical because ammonia is a precursor to particulate ammonium nitrate. EPA's use of 1.0 parts per billion (ppb) for ammonia background concentration for all months of the year will tend to overestimate the visibility benefits associated with reductions of NO

, particularly in the winter months. TEP noted that monthly ammonia measurement data from the IMPROVE monitoring network site in southern Arizona (Chiricahua) indicate that ammonia concentrations below 1.0 ppb (e.g., 0.5 ppb) are present at this site during the winter months. TEP asserted that use of those values will more accurately predict the visibility improvements expected from the reductions in NO

emissions. Although TEP did not perform any new modeling for comparison to EPA's results in the proposal, TEP sent a letter to EPA in May 2013 that provided clarification regarding certain modeling parameters and the results of modeling performed by TEP's contractor (AECOM). According to TEP, the modeling performed by AECOM included a BART control scenario involving SNCR and DSI, similar to EPA's proposed BART determination for Sundt Unit 4. The results of AECOM's modeling was a maximum visibility improvement of 0.16 dv at Saguaro National Park East compared to the baseline case. The TEP noted that EPA's modeling representing the same control configuration (SNCR and DSI) reported a maximum visibility improvement of 0.49 dv. TEP acknowledged that these differences in modeling results have little practical effect, as EPA has proposed that its results support a BART determination involving application of SNCR and DSI on Sundt Unit 4, and TEP does not dispute that overall finding. However, should EPA find a need to do additional modeling to support its final BART determination for Sundt Unit 4, TEP recommended that EPA incorporate the modeling improvements suggested in TEP's letter of May 10, 2013.

Response:

We disagree that the 1.0 ppb ammonia background we assumed for CALPUFF modeling is too high. It is consistent with EPA guidance given that some ammonia measurements are higher than 1.0 ppb, and the available ammonia data is variable over the areas included in the visibility modeling. The uncertainty over appropriate ammonia values leaves us without a reasonable basis for choosing a different constant value, or a more complex monthly varying scheme as recommended by the commenter. Ambient ammonia measurements for use as input to modeling are scarce, and measurements that include it in the form of ammonium still scarcer. In the absence of compelling ammonia background estimates, the EPA Interagency Work Group on Air Quality Modeling (IWAQM) Phase 2 guidance recommends the use of a 1.0 ppb ammonia background for arid lands, which includes Arizona.

This is the only guidance available on this issue. It is worth noting that there are measurements of gaseous ammonia (NH

) that by themselves are close to or greater than 1.0 ppb, even in winter.

Therefore, we consider the 1.0 ppb ammonia background that we used to be appropriate for this action. Finally, we agree with the commenter that the recommended modeling changes would have little practical effect on the BART determination for Sundt Unit 4.

Interagency Work Group on Air Quality Modeling (IWAQM) Phase 2 Summary Report And Recommendations For Modeling Long Range Transport Impacts (EPA-454/R-98-019), EPA OAQPS, December 1998,

http://www.epa.gov/scram001/7thconf/calpuff/phase2.pdf.

Memorandum in docket, “Full Technical Response to Modeling Comments for June 2014 Final Arizona Regional Haze FIP (Phase III),” Colleen McKaughan and Scott Bohning, EPA, June 16, 2014.

B. Nelson Lime Plant Kilns 1 and 2

1. Subject to BART Determination

Comment:

ADEQ asserted that EPA improperly disapproved ADEQ's finding that Nelson Lime Plant is not subject to BART. ADEQ argued that ADEQ's use of a three-year average 98th percentile value “appropriately recognizes the highly variable visibility conditions that prevail in western states due to periodic wildfires that can result in short-term spikes in visibility impairment” and is consistent with how EPA determines compliance with certain NAAQS.

Response:

These comments largely pertain to EPA's partial disapproval of the Arizona RH SIP and are therefore untimely, as EPA has already taken final action on the SIP.

To the extent that the comments dispute EPA's proposed determination that the Nelson Lime Plant is subject to BART under the FIP, we disagree with the substance of their argument. The BART Guidelines recommend use of the 98th percentile modeled visibility impact across multiple years of modeling in order to identify sources that cause or contribute to visibility impairment in a Class I area.

