Amicus Curiae Brief — Util. Air Regulatory Grp. v. Envtl. Prot. Agency, 135 S. Ct. 702 (2014) (No. 14-47)

Supreme Cour, US.

FILED

MAR § - 2015

OFFICE OF THE CLERK

Nos. 14-46, -47, -49

In the Supreme Court of the United States

MICHIGAN, ET AL., PETITIONERS,

.

ENVIRONMENTAL PROTECTION AGENCY, ET AL.,

UTILITY AIR REGULATORY GROUP, PETITIONER,

Vv.

ENVIRONMENTAL PROTECTION AGENCY, ET AL.,

NATIONAL MINING ASSOCIATION, PETITIONER,

Vv.

ENVIRONMENTAL PROTECTION AGENCY, ET AL.,

On Writs of Certiorari To The United States Court

of Appeals For The District of Columbia

BRIEF OF AMICI CURIAE EXPERTS IN AIR

POLLUTION CONTROL AND AIR QUALITY

REGULATION IN SUPPORT OF RESPONDENTS

ELIZABETH J. HUBERTZ

Counsel of Record

INTERDISCIPLINARY ENVIRONMENTAL CLINIC

WASHINGTON UNIVERSITY SCHOOL OF LAW

ONE BROOKINGS DRIVE, CAMPUS Box 1120

ST. LOUIS, MISSOURI 63130

314.935.8760 * ejhubertz@wulaw.wustl.edu

i

TABLE OF CONTENTS

Page

TABLES OF CORTE Beth 2 O vcscceveseosessecsecsasacessnaesincemaasa i

TABLES OF AUTEROIE 8 BIE cesecssesecssisxesacnvessgregnane iv

INTEREST OF THE AMICI CURIAE ....................... 1

SUMMARY OF THE ARGUMENT .....................::::ee 4

PIS IBEN 1 o000s00scsecsevecensosexconsctasacasaensieamener nae 5

I. IN 2000, EPA REASONABLY FOUND IT

APPROPRIATE TO REGULATE EGU HAP

EMISSIONS WITHOUT FORMALLY

CONSIDERING COSTS BECAUSE CONTROL

TECHNOLOGIES FOR REDUCING SUCH

EMISSIONS WERE ALREADY AVAILABLE AND

WIDELY USED, AND THE COSTS OF THOSE

TECHNOLOGIES WERE NOT PROHIBITIVE. ...... 6

A. The HAPs Emitted By EGUs Include Mercury,

Non-Mercury Metals, Acid Gases, And Organic

HAPs; Mercury Has Been Identified As The

HAP Of Greatest Potential Concern For Public

B. Technologies Were Readily Available For

Controlling Mercury And Other HAPs Emitted

By Ue Bae DGD, cnsscnvecsdsscsssciaccee eae 9

i

1. Controls For Non-Mercury Metallic HAPs

And Particle-Bound Mercury ................... 10

2. Controls For Acid Gas HAPs And Ionized

EEE Ee 11

3. Controls For Elemental Mercury............. 14

C. Technologies For Controlling Mercury And

Other HAPs Emitted By EGUs Were Already

SIE EE II, ci dincccedecussvcacconssesccsccsectes 15

1. Installed Base Of Non-Mercury Metallic

HAP And Particle-Bound Mercury

i ac csemnnceanvabavesseoes 16

2. Installed Base Of Acid Gas HAP And

Ionized Mercury Controls......................... 17

3. Installed Base Of Elemental Mercury

i nina ncanencnunncdisucvsenecse 18

D. Because Control Technologies For Reducing

Mercury And Other HAPs Emitted By EGUs

Were Available And Widely Used, And The

Costs Of Those Technologies Were Not

Prohibitive, EPA’s 2000 Finding Was

slain tiadiminapncaestecsesscsecseeeoes 19

1. ESPs And Fabric Filters .......................... 20

2. Flue Gas Desulfurization (Wet/Dry

tae ES Ee 21

il

3. ACI And Other Sorbent Injection

I vic scnscicsanteniteinieibtnisneditediancatdidentedasdomtandiptens 23

Il. EPA REASONABLY AFFIRMED ITS 2000

FINDING THAT REG ULATION WAS

APPROPRIATE BECAUSE COST-EFFECTIVE

EMISSION CONTROL TECHNOLOGIES FOR

MERCURY AND OTHER HAPS WERE IN WIDE

CIEE BS SU: kinscccscteseninssncesitnsmiibinnaninnmanittonsnnaannion 26

A. Between 2000 And 2012, Many EGUs

Installed Mercury Emission Control

Equipment In Response To State

SID: sesisencccnlbsicniiianibiannddcumaiidantdagnaian 27

B. Between 2000 And 2012, Many EGUs

Installed Control Technologies That Will

Reduce Mercury And Other HAP Emissions In

Response To Federal Regulation Of Other

Pollutants And Were Able To Meet The MATS

Requirements Before They Took Effect. ....... 33

ae RIN stirsnerinsctemnseceentetesuaanianéannamiaiananealeinnd 35

iV

TABLE OF AUTHORITIES

Page

STATUTES

PS Oe eT Oe: APNEA 4,8

RROD nee Pore NCS 29

i oda ee 20

abe 8

OTHER AUTHORITIES

AMERICAN LUNG ASSOCIATION, EMISSIONS OF

HAZARDOUS AIR POLLUTANTS FROM COAL-FIRED

POWER PLANTS (Mar. 7, 2011). .................c0ee0ee0es 12, 13

B&W, Southern Co. announce DOE-Based Projects to

Reduce Mercury, INSIDE ENERGY (Dec. 25, 2000),

available at 2000 WL 2108218........................cceeeeees 23

Curtis Carlson, et al., Sulfur Dioxide Control by

Electric Utilities: What Are the Gains from Trade?,

Resources for the Future Discussion Paper ©3-44-

Be CE, Fi iircitiincccanasetcnsnnnsecusaisapanistsisvenginnenanses 22

CLEAN ENERGY GROUP, ENSURING A CLEAN, MODERN

ELECTRIC GENERATING FLEET WHILE MAINTAINING

ELECTRIC SYSTEM RELIABILITY: SUMMER 2011 UPDATE

ID HO sececanss kectenenencabaensacpcudsncumaeinenimecios 28, 29, 32

Vv

DEPARTMENT OF ENERGY/NATIONAL ENERGY

TECHNOLOGY LABORATORY, PHASE II MERCURY

CONTROL TECHNOLOGY FIELD TESTING PROGRAM:

UPDATED ECONOMIC ANALYSIS OF ACTIVATED CARBON

I SEE ra LOD 18-19, 31

ELECTRIC POWER RESEARCH INSTITUTE, FGD

OPTIMIZATION WORKBOOK, available at

http://www.epri.com/abstracts/Pages/ProductAbstract

MARTE TOGUCCIS MT He 1111.26 .........0000cccccccccceccveeceeses 21

ENVIRONMENTAL PROTECTION AGENCY, EPA/600/R-

99/056, COAL UTILITY ENVIRONMENTAL COST

(CUECOosT) WORKBOOK USER’S MANUAL (1999) ...... 21

ENVIRONMENTAL PROTECTION AGENCY, EPA/600/R-

00/093, CONTROLLING SO2 EMISSIONS: A REVIEW OF

TECHNOLOGIES (Nov. 2000)

ENVIRONMENTAL PROTECTION AGENCY, CONTROL OF

MERCURY EMISSIONS FROM COAL-FIRED ELECTRIC

UTILITY BOILERS (Feb. 26, 2004) ........20000..0.000cccceeen 24

ENVIRONMENTAL PROTECTION AGENCY, EPA Sets

National Air Quality Standards (Apr. 30, 1971)

avatlable at http://www?2.epa.gov/aboutepa/epa-sets-

national-air-quality-standards...........................0000 16

ENVIRONMENTAL PROTECTION AGENCY, Mercury

Study Report to Congress (Dec. 1997).......... 21, 22, 31

vl

ENVIRONMENTAL PROTECTION AGENCY, National

Electric Energy Data System (NEEDS) Database

v.3.02, http://www.epa.gov/airmarkets/documents/

ipm/NEEDSV3.02_EISA.xls (Mar. 2002) .... 16, 18, 20

ENVIRONMENTAL PROTECTION AGENCY, National

Electric Energy Data System (NEEDS) Database

v.4.10, http://www.epa.gov/airmarkets/documents/

ipm/NEEDSv410.zip (Apr. 2010)....................... 30, 31

ENVIRONMENTAL PROTECTION AGENCY, Particulate

Matter (PM) Standards — Table of Historical PM

NAAQS, http://www.epa.gov/ttn/naaqs/standards/pm/

Sk LIE SSeS career Pees rer me 16

ENVIRONMENTAL PROTECTION AGENCY, EPA-600/R-

03-110, PERFORMANCE AND COST OF MERCURY AND

MULTIPOLLUTANT EMISSION CONTROL TECHNOLOGY

APPLICATIONS ON ELECTRIC UTILITY BOILERS

ERE RSE eY DIRT are aca eet vara Way 14

ENVIRONMENTAL PROTECTION AGENCY, EPA-600/R-

00-083, PERFORMANCE AND COST OF MERCURY

EMISSION CONTROL TECHNOLOGY APPLICATIONS ON

ELECTRIC UTILITY BOILERS (Sep. 2000).................... 24

ENVIRONMENTAL PROTECTION AGENCY, EPA430-R-99-

011, PROGRESS REPORT ON THE EPA AcID RAIN

rr rr, a ceideueueenaansinnmalennens 18

ENVIRONMENTAL PROTECTION AGENCY, Study of

Hazardous Air Pollutant Emissions from Electric

Utility Steam Generating Units -- Final Report to

IIIS i ccicddimisctagdendecnienchenstumnation 8, 9, 20

vil

ENVIRONMENTAL PROTECTION AGENCY, Sulfur Dioxide

(SOz) Primary Standards — Table of Historical SOz

NAAQS, http://www.epa.gov/ttn/naaqs/standards/so2/

Oe MII is cuicidusdceaicipuminaneictbiniendeniasauaanomiaiates 17

FirstEnergy to Demonstrate Multi-Pollutant Control

System, INDUSTRIAL ENVIRONMENT (July 1, 2000),

available at 2000 WL 9960640........00.00-.....c eee eee 23

LARRY GRAY, REVIEW OF CONTROL TECHNOLOGIES FOR

MERCURY EMISSIONS FROM COAL-FIRED POWER

I I a as a 15

JAMES E. MCCARTHY, CONG. RESEARCH SERV.,

RL33535, MERCURY EMISSIONS FROM ELECTRIC

POWER PLANTS: STATES ARE SETTING STRICTER LIMITS

Memorandum, Mercury Costs Calculations:

Assumptions, Approach, and Results (Sep. 2000),

available at http://www.epa.gov/ttn/atw/combust/

an IAS IE GN Re EPID L RES Pps 25

NATIONAL ASSOCIATION OF CLEAN AIR AGENCIES,

COMMENTS ON Docket ID No. EPA—HQ-—OAR-—2009—

I I i a a a 32

National Emission Standards for Hazardous Air

Pollutants From Coal and Oil-Fired Electric Utility

Steam Generating Units and Standards of

Performance for Fossil-Fuel-Fired Electric Utility,

Industrial-Commercial-Institutional, and Small

Industrial—-Commercial-Institutional Steam

Vill

Generating Units, 77 Fed. Reg. 9.304

a, Gi aaa 6, 27

NORTHEAST STATES FOR COORDINATED AIR USE

MANAGEMENT, COMMENTS ON DOCKET ID No. EPA-—

HQ-OAR~2009-0234 (AUG. 2, 2011) ..............00..0060. 32

NORTHEAST STATES FOR COORDINATED AIR USE

MANAGEMENT, CONTROL TECHNOLOGIES TO REDUCE

CONVENTIONAL AND HAZARDOUS AIR POLLUTANTS

FROM COAL-FIRED POWER PLANTS

I i i i ar passim

NORTHEAST STATES FOR COORDINATED AIR USE

MANAGEMENT, ENVIRONMENTAL REGULATION AND

TECHNOLOGY INNOVATION: CONTROLLING MERCURY

EMISSIONS FROM COAL-FIRED BOILERS

I a aoe ae 22-23, 24

NORTHEAST STATES FOR COORDINATED AIR USE

MANAGEMENT, MERCURY EMISSIONS FROM COAL-

FIRED POWER PLANTS (BOUD)....0.ccccscescsccsccevcccccccessce0ss 31

Regulatory Finding on the Emissions of Hazardous

Air Pollutants from Electric Utility Steam

Generating Units, 65 Fed. Reg. 79,825

SI, BI IIE cbccrccnci cukulesicosaansoncsdsbumueccabuineeinaann passim

Revision of December 2000 Regulatory Finding on

the Emissions of Hazardous Air Pollutants From

Electric Utility Steam Generating Units and the

Removal of Coal- and Oil-Fired Electric Utility Steam

Generating Units From the Section 112(c) List, 70

Fed. Reg. 15,994 (Mar. 29, 2005) ..................0cceeeeee 29

1X

Rulemaking To Amend Dates in _ Federal

Implementation Plans Addressing Interstate

Transport of Ozone and Fine Particulate Matter,

ca, @ weet fe |e | eo 27

DAVID G. SLOAT AND PAUL S. FARBER, PARTICULATE

CONTROL FOR INDUSTRIAL APPLICATIONS (Mar. 25,

SEPT nin sc civatacdsacepinasavauudcenueabcbddasaudsainnlanisadciicesadonen es 13

Standards of Performance for New and Existing

Stationary Sources: Electric Utility Steam

Generating Units, Final Rule, 70 Fed. Reg. 28,606

EB, SE cissesidisscissnnnsnntesqdaccontanhstanavatienebsabaciesunios 29

U.S. ENERGY INFO. ADMIN., DOE/EIA-0384(2011),

ANNUAL ENERGY REVIEW 2011 (Sep. 2012), available

at http://www.eia.gov/totalenergy/data/annual/pdf/

ID IIIs saeiinscebn sia oseuaukcwingrces sconunidcavlaidudanentanccseisaumaaeis 24

U.S. ENERGY INFO. ADMIN., Coal-fired power plant

operators consider emissions compliance strategies

(Mar. 28, 2014), http://www.eia.gov/

todayinenergy/detail.cfm?id=15611 ...................0..... 34

U.S. GOov’T ACCOUNTABILITY OFFICE, GAO-10-47,

MERCURY CONTROL TECHNOLOGIES AT COAL-FIRED

POWER PLANTS HAVE ACHIEVED SUBSTANTIAL

EMISSIONS REDUCTION (2009).................ccceceeeeees 30, 32

INTEREST OF THE AMICI CURIAE'!

Amici Curiae are Paul Miller, John Paul, Ranajit

Sahu, and Eric Svenson. The Amici wish to supply

the Court with information § regarding’ the

practicability of reducing hazardous air pollutant

(HAP) emissions from coal-fired electric utility steam

generating units (EGUs) as of 2000, the date of EPA’s

original finding that regulation was appropriate and

necessary. They further explain why in their view,

based on the availability of various. control

technologies for reducing HAP emissions — the costs

of which were reasonably well known and had been

proven to not be prohibitive — it was reasonable for

EPA not to formally consider costs in making this

finding. Amici also wish to provide the Court with an

understanding of the advancements in control

technologies from 2000 to 2012 and how these

advances support EPA’s affirmation of the

appropriate and necessary finding

Amicus Dr. Paul Miller is the Deputy Director

and Chief Scientist of Northeast States for

Coordinated Air Use Management (NESCAUM)

where he provides the organization with legal,

technical, and policy support for all NESCAUM

initiatives. He plays a leading role in supporting

! No counsel for any party had any role in authoring this

brief, and no persons other than the amici curiae and their

counsel made any monetary contribution to its preparation or

submission. Written consents from the parties to the filing of

this brief are on file with the Clerk.

2

state efforts to address acid deposition, mercury

emissions and other air and climate issues. Dr. Miller

has been a Senior Fellow at Princeton University’s

Center for Energy and Environmental Studies, and a

National Research Council Associate at the Joint

Institute for Laboratory Astrophysics, University of

Colorado, Boulder. He has a Ph.D. in Chemical

Physics from Yale University and a J.D. from

Stanford University.

Amicus Mr. John Paul recently retired from his

position as Administrator of the Regional Air

Pollution Control Agency (RAPCA) of Dayton, Ohio.

Mr. Paul worked for the agency from 1973 through

June of 2014 and served as its director from 1985. He

holds a Master of Science from Iowa State University

and a Bachelor of Science from Michigan State

University. Mr. Paul was active in air pollution

control issues in the State of Ohio and nationwide.

He served as co-chair of the National Association of

Clean Air Agencies’ New Source Review Committee,

as well as several terms as an officer and member of

the Board of Directors. He served two six-year terms

on EPA’s Clean Air Act Advisory Committee

(CAAAC) and co-chaired the CAAAC working group

on the utility MACT.

Amicus Dr. Ranajit Sahu is an environmental

engineer and consultant based in Southern

California. Dr. Sahu has more than 25 years of

experience in the fields of environmental, mechanical

and chemical engineering, including: design and

specification of environmental control equipment,

hazardous waste remediation, air pollution control

and equipment’ design, combustion process

engineering, energy studies, multimedia

environmental regulatory compliance, transportation

air quality impact analysis, multipathway health risk

assessments for toxics, air dispersion modeling,

regulatory strategy development, and design for

pollution prevention. He has provided numerous

industrial, government, commercial and _ public

interest group’ clients with comprehensive

multimedia compliance assistance encompassing all

media (air, water, solid and hazardous waste, mixed

waste, noise and community issues). He has Ph.D.

and M.S. degrees in Mechanical Engineering from

the California Institute of Technology and a B.Tech

degree from the Indian Institute of Technology.

Amicus Mr. Eric Svenson is a Senior Advisor with

M.J. Bradley & Associates (MJB&A) where he

focuses on strategic planning and analysis for the

electric and gas utility industry. He has over 39 years

of experience in many aspects of the industry

including: electric power plant operations,

engineering and construction, strategic planning, and

electric and gas transmission and distribution

systems. He also has significant expertise in state,

regional, and federal public policies pertaining to

economic, energy, and environmental regulation.