There are at least three different ways to determine the 98th percentile impact across three years of modeling: The maximum 8th high in any one year, the 22nd high impact over all three years, or the three-year average of the 8th high impacts from each year. Of these three methods, the three-year average is the least conservative way of determining the 98th percentile impact. Depending on the yearly distribution of the results, the most conservative 98th percentile impact may come from the maximum 8th highest value for each of the three years or the 22nd highest value for all years merged. While the BART Guidelines do not specify which value to use, given that the subject-to-BART determination is a screening test, EPA's position is that a more conservative approach, i.e., the 22nd high of three merged years or the maximum 8th high of any one year, is more appropriate for this screening test. The FLMs also recommend a more conservative approach and have noted that other states have used such an approach.

78 FR 46142.

40 CFR part 51, appendix Y, section III.A.3.

Federal Land Managers' Air Quality Related Values Work Group (FLAG) Phase I Report—Revised (2010) (FLAG 2010) at 23; National Park Service Comments on EPA Review of Arizona Department of Environmental Quality (ADEQ) Determinations of Best Available Retrofit Technology (BART) at 2-3, and Reasonable Progress (RP) March 6, 2013.

We also do not agree with ADEQ that a three-year average approach “appropriately recognizes the highly variable visibility conditions that prevail in western states due to periodic wildfires that can result in short-term spikes in visibility impairment.” The visibility impacts of individual sources, including the Nelson Lime Plant, are determined by calculating the change in deciviews caused by the source compared to natural visibility conditions.

While natural conditions could include short-term spikes from wildfires, the effect of such a spike in the background level of pollution is to decrease the relative deciview impact of

a given source.

Thus, the possibility of short-term spikes from wildfires would, if anything, argue for a more conservative approach to evaluate an individual source's contribution. Moreover, we do not agree that the use of a three-year average is appropriate here simply because certain NAAQS use a three-year averaging period. Thus, consistent with the FLMs' recommendation and with the approach used by EPA and other states for making subject-to-BART determinations, we find that use of the 98th percentile impact of any one year is appropriate for making subject-to-BART determinations for purposes of this action.

40 CFR part 51, appendix Y, section III.A.3.

See 70 FR at 39124 (“as a Class I area becomes more polluted, any individual source's contribution to changes in impairment becomes geometrically less”).

With regard to the modeling performed for the Nelson Lime Plant, ADEQ's comments refer to three different modeling analyses: (1) The initial modeling performed by LNA; (2) the refined modeling analysis performed by LNA using the revised IMPROVE equation; and (3) an additional analysis referred to by LNA in its comments on the Phase 2 proposal. ADEQ included the results of the first two analyses in the Arizona RH SIP. Both sets of results showed that for a single year, 2003, the Nelson plant's 8th high visibility impact exceeded 0.5 dv.

Under EPA's interpretation of the 0.5 dv threshold, this makes the facility subject to BART. The complete results of the third analysis performed by LNA were not submitted to EPA.

However, more recent modeling performed by LNA shows that the 98th percentile impact of the facility exceeds 0.5 dv in each of the three years modeled.

Thus, even under the three-year averaging approach preferred by the State, the Nelson Lime Plant is subject to BART, according to the most recent modeling performed by the facility's owner. As explained above, under EPA's interpretation of the 0.5 dv threshold, the Nelson Lime Plant is subject to BART based on prior modeling. Therefore, for the reasons set out in our Phase 2 proposed and final rulemakings and in this response, we are finalizing our determination that the Nelson Lime Plant is subject to BART.

Arizona Regional Haze SIP at 152-53, Table 10.9 and Table 10.10.

See 78 FR at 46154.

BART Five Factor Analysis, Lhoist North America Nelson Lime Plant; Prepared by Trinity Consultants in conjunction with Lhoist North America of Arizona, Inc. (Public version dated September 27, 2013), Table 4-7. As explained in our proposal, these results are conservative (i.e., tending to overestimate rather than underestimate the impacts), but appropriate for purposes of a subject-to-BART determination.