Prior to joining MJB&A, Eric was the Vice President

for Environment, Health and Safety for Public

Service Enterprise Group, a $16 billion market cap

Fortune 500 electric and gas utility headquartered in

Newark, NJ. He co-chaired EPA’s Greenhouse Gas

Best Available Control Technology (BACT)

committee that provided advice to EPA for its

4

development of BACT regulation and co-authored

with the Natural Resources Defense Council several

reports benchmarking electric power industry

emissions. He holds a Master of Engineering —

Mechanical degree from Stevens Institute of

Technology.

All Amici file this brief solely as individuals and

not on behalf of the institutions with which they are

affiliated.

SUMMARY OF THE ARGUMENT

Fossil fuel-fired power plants account for nearly

seventy percent of electricity generation in the

United States. Coal-fired power plants alone account

for nearly forty percent of total generation. Coal-fired

generation results in emissions of numerous air

pollutants, some of which are toxic and cause or may

cause cancer or other serious health effects. As coal is

burned in utility boilers, mercury and other heavy

metals are released, as are halogens, which form acid

gases such as hydrogen chloride and hydrogen

fluoride. These metals and gases are considered

HAPs under Section 112 of the Clean Air Act.

Based on studies EPA performed and other

information available to it at the time, EPA

concluded in 2000 that it was appropriate and

necessary to regulate HAP emissions from coal- and

oil-fired? EGUs. It based this finding in part on the

* Coal-fired EGUs emit by far the largest amount of utility-

originated HAPs. Oil-fired EGUs contributed to less than two

o

health hazards posed by HAPs emitted by EGUs,

particularly mercury, and on its determination that

when it eventually promulgated a regulation, EGUs

would be able to control their HAP emissions using

technologies that were, for the most part, already

available and already in use at many EGUs in 2000.

In 2012, after an additional decade of experience with

and study of the technological, economic and

operational feasibility of HAP controls, EPA affirmed

its 2000 finding. Both decisions were reasonable and

supported by the facts known to EPA and throughout

the industry.

ARGUMENT

In December 2000, after nearly a decade of study

of the health effects posed by EGU HAP emissions

and the control strategies available to reduce such

emissions, EPA determined that the regulation of

HAPs emitted by coal-fired EGUs was appropriate

and necessary. This decision was based in part on

EPA’s conclusion that technologies for controlling

HAP emissions that posed a threat to the public

health would be available for use (and in some cases

were already in use) once EPA promulgated its

reguiation. More than eleven years later, EPA

affirmed this decision, finding that the control of

EGU HAP emissions was achievable with existing

percent of the total HAP emissions for 1990 and 2010.

Regulatory Finding on the Emissions of Hazardous Air

Pollutants from Electric Utility Steam Generating Units, 65

Fed. Reg. 79,828 (Dec. 20, 2000). Therefore, this brief does not

discuss oil-fired EGUs and focuses solely on coal-fired EGUs.

6

technologies and without placing the reliability of

electric service at risk.

I. IN 2000, EPA REASONABLY FOUND IT

APPROPRIATE TO REGULATE EGU HAP

EMISSIONS WITHOUT FORMALLY

CONSIDERING COSTS BECAUSE CONTROL

TECHNOLOGIES FOR REDUCING SUCH

EMISSIONS WERE ALREADY AVAILABLE

AND WIDELY USED, AND THE COSTS OF

THOSE TECHNOLOGIES WERE NOT

PROHIBITIVE.

A. The HAPs Emitted By EGUs Include

Mercury, Non-Mercury Metals, Acid

Gases, And Organic HAPs; Mercury Has

Been Identified As The HAP Of Greatest

Potential Concern For Public Health.

EGUs emit a variety of HAPs including mercury,

non-mercury metals, acid gases, and organic HAPs.*

3 Organic HAPs include dioxins and furans. The significant

majority of data obtained by EPA for measured organic HAP

emissions from EGUs were below the detection levels of the

EPA test methods, leading EPA to conclude that it is

impracticable to measure organic HAP emissions from EGUs

and to propose work practice standards for organic HAPs

instead of the emission limitations it set for the other HAPs.

National Emission Standards for Hazardous Air Pollutants

From Coal and Oil-Fired Electric Utility Steam Generating

Units and Standards of Performance for Fossil-Fuel-Fired

Electric Utility, Industrial-Commercial-Institutional, and Small

Industrial-Commercial-Institutional Steam Generating Units,

7

Mercury and non-mercury metals (arsenic, cadmium,

chromium, lead, nickel, selenium, and others) are

naturally occurring elements found in coal in trace

amounts. Non-mercury metals generally do not

volatilize (convert into a gaseous state) when coal! is

burned in utility boilers and instead remain as solid

particles or bound to solid particles in the residual

ash. Mercury, by contrast, is highly volatile and

tends to vaporize and become entrained in the flue

gas (combustion exhaust) as either elemental or ionic

mercury vapor. Coal characteristics, e.g., rank,‘ and

combustion conditions dictate which form of vapor

phase mercury predominates. A small amount of

mercury, typically less than 10 percent, does not

volatilize when coal is burned in utility boilers and

remains in the residual ash as particle-bound

mercury.

Coal also contains halogens, highly reactive

elements in a group that includes fluorine, chlorine,

bromine, aud iodine. When coal is burned in utility

boilers the halogens are released and form strong

acids — primarily hydrogen chloride (HCl) and

77 Fed. Reg. 9,369 (Feb. 16, 2012). Therefore, organic HAPs will

not be further discussed in this brief.

‘ Rank is a measure of the degree of alteration that occurs as

buried organic matter undergoing coalification is subjected to

increasing temperature and pressure. Lower rank coals contain

less carbon and have a lower energy content than higher rank

coals. Lower rank coals also generally contain less chlorine,

which oxidizes mercury making it easier to capture in air

pollution control devices. The major ranks, from lowest to

highest, are lignite, subbituminous coal, bituminous coal, and

anthracite.

8

hydrogen fluoride (HF) — as the flue gas cools. The

concentration of halogens in coal varies by coal type,

although chlorine is usually the most abundant.

Coal-fired EGUs emit far more HC] than any other

HAP. EPA, Study of Hazardous Air Pollutant

Emissions from Electric Utility Steam Generating

Units -- Final Report to Congress, Vol. I 3-15 tbl.3-3

(Feb. 1998) (Utility RTC).

Based on the results of the Utility RTC and on

information obtained subsequent to that study,

EPA’s 2000 finding that regulation of EGU HAP

emissions was appropriate identified mercury from

coal-fired EGUs as the HAP of greatest concern for

public health. Regulatory Finding on the Emissions

of Hazardous Air Pollutants from Electric Utility

Steam Generating Units, 65 Fed. Reg. 79,825, 79,826

(Dec. 20, 2000) (2000 Finding). Pursuant to

§112(n)(1) of the Clean Air Act, EPA performed a

study in the late 1990s of the hazards to public

health reasonably anticipated to occur as a result of

EGU HAP emissions after imposition of the other

requirements of the Clean Air Act. EPA examined

HAP emissions test data from 52 coal-, oil-, and

natural gas-fired utility units for the study and

identified 67 of the more than 180 HAPs listed in

Section 112 of the Act as potentially emitted by

EGUs. Utility RTC at ES-2 to ES-4. EPA then

performed an assessment of inhalation and/or

multipathway exposure risks for a subset of priority

HAPs and concluded that mercury from coal-fired

EGUs is the HAP of greatest potential concern for

public health due to its high toxicity, its persistence

in the environment, and its tendency to

9

bioaccumulate in food chains. Jd. at 7-45. EPA also

concluded that arsenic and a few other metals are of

potential concern for carcinogenic effects, and that

dioxins and acid gas HAPs are of potential concern as

well. 2000 Finding at 79,827.

B. Technologies Were Readily Available For

Controlling Mercury And Other HAPs

Emitted By EGUs In 2000.

In 2000, when EPA made its finding that

regulation of EGU HAP emissions was appropriate,

EPA specifically identified a number of strategies for

controlling HAP emissions, including the use of pre-

combustion controls (fuel switching, coal switching,

coal cleaning, coal gasification), combustion controls

(boiler type, low NO, burners), post-combustion

controls (flue gas cleaning technologies), and

alternative controls (demand side management,

energy conservation). Utility RTC at 13-1 to 13-32.

EPA determined that the qualitative effects of these

strategies on EGU HAP emissions varied and could

not be predicted in some cases. Jd. at 13-33. However,

the ability of readily available post-combustion

controls to effectively capture most HAPs emitted by

EGUs was well understood. The available

technologies for controlling particulate matter had

the important co-benefit of capturing metallic HAPs

and particle-bound mercury, while the available

technologies for controlling sulfur dioxide had the

important co-benefit of capturing acid gases and the

ionic form of mercury. In addition, recent data

indicated that post-combustion control of elemental

10

mercury, perhaps the most difficult HAP to capture,

was also possible.

1. Controls For Non-Mercury Metallic

HAPs And Particle-Bound Mercury

Conventional controls for particulate matter (PM)

are highly effective for controlling non-mercury

metallic HAPs. These HAPs normally form or attach

to ash particles and are captured by standard PM

control devices, including electrostatic precipitators

(ESPs) and fabric filters. Mercury bound to ash

particles or other PM is also captured by these

devices.

ESPs use an electrical charge to remove particles

from flue gas under the influence of an electric field.

An ESP imparts a positive or negative charge to

incoming particles, then collects the particles on

oppositely charged plates or tubes. The collection

surfaces are rapped or vibrated periodically to

remove the accumulated particles, which are

collected in a hopper for disposal. ESP effectiveness

depends on the electrical resistivity of the particles

and on particle size. High resistivity particles

(produced by low-sulfur coal) are more difficult to

capture, as are smaller particles. Despite these

difficulties, ESPs can capture more than 99 percent

of total PM and 80 to 95 percent of PMos.°

NORTHEAST STATES FOR COORDINATED AIR USE

5 PMoes, also known as fine particulate matter, consists of

particles 2.5 micrometers in diameter or smaller.

11

MANAGEMENT, CONTROL TECHNOLOGIES TO REDUCE

CONVENTIONAL AND HAZARDOUS AIR POLLUTANTS

FROM COAL-FIRED POWER PLANTS 23 (Mar. 31, 2011)

(2011 NESCAUM Report).

Fabric filters trap and collect particles in flue gas

as it passes through the filter. The filters are made of

woven or felted material in the form of sheets,

cartridges, or bags, although bags are the most

common type. Gas passes freely through fabric

filters, but particles are retained and gradually build

up a cake on the fabric which is periodically removed

by one of a number of different cleaning mechanisms

and collected in a hopper for disposal. Fabric filters

are more efficient than ESPs and can capture up to

99.9 percent of total PM and 99 to 99.8 percent of

PMo2:5. Id. at 24.

ESPs and fabric filters generally capture greater

than 90 percent of all non-mercury metallic HAPs.

2000 Finding at 79,829. Capture rates for particle-

bound mercury are comparable to total PM capture

rates.

2. Controls For Acid Gas HAPs And

Ionized Mercury

Acid gas HAPs (HCl and HF) are effectively

captured by conventional controls for sulfur dioxide

(SOz), a highly reactive gas and criteria pollutant

emitted in large amounts by coal-fired EGUs.

Technologies for capturing SOe are called Flue Gas

12

Desulfurization (FGD) systems and include wet and

dry scrubbers and dry sorbent injection.§

Wet scrubbers inject an aqueous lime or limestone

slurry into the flue gas within a spray tower. SO2 and

acid gases in the flue gas, including HC) and HF, are

absorbed by and react with the alkaline slurry to

produce a wet solid residue, commonly called FGD

sludge, which is then collected for disposal or use as a

by-product. Lime is more reactive than limestone and

offers the potential for higher removal rates but is

also more expensive, so limestone is the most

commonly used reagent. Wet scrubbers typically

capture over 95 percent of SO2 and are sometimes

capable of removal rates in excess of 98 percent.

Removal rates for HCl are even higher. 2011

NESCAUM Report at 10. See also AMERICAN LUNG

ASSOCIATION, EMISSIONS OF HAZARDOUS’ AIR

POLLUTANTS FROM COAL-FIRED POWER PLANTS 32

(Mar. 7, 2011).