2. BART Analysis and Determination for NO

Comment:

NPS indicated that it agrees with EPA that visibility improvements expected as a result of applying SNCR support this technology as BART for NO

Response:

We agree with NPS, and acknowledge its support on this issue.

Comment:

ADEQ asserted that the three-year compliance time provided in the rule does not provide enough time to retrofit SNCR on Kilns 1 and 2 because of the difficulty of installing such controls. In contrast, Earthjustice argued that EPA should set a one-year compliance deadline for the installation of SNCR at the plant. According to Earthjustice, EPA recognized in the proposal that SNCR can be installed in one year, but speculated without any support that it might take longer at the Nelson Lime Plant because of a “lack of information regarding SNCR installation schedules on lime kilns.” The commenter stated that allowing an extra two years without any supporting record violates the CAA's requirement that BART be installed as expeditiously as practicable.

Response:

We disagree with ADEQ's assertion that a three-year compliance schedule is too short and with Earthjustice's assertion that it is too long. ADEQ has not provided any support for its assertion that three years is an insufficient period of time for installation, nor has the facility's owner made such an assertion. Regarding Earthjustice's contention that a shorter deadline is required, we note that the examples cited are for SNCR installations on cement kilns. There are multiple operational and design differences between cement and lime production.

Cement and lime production processes are sufficiently different that it is not appropriate to assume that SNCR installation times for cement kilns are directly transferable to the application of SNCR on lime kilns. To our knowledge, SNCR has never been installed on a lime kiln. Given that this control technology will be retrofitted to a new source category for the first time, it is not unreasonable to expect unforeseen challenges and delays. EPA's timeline is conservative and takes into account this possibility. Therefore, we find that a requirement to install SNCR within three years is consistent with the provisions of the CAA and the RHR requiring compliance with BART emission limits as expeditiously as practicable.

“Comments on Draft NO

Control Measure Summary for Lime Kilns”, National Lime Association, March 30, 2006; AP-42, Section 11.6, Portland Cement Manufacturing; AP-42, Section 11.17, Lime Manufacturing.

Comment:

Earthjustice agreed that SNCR is a technically feasible control technology at the Nelson Lime Plant, but disagreed that the control efficiency for SNCR should be limited to 50 percent. Earthjustice stated that EPA's analysis must include the most stringent emissions reductions possible with SNCR (citing the BART Guidelines), and asserted that SNCRs can achieve control efficiencies significantly higher than 50 percent for the reasons discussed by Earthjustice in relation to the Clarkdale and Rillito cement plants. Earthjustice added that higher NO

reductions are especially appropriate at Nelson Lime Plant given the facility's high baseline NO

emissions. Earthjustice also noted that EPA provided no support in the record for the CEMS emissions data used in the development of the NO

emissions baseline.

Response:

We disagree with this comment. The information provided by Earthjustice consists of examples of SNCR on cement kilns. There are substantial differences between cement kilns and lime kilns that do not allow for direct comparisons of technical feasibility or control effectiveness. As noted previously, neither we nor the commenter were able to identify an instance of a lime kiln operating with SNCR in the United States. In addition, Earthjustice has not provided any information supporting an SNCR control efficiency more stringent than 50 percent on a lime kiln.

LNA has provided a summary of CEMS emission data, but considers it CBI since it also includes lime production data. We have included a summary of the lb/ton values from the testing period in our docket for the final rule because the BART limit is established in terms of lb/ton.

We have not included the mass emission rates from the testing period, since including both the lb/hour and lb/ton data in the docket would allow for the back-calculation of the lime production data.

Non CBI—Summary of LNA Nelson March, May and June 2013 CEMS Testing.xlsx.