Dry scrubbers are similar to wet scrubbers in that

they inject an aqueous lime slurry into the flue gas to

react with acid gases. However, in dry scrubber

systems the slurry has aé_ée higher’ sorbent

concentration, and the water is evaporated by the

heat of the flue gas. As a result, a dry waste product

is formed instead of a wet sludge, which is then

captured in a standard PM control device. Dry

scrubbers are slightly less efficient than wet

¢ A sorbent is a material that collects molecules of another

substance by absorption (drawing molecules into its interstices)

or by adsorption (attracting molecules to its surface).

13

scrubbers but still typically capture more than 90

percent of SOz2. 2011 NESCAUM Report at 11. When

used in conjunction with fabric filters, dry scrubbers

typically capture about 95 percent of HCl. DAviD G.

SLOAT & PAUL S. FARBER, PARTICULATE CONTROL FOR

INDUSTRIAL APPLICATIONS 7 (Mar. 25, 2007).

Dry sorbent injection (DSI) is the pneumatic

injection of a powdered sorbent directly into the

boiler or the downstream ductwork to react with acid

gases in the flue gas. The dry reaction products are

then captured in a standard PM control device.

Capture efficiency depends on the sorbent and the

type of PM control device used. The mineral trona,

the most common sorbent in use, can capture 30-60

percent of SOz when injected upstream of an ESP

and up to 90 percent when injected upstream of a

fabric filter. 2011 NESCAUM Report at 13. EGUs

using DSI with a downstream ESP for particle

collection have demonstrated HCl] captures rates in

the 95-98 percent range. AMERICAN LUNG

ASSOCIATION, EMISSIONS OF HAZARDOUS AIR

POLLUTANTS FROM COAL-FIRED POWER PLANTS 33

(Mar. 7, 2011).

Technologies for SOz and acid gas HAP control

also capture ionic mercury. Ionic mercury vapor is

water-soluble and dissolves in and reacts with the

aqueous slurry used in wet and dry scrubbers. In wet

scrubbers it is collected in the FGD sludge; in dry

scrubbers it is captured in the downstream PM

control device. Removal rates are highly variable and

depend on a number of factors. Wet FGD systems

have exhibited capture rates in the 23 to 97 percent

14

range, while dry FGD systems have exhibited rates

in the 3 to 98 percent range. EPA, EPA-600/R-03-110,

PERFORMANCE AND COST OF MERCURY AND

MULTIPOLLUTANT EMISSION CONTROL TECHNOLOGY

APPLICATIONS ON ELECTRIC UTILITY BOILERS 17-18

(Oct. 2003).

3. Controls For Elemental Mercury

Elemental mercury is insoluble in water and has

poor reactivity with other species so it is not prone to

adsorption onto ash. Approaches for controlling it

include converting it into particle-bound mercury by

means of adsorption onto activated carbon, or

converting it into ionic mercury by adding an

oxidizing agent (e.g., chlorine or bromine) or through

the use of selective catalytic reduction (SCR)

technology.’

Conversion of elemental mercury into particle-

bound mercury is achieved using activated carbon

injection (ACI), which is the pneumatic injection of

powdered activated carbon directly into the flue gas.

Unlike ash, activated carbon particles effectively

adsorb elemental mercury due to their increased

surface area, converting it into particle-bound

mercury and allowing it to be captured in a standard

PM control device. Conversion of elemental mercury

into ionic mercury is achieved by adding a halogen to

7 SCR technology is used to control nitrogen oxides (NQO,).

Flue gas is passed through a catalyst where NO, reacts with the

catalyst and anhydrous ammonia, converting it to nitrogen and

water.

15

the coal or, if ACI is used, to the activated carbon

prior to injection. Halogens help oxidize? elemental

mercury, turning it into ionic mercury which is

captured in FGD systems. SCR technology also

oxidizes elemental mercury for capture in FGD

systems.

EPA was aware of the potential to capture

elemental mercury through the use of sorbents such

as activated carbon at the time of its 2000 finding.

2000 Finding at 79,829. ACI has since been proven

capable of capturing about 65 percent of elemental

mercury using untreated carbon, and in the 90

percent range using carbon treated with chemical

additives. LARRY GRAY, REVIEW OF CONTROL

TECHNOLOGIES FOR MERCURY EMISSIONS FROM COAL-

FIRED POWER PLANTS 8-9 (Oct. 24, 2013).

C. Technologies For Controlling Mercury

And Other HAPs Emitted By EGUs Were

Already Widely Used In 2000.

Technologies for controlling two of the three forms

of mercury and other HAPs emitted by EGUs were

not only available in 2000, when EPA made its

finding that regulation of EGU HAP emissions was

appropriate, they were already widely used. Several

of these technologies had already been in use for

decades by that time.

8 “QOxidize” in this situation refers to the removal of electrons

to make a positively charged mercury ion.

16

1. Installed Base Of Non-Mercury

Metallic HAP And Particle-Bound

Mercury Controls

EPA has regulated some forms of particulate

matter since at least 1971, when the first National

Ambient Air Quality Standards (NAAQS) were

promulgated. EPA, EPA Sets National Air Quality

Standards (Apr. 30, 1971), available at

http://www 2.epa.gov/aboutepa/epa-sets-national-air-

quality-standards. The original PM NAAQS set

limits for total PM. These were replaced in 1987 with

limits for PMio%, and separate limits for PM2s were

added in 1997. EPA, Particulate Matter (PM)

Standards — Table of Historical PM NAAQS,

http://www.epa.gov/ttn/naaqs/standards/pm/s_pm_his

tory.html (last visited Feb. 28, 2015).

For purposes of compliance with the NAAQS and

other Clean Air Act requirements, nearly every EGU

in the U.S. had already installed PM control devices

by 2000, when EPA made its finding. At that time,

300 out of 302 total gigawatts (GW) of coal-fired

capacity (99.2 percent) already had PM controls,

primarily ESPs (256 GW; 84.9 percent) and/or fabric

filters (51 GW; 16.9 percent). EPA, National Electric

Energy Data System (NEEDS) Database v.3.02,

http://www.epa.gov/airmarkets/documents/ipm/NEE

DSV3.02_ EISA.xls (last visited Mar. 2, 2015)

(NEEDS v.3.02). Because these concrol devices have

9 PMho consists of particles 10 micrometers in diameter or

smaller.