3. BART Analysis and Determination for SO

Comment:

Earthjustice disagreed with EPA's rejection of DSI technology based on cost considerations, and with EPA's BART reduction approach that relies on a change in fuels. Earthjustice disagreed with what it considers EPA's uncritical agreement with the company (i.e., DSI at 40 percent reduction) and asserted that, given the almost 4,000 tpy of SO

emitted from the two kilns, EPA's determination of the most stringent control efficiencies achievable should have been more thorough and technically grounded. Earthjustice asserted that a DSI can be optimized and can achieve far greater than 40 percent reduction, as the company's own tests show (i.e., short-term efficiencies ranging from 17 to 84 percent). Earthjustice also asserted that even with what it considers a flawed analysis, the calculated cost-effectiveness value of about $4,000/ton reduced is well within acceptable ranges. As a result, Earthjustice disagreed with the weight that EPA gave to the incremental cost-effectiveness values and urged EPA to reconsider its SO

BART determination for the Nelson Lime Plant in the final rule.

By contrast, NPS said that it supports EPA's conclusion, noting that it is most important to reduce process emissions before adding expensive emissions controls. NPS indicated support for EPA's decision because it generally favors moving toward cleaner fuels. After changing the fuel at the plant, however, NPS noted that it may be appropriate to revisit requiring emissions controls at that time.

Response:

We acknowledge NPS's support on this issue. We disagree with Earthjustice that a more stringent DSI control efficiency is appropriate. Although the commenter notes that site-specific test data suggest short-term control efficiencies as high as 84 percent, there is no evidence that the upper range of short-term control efficiencies is sustainable over longer periods. As a result, when calculating annual emissions reductions in tpy, which is performed on an annual average basis, we do not consider it appropriate to use a control efficiency achieved over a short-term period because it might not achievable over a long-term annual average. Although Earthjustice asserted that the determination of a DSI control efficiency in our proposed rule should be more thorough and technically grounded, it has not provided any information regarding how, specifically, we should revise our analysis or that supports a more stringent control efficiency.

Furthermore, as explained in more detail in a response to a comment from LNA below, the total cost figures in our proposed rule inadvertently omitted annual indirect costs. Correcting this error results in approximate average and incremental cost-effectiveness values of $4,800/ton and $10,200/ton for Kiln 1 and $4,500/ton and $9,500/ton for Kiln 2.

The largest incremental visibility benefit of DSI relative to the visibility benefit of the proposed fuel mixture change at a single Class I area is 0.11 dv at Grand Canyon National Park.

We do not consider this level of incremental cost to be warranted by the incremental visibility benefit of DSI relative to the fuel mixture change. However, additional controls for the Nelson Lime Plant, such as DSI, should be considered for purposes of ensuring reasonable progress in future planning periods.

“LNA Nelson Control Costs (revised for Final Rule).xlsx.”

See 79 FR 9341, Table 26.

Comment:

LNA determined that compliance with the SO

emission limits within six months after the effective date of the final rule in the

Federal Register

, likely in July 2014, is not feasible. Therefore, the proposed six-month compliance window is unreasonable. Compliance with the SO

emission limits is based on a two-step process: (1) Use of a CEMS to determine actual SO

emissions from each kiln; and (2) use of daily production tonnage. LNA estimated that an 18-month period is a more reasonable compliance timeframe for a system that supports both NO

and SO

CEMS as well as new weigh scales on lime storage silo transfer belts.

Response:

We agree that a six-month time period is an insufficient amount of time for the design, installation, and optimization of an SO

CEMS in this case. In other cases in which compliance with a BART limit does not involve construction of add-on controls, but does involve installation of CEMS, we have provided a twelve-month window for compliance.

In this case, taking into account that multiple CEMS (NO

and SO

) will need to be installed, and the fact that the facility does not currently operate with CEMS, may not have existing systems or infrastructure in place, and is replacing lime weigh scales, we consider an 18-month compliance time frame to be as expeditious as practicable. Therefore, we are revising the compliance deadline for SO

at Nelson Lime Plant to eighteen months from the effective date for the final rule in the

Federal Register

77 FR 72578. The Cholla Power Plant SO

BART limit required installation of inlet CEMS, with a twelve-month compliance deadline.

Comment:

LNA stated that in its BART analysis submitted to EPA, the fuel mixture control option was based upon a maximum of 6.5 percent ash content in the proposed fuel mixture. LNA asserted that it did not choose this value arbitrarily, but based the value on operational knowledge and on information provided by the manufacturer of the kilns, Kennedy Van Saun (KVS).

Response:

As noted in the proposed rule, we used a fuel mixture consistent with a maximum 6.5 percent ash content in the SO

BART analysis. We have not received any other comments regarding this issue, and the final SO

limits finalized in today's rule reflect this maximum ash content.

Comment:

LNA asserted that EPA's estimate of the costs for DSI is unrealistic because EPA did not use site-specific input values. In addition, LNA said that there are errors in EPA's cost calculations. LNA noted various issues with EPA's cost analysis for DSI and asserted that the value of $4,200/ton of SO

removed is too low.

Response:

We agree that our cost calculations contain an error in the “cost summary” tab, which is also reflected in the TSD and in the

Federal Register

preamble to our proposed rule. The total annual cost for DSI should represent the sum of annual direct and annual indirect costs, but did not include the annual indirect cost. We corrected this error in a new version of the spreadsheet for today's final rule.

As a result, the average cost-effectiveness values for DSI on kilns 1 and 2 increase to about $4,800/ton and $4,500/ton (from $4,174/ton and $4,085/ton, respectively). The incremental cost-effectiveness values of DSI, relative to the fuel mixture change, are about $10,200/ton and $9,500/ton (from $8,803/ton and $8,576/ton, respectively). As noted in the proposed rule, we did not consider DSI to be cost-effective on an incremental basis relative to the fuel mixture change, given the relatively small visibility benefits expected from DSI (0.10 dv at the most improved class I area and 0.29 dv cumulative). Therefore, we do not consider DSI to be cost-effective, relative to the fuel mixture change, based on these revised and even higher dollar/ton values.

“LNA Nelson Control Costs (revised for Final Rule).xlsx”.

LNA provided EPA with a detailed version of DSI cost calculations that was designated as CBI along with a public version with most of the calculations redacted. Because we are generally prohibited from disclosing CBI, we relied on the publicly available information to develop a separate set of calculations for the proposed rule. While there are several elements of our cost estimates that differ from LNA's CBI-protected cost calculations, these differences are immaterial in light of our finding that DSI is not a cost-effective control option relative to the fuel

mixture change. Therefore, we have not further revised our cost analysis for DSI based on LNA's comments because the changes suggested by LNA would not alter our determination that DSI is not cost-effective for either kiln on an incremental basis.

4. BART Analysis and Determination for PM

Comment:

ADEQ expressed support for EPA's determination that the existing baghouse at the Nelson Lime Plant is BART for PM

10.

Response:

We acknowledge ADEQ's support on this issue.

5. Other Comments

Comment:

LNA asserted that EPA's BART proposal does not provide for differing emission rates during startup, shutdown, and malfunction (SSM), and stated that EPA should reconsider this decision that is not supported by the available information. The CEMS data for NO

and SO

that LNA submitted in its BART analysis is based on periods of steady-state operation that does not include periods of startup and shutdown. Since the CEMS data do not include these emissions, LNA did not consider it appropriate for the proposed limits to include emissions from startup and shutdown. LNA proposed that the rolling 30-day limits in the proposed rule should apply only during periods of normal operation, and proposed establishing separate emission limits during periods of startup and shutdown. LNA provided emissions data for each of the various types of startup and shutdown events.

Response:

We agree that the emission limits in the proposed rule did not account for emissions from periods of startup and shutdown and we agree that the emission limits should include such periods. Because Section 302(k) of the CAA requires emission limits such as BART to be continuous,

BART emission limits must apply at all times, including during periods of startup, shutdown, and malfunction. We therefore consider it appropriate to revise the proposed emission limits for NO

and SO

to account for emissions from periods of startup and shutdown. In order to revise the emission limits to appropriately account for startup and shutdown emissions, we sought additional information from LNA following the close of the public comment period.

In response, LNA suggested retaining the rolling 30-day limits that would apply at all times, but revising them upward to accommodate startup and shutdown emissions.

Following further discussions between EPA and LNA,

LNA proposed revising the rolling 30-day limit to an annual average limit that would apply at all times.