17

the co-benefit of capturing metallic HAPs and

particle-bound mercury, the installed base of metallic

HAP/particle-bound mercury controls was also over

99 percent in 2000.

2. Installed Base Of Acid Gas HAP And

Ionized Mercury Controls

Just as the PM NAAQS set limits for that

pollutant, the SOz2 NAAQS have set limits for SO2

since 1971. These limits did not change between 1971

and 2000, except for the revocation of annual and 3-

hour secondary standards in 1973 and 1996,

respectively..° EPA, Sulfur Dioxide (SO2) Primary

Standards — Table of Historical SOz NAAQS,

http://www.epa.gov/ttn/naaqs/standards/so2/s_so2_hi

story.html] (last visited Feb. 28, 2015). Additionally,

in 1990 Congress recognized that SO: pollution was

causing a serious problem that was not being

addressed through the NAAQS -— acid deposition, also

known as acid rain.!! The 1990 Clean Air Act

amendments directed EPA to establish a program to

control acid rain. EPA did so, implementing the Acid

10 Primary NAAQS provide public health protection.

Secondary NAAQS provide public welfare protection, e.¢.,

protection against decreased visibility and damage to animals,

crops, vegetation, and buildings.

11 The presence of acid gases in the air can cause atmospheric

water vapor to shift from a neutral pH to an acidic pH. When

the acidic water vapor condenses, it becomes acid rain, or acid

snow, and can damage trees and other vegetation as well as

cars and buildings. In some cases, the acid gases remain in gas

form or cause dust particles to become acidic. In this form, they

can be inhaled and cause health problems.

18

Rain Program in 1995 which regulated SO2 emissions

from EGUs through a cap-and-trade program. For

purposes of compliance with the NAAQS and other

Clean Air Act requirements and programs, including

the Acid Rain Program, some EGUs — primarily in

the Northeast and the Midwest — had already

installed scrubbers by 2000, when EPA made its

finding. EPA, EPA430-R-99-011, PROGRESS REPORT

ON THE EPA ACID RAIN PROGRAM (Nov. 1999). At

that time, 73 out of 302 total GW of coal-fired

capacity (24 percent) had already installed wet or dry

scrubbers. NEEDS v.3.02. Because these control

devices have the co-benefit of capturing acid gas

HAPs and ionized mercury, the installed base of acid

gas HAP/ionized mercury controls was also 24

percent in 2000.

3. Installed Base Of Elemental Mercury

Controls

Elemental mercury is the only HAP not captured

by controls for other pollutants. However, elemental

mercury can be captured using ACI, which converts

it to particle-bound mercury for capture in PM

control devices, or by adding halogens to the coal (or

to the activated carbon, if ACI is used) to convert it to

ionized mercury for capture in FGD systems. Neither

of these technologies was commercially available

when EPA made its regulatory finding in 2000, but

their theory of operation was well understood, they

had already undergone pilot-scale testing, and full-

scale testing on several coal-fired utility boilers was

approximately one year away. U.S. DEPARTMENT OF

ENERGY, DOE/NETL’S PHASE II MERCURY CONTROL

19

TECHNOLOGY FIELD TESTING PROGRAM: UPDATED

ECONOMIC ANALYSIS Or ACTIVATED CARBON

INJECTION 14 tbl.5 (May 2007) (showing initial

testing at Alabama Power's E.C. Gaston Unit 3 in

April 2001).

D. Because Control Technologies For

Reducing Mercury And Other HAPs

Emitted By EGUs Were Available And

Widely Used, And The Costs Of Those

Technologies Were Not Prohibitive, EPA’s

2000 Finding Was Reasonable.

By the end of 2000, it was apparent that almost

all HAPs, including two of the three species of

mercury, could be removed from EGU emissions

through the use of control devices that also removed

already-regulated criteria pollutants like PM and

SO. as described above. ESPs and fabric filters,

which capture PM, also capture particle-bound

mercury and other metallic HAPs that remain as

particles or bound to solid particles in the residual

ash after coal combustion. See supra Part I.B.

Additionally, scrubbers and DSI, which capture SOz,

also capture acid gas HAPs and ionized mercury from

flue gas. See supra Part L.B.

EPA was not blind to the practical implications of

controlling EGU HAP emissions when it made its

2000 finding that regulation was appropriate. In fact

EPA made its finding based in part on the

technological feasibility of controlling HAP

emissions, coupled with an understanding of the

general expense of mercury and other HAP controls.

20

2000 Finding at 79,828-30 (noting feasibility of

control and technologies that could “greatly reducfe]”

mercury control costs). The general cost of

installation and operation of these control devices

was well known throughout the electric power

generation industry and to EPA. However, the

ultimate cost of emission controls depended heavily

on the degree of removal efficiency that would

eventually be required by regulation after the

completion of the administrative rulemaking

process.'? At the finding stage, EPA’s analysis

demonstrated an understanding of the expense of

control devices relative to the overall costs of EGU

operation, as an inherent aspect of technological

feasibility.

1. ESPs And Fabric Filters

ESPs were the most well-established technology.

At the time of EPA’s finding, they had been used on

boilers for about 80 years. Utility RTC at 2-12.

Similarly, as of 2000, fabric filters had been used on

utility boilers for about two decades, although not on

as wide a scale as ESPs. Id. at 2-13. Of course, not all

installations were the same and they did not involve

the same capital investment or operation costs, but

nonetheless, roughly 99 percent of coal-fired EGU

capacity had installed ESPs, fabric filters or other

PM controls by the end of 1999. NEEDS v.3.02. These

control technologies were in place at a near-universal

12

In fact, Section 112(d)(2) of the Clean Air Act requires this

kind of inquiry at the standard-setting stage for all source

categories, not just EGUs.

21

level and were therefore providing a significant

amount of metallic HAP and particle-bound mercury

control prior to MATS at little additional cost to

operators.'!3 EPA, Mercury Study Report to Congress,

ES-14; Vol. VIII 5-20 (Dec. 1997) (Mercury RTC).