LNA also proposed establishing short-term ton/day limits for the Kilns, which would correspond to the short-term 24-hour average emission rates used in the visibility modeling.

42 U.S.C. 7602(k).

Phone call between Colleen McKaughan, EPA, and Ed Barry, LNA, on April 10, 2014.

Letter from Ed Barry, LNA, to Colleen McKaughan, EPA (April 29, 2014).

Conference calls between EPA and LNA, May 2 and 7, 2014.

Letter from Ed Barry, LNA, to Colleen McKaughan, EPA (May 9, 2014).

Id.

Based on our evaluation of the additional information provided by LNA, we are making the following revisions to the proposed emission limits. First, we are revising the lb/ton limits from a rolling 30-day basis to a rolling 12-month basis. As described in LNA's comments, periods of startup can exhibit substantial emissions, but with little to no lime production. While these startup emissions are not higher than those observed during normal operation on a simple mass basis (e.g., lb/hour, or ton/day), the fact that there is no production associated with these emissions complicates their inclusion when determining compliance with a lb/ton limit. As a result, the particular day(s) during which a startup event occurs will appear as a short-term spike in the kiln's emission rate (lb/ton). When combined with the preceding 29 days of emission data, this emission spike has the effect of driving the rolling 30-day emission rate (lb/ton) upwards. It may then be necessary for the unit to operate at a much lower rate of emissions over the next 29 days in order to ensure compliance with the 30-day limit, which may not be technically feasible. By establishing the limit on a rolling 12-month basis, such short-term spikes are averaged with data values from over an entire year, making its impact on the rolling emission rate less pronounced.

Second, in order to ensure that performance of the SNCR system installed at the Nelson Lime Plant is optimized, we are including in the final rule a series of control technology demonstration requirements. In particular, LNA is required to prepare and submit to EPA: (1) A design report describing the design of the ammonia injection system to be installed as part of the SNCR system; (2) data collected during a baseline period; (3) an optimization protocol; (4) data collected during an optimization period; (5) an optimization report establishing optimized operating parameters; and (6) a demonstration report including data collected during a demonstration period. While this type of control technology demonstration is not typically required as part of a regional haze plan, we consider it to be appropriate here, given the minimal data available about the performance of SNCR at lime kilns. Based upon the data collected during this process, EPA may revise the rolling 12-month average for the NO

emission limit in a future notice-and-comment rulemaking action.

Third, we are establishing short-term 24-hour average emission limits (ton/day) consistent with the emission rate used in the visibility modeling for each respective control option. As noted above, revising the averaging period to an annual basis minimizes the effect of short-term spikes in emissions over a greater data set. In effect, this allows the Nelson Plant greater short-term emissions variability while still demonstrating compliance with the BART limit. To ensure that this variability does not interfere with the modeled visibility benefit, which is based upon reductions from the highest 24-hour average emission rate, we are establishing short-term ton/day emission limits. These limits are combined limits that apply across both Kiln 1 and 2, on a rolling 30-kiln operating day basis. We are finalizing a combined Kilns 1 and 2 NO

limit of 3.20 tons/day and SO

limit of 10.10 tons/day.

C. Comments on the Hayden Smelter

1. General Comments

Comment:

ASARCO agreed with the BART Guidelines “that BART is not `to redesign the source,' ” and stated this understanding is inherent in Congress' denomination of the technology as “best available retrofit technology.”

Response:

We agree that BART does not require redesign of the source.

Comment:

ASARCO noted that the BART Guidelines are not “mandatory” as applied to the Hayden Smelter, and that EPA must depart from them if presented with sound technical justification.

Response:

We agree that the BART Guidelines are not binding with respect to the Hayden Smelter, but note that the BART Guidelines serve as persuasive guidance for all BART sources.

Comment:

ASARCO stated that, as further changes to air pollution controls at the Hayden Smelter will be required to demonstrate attainment of the 1-hour SO

NAAQS, ASARCO supports EPA's proposal to promulgate “a performance standard as BART rather than

prescribing a particular method of control,” if EPA determines additional controls are needed. ASARCO stated that reconfiguration of the smelter might be required to attain the 1-hour SO

NAAQS in the form of a “converter retrofit project” or CRP. ASARCO argued that while detailed engineering of the CRP is substantially completed, details must be worked out before the final project can be permitted. Thus, ASARCO concluded that it is critical for EPA not to finalize a BART FIP for SO

that interferes with the Hayden area's attainment of the SO

NAAQS. Similarly, ADEQ urged EPA to reevaluate its SO

BART decision for the Hayden Smelter and align it with controls that ASARCO has to implement in order to comply with the 1-hour SO

NAAQS.

Response:

Following the close of the public comment period, ASARCO provided us with additional information concerning the CRP, including a description of plans to replace the BART-eligible Peirce-Smith converters with new converters. If the BART-eligible converters are replaced prior to the BART compliance deadline, then the BART requirements would no longer apply. Accordingly, there is no basis to expect that the RH FIP will interfere with ASARCO's ability to ensure attainment of the SO

NAAQS. We also agree that a performance standard rather than a particular method of control is appropriate for BART. As explained further below and in a revised BART determination included in the docket for this final rule, ASARCO has demonstrated that separate levels of control are necessary for the primary and secondary capture systems. Therefore, we are setting the level of control to 99.8 percent (equivalent to the existing double contact acid plant) for the primary capture system and 98.5 percent for the secondary capture system. These limits only apply if ASARCO does not replace the BART-eligible converters prior to the BART compliance deadline.

2. BART Analysis and Determination for SO

From Converters

Comment:

ADEQ said that EPA's disapproval of ADEQ's SO

BART determination for the Miami and Hayden Smelters is unsupported. Similarly, AMA requested that EPA reconsider its decision to disapprove parts of the Arizona RH SIP because the State should make a BART determination for the smelters according to the CAA.

Response:

These comments concern EPA's partial disapproval of the Arizona RH SIP and are therefore untimely, as EPA has already taken final action on the SIP.

The commenters have provided no legal basis for EPA to reconsider that action.

78 FR 46142.

Comment:

NPS expressed support for EPA's decisions based on the expected substantial visibility improvements associated with installing a new acid plant as BART for SO

at the Hayden Smelter. In particular, NPS agreed with EPA's decisions to protect many Class I areas.

Response:

We appreciate NPS's support and note that the BART level of control for the converters is a performance standard and not any particular method of control.

Comment:

ASARCO, ADEQ, and AMA expressed doubt over the technical feasibility of a double contact acid plant for controlling secondary ventilation gases. ASARCO asserted that acid plants are not an “applicable” technology, and therefore, not an “available” technology for controlling secondary ventilation gases because of low concentrations of SO

and high variability in the exhaust gas stream. ASARCO stated that EPA failed to evaluate the technical feasibility of double contact acid plants when applied to these low-strength gases, which is the second step of a BART analysis. ASARCO argued that, had EPA conducted an adequate analysis, it would have concluded that double contact acid plants are not an “applicable” technology because they do not have a “practical potential for application” to the secondary ventilation gases and hence are not an “available” technology. ADEQ and AMA echoed ASARCO's comments, urging EPA to look at the information submitted by ASARCO and reconsider its proposal.

Response:

We do not agree that a double contact acid plant is technically infeasible for the secondary gas stream at the Hayden Smelter. As explained in the BART Guidelines, control technologies are technically feasible if either (1) they have been installed and operated successfully for the type of source under review under similar conditions, or (2) the technology could be applied to the source under review.

The BART Guidelines further explain that the regulatory authority must exercise technical judgment in determining whether a control alternative is applicable to the source type under consideration. In most cases, a commercially available control option is presumed applicable if it has been used on the same or a similar source type. Absent a showing of this type, one must evaluate technical feasibility by examining the physical and chemical characteristics of the pollutant-bearing gas stream, and comparing them to the gas stream characteristics of the source types to which the technology had been applied previously.

In this instance, a double contact acid plant is already in use at the Hayden Smelter. Therefore, it is presumed to be an applicable technology, absent a demonstration that specific circumstances preclude its application to a particular emission unit. Generally, such a demonstration involves an evaluation of the characteristics of the pollutant-bearing gas stream and the capabilities of the technology.

No such demonstration of technical infeasibility has been made here. On the contrary, the record establishes that a double contact acid plant is feasible for the secondary gas stream at the Hayden Smelter.

40 CFR part 51, appendix Y, section IV.D.2.

Id.

See Id.

In particular, while the secondary gas stream has a lower SO

concentration and higher volumetric flow rate than the primary gas stream, these differences do not render a double contact acid plant technically infeasible. Indeed, EPA concluded more than 30 years ago that “[i]t is technically feasible . . . to design acid plants that will operate auto-thermally on feed streams that exhibit SO

concentrations below the 3.5 to 4.0 percent range.”

The commenters have offered no evidence to refute this conclusion. Contrary to the commenters' suggestions, ASARCO's contractors, Gas Cleaning Technologies (GCT) and MECS,

have not stated that use of a double contact acid plant is technically infeasible.

Rather, they have indicated that use of this technology would present additional technical challenges that would make it more costly and less effective than estimated by EPA. In particular, GCT states that “[a] more realistic 60 ppmv [parts per million by volume] outlet concentration would mean only 96 [percent] SO

removal efficiency by such an acid plant at ASARCO. . . . when a realistic capital cost and removal efficiency is used for the acid plant, the $/ton SO

removed estimate will be more than double the $872/ton

indicated by EPA.”

However, as explained in the BART Guidelines, where the resolution of technical difficulties is merely a matter of increased cost, you should consider the technology to be technically feasible.

100

Therefore, in this instance, EPA considers a double contact acid plant to be a technically feasible option for control of the secondary gas stream. ASARCO's assertions regarding cost-effectiveness are addressed below.

1984 NSPS Review at 4-3.

This is the name of the company.

See Letter from Steven Puricelli, MECS, to Matt Russell, GCT (March 5, 2014)(“MECS Letter”) (“A double acid plant could operate with this low secondary gas concentration . . .”); Letter from Matt Russell, GCT, to Jack Garrity, ASARCO (“GCT Letter”)(February 12, 2014) at 2 (“it may be technically feasible to operate an acid plant on the converter secondary gases . . .”).

GCT Letter at 2 (“A more realistic 60 ppmv outlet concentration would mean only 96% SO

removal efficiency by such an acid plant at ASARCO . . . when a realistic capital cost and removal efficiency is used for the acid plant, the $/ton SO

removed estimate will be more than double the $872/ton SO

indicated by EPA.”).

100

40 CFR part 51, appendix Y, section IV.D.2.

Comment:

ASARCO stated that there are deficiencies in EPA's cost analysis for an acid plant. First, ASARCO asserted that EPA cannot rely upon the cost formula from the 1984 NSPS Review for an acid plant without validating current costs and, as a result, has substantially underestimated the cost of the proposed acid plant for the secondary ventilation gases. ASARCO stated that the equation that EPA used was derived from double-contact acid plants that were processing primary ventilation gases with significantly higher SO

concentration (4.5 percent to 8.0 percent) and flow rates up to 140,000 standard cubic feet per minute (scfm). This compared to rates for secondary ventilation gases at 0 to1 percent SO

and 275,000 scfm.

101

ASARCO stated that EPA's extrapolation to lower concentrations cannot be justified because none of the data points included double-contact acid plants treating secondary ventilation gases, for which MECS gave a significantly higher cost estimate.

101

The original comment referred to a “0-0.1” percent concentration for secondary ventilation gases. ASARCO Comment Letter at 9. However, this appears to be an error, as the same letter also states that “[a]t the Hayden Smelter, the SO

content of secondary ventilation gas ranges from 0 to 1 [percent] SO

or approximately 0 to 10,000 ppm, and averages 1580 ppm.” ASARCO Commenter Letter at 5.

Second, ASARCO stated that supplemental heating of the acid plant influent gas is required, but there is no supplemental heat available to reduce heat load requirements as suggested by E

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.