2. Flue Gas Desulfurization (Wet/Dry

Scrubbers)

By 2000, extensive research had been conducted

by EPA and others into the capital and operational

costs of various scrubber configurations that could be

used to remove acid gases and ionized mercury. See,

e.g., EPA, EPA/600/R-00/093, CONTROLLING S02

EMISSIONS: A REVIEW OF TECHNOLOGIES 43-84 (Nov.

2000); Id. at 84 tbl.6-12 (summarizing costs for

various configurations). Industry publications aimed

at EGU operators focused on reducing the costs of

scrubber operation by examining the factors that

could lower the costs of operation, such as age of

equipment, use of different coal types and equipment

selection. EPRI, FGD OPTIMIZATION WORKBOOK, app.

B at 16 tbl.B-8 (Aug. 22, 1998), available at

http://www.epri.com/abstracts/Pages/ProductAbstract

.aspx?ProductId=TR-111118. EPA also offered a tool

for estimating capital and operating costs. See EPA,

EPA/600/R-99/056, COAL UTILITY ENVIRONMENTAL

Cost (CUECOST) WORKBOOK USER’S MANUAL (1999).

Indeed, the basic facts about the conditions most

favorable for scrubber use were well known

13 Some older ESPs might require upgrades or add-on

equipment to enhance efficiency.

22

throughout the power-generation sector. For

example, it was known that wet scrubbers had a

higher efficiency, and that scrubbers generally were

more cost effective on higher capacity generating

units. EPA, EPA/600/R-00/093, CONTROLLING S02

EMISSIONS: A REVIEW OF TECHNOLOGIES 24, 56 (Nov.

2000). It was also well known that acid gas HAP and

mercury removal through the use of SQOz2 control

technology was a co-benefit and, to the extent that

such equipment was already installed, provided acid

gas HAP and ionized mercury control at little

additional cost to operators. Mercury RTC, Vol. VII,

5-20.

Moreover, it was well known that the marginal

cost of SOz abatement, including the cost of

scrubbers, had dropped significantly between 1990

and 2000, due in part to advances in control

technology. Curtis Carlson, et. al., Sulfur Dioxde

Control by Electric Utilities: What Are the Gains from

Trade?, Resources for the Future Discussion Paper

98-44-REV 34 (Apr. 2000) (the marginal abatement

cost of FGD due to technology improvements had

dropped by $50 per ton of SO2). Studies also indicated

that industry had significantly overestimated the

cost of SOz removal during the 1980s and early

1990s. NORTHEAST STATES FOR COORDINATED AIR USE

MANAGEMENT, ENVIRONMENTAL REGULATION AND

TECHNOLOGY INNOVATION: CONTROLLING MERCURY

EMISSIONS FROM COAL-FIRED BOILERS V-6 to V-7

23

(Sep. 2000) (2000 NESCAUM Report).'* At any rate,

between 1991 and 1995, 19 GW of U.S. generating

capacity installed scrubbers, bringing the installed

base to about 24 percent of total generating capacity

by 1999.15 See supra Part I.C.2.

2. ACI And Other Sorbent Injection

Systems

Although activated carbon injection and other

controls for elemental mercury were not yet

commercially available in 2000, their theory of

operation was well understood, bench- and pilot-scale

projects had been completed, and _ full-scale

demonstration projects were in the initial stages,

such as the one at FirstEnergy’s Eastlake facility

near Cleveland, Ohio, with more scheduled to follow,

such as the ones at Alabama Power Company’s

Gaston facility and at the Pleasant Prairie plant in

Wisconsin. FirstEnergy to Demonstrate Multi-

Pollutant Control System, INDUSTRIAL ENVIRONMENT

(July 1, 2000), available at 2000 WL 9960640

(demonstration of ECO technology to control

“nitrogen oxide (NO,), sulfur dioxide, fine particulate

matter, mercury and other substances”); B&W,

Southern Co. announce DOE-Based Projects to

‘4 See also 2000 NESCAUM Report at xiv (“early estimates

consistently overstate actual compliance costs, often by a factor

of two or more”).

‘6 This was lower than expected due to the significant drop in

the cost of low-sulfur coal. Burning low-sulfur coal proved a

cheaper e!ternative than installing FGD technology. See 2000

NESCAUM Report at [V-25 to IV-26.

24

Reduce Mercury, INSIDE ENERGY (Dec. 25, 2000),

available at 2000 WL 2108218; EPA, CONTROL OF

MERCURY EMISSIONS FROM COAL-FIRED ELECTRIC

UTILITY BOILERS 4-6 (Feb. 26, 2004).

Using the results of pilot and bench-scale studies

and cost-projection modeling, experts were able to

estimate the cost to EGUs of elemental mercury

removal. According to EPA, the capital costs of

activated carbon technology were relatively low.

EPA, EPA-600/R-00/083, PERFORMANCE AND CosT OF

MERCURY EMISSION CONTROL TECHNOLOGY

APPLICATIONS ON ELECTRIC UTILITY BOILERS 7 (Sep.

2000). The then-best estimate of mercury removal

costs ranged from 0.305 mills per kilowatt-hour at

the low end to 3.783 mills per kilowatt-hour at the

highest,'§ just 0.4 to 4.9 percent of the average retail

price of electricity in 2000.17 The costs of elemental

mercury removal were roughly comparable to the

cost of removing nitrogen oxides. Jd. at 23.'8

16

The use of hot-side ESP technology at a smal] number of

power plants is responsible for the highest numbers. If those

are excluded, the upper end of the range is only 1.915 mills per

kilowatt-hour. Id. at 22-23.

17 On average, consumers paid six to eight cents per kilowatt-

hour for electricity. U.S. ENERGY INFO. ADMIN., DOE/EIA-

0384(2011), ANNUAL ENERGY REVIEW 2011 (Sep. 2012),

available at http://www.eia.gov/totalenergy/data/annual/pdf/

aer.pdf. A mill is Viooo of a dollar.

18 See also 2000 NESCAUM Report at VI-19. Both the EPA

study and the NESCAUM report were part of the

administrative record EPA assembled in conjunction with the

2000 Finding in preparation for the rulemaking. Docket Nos. I-

25

Research indicated that the total costs of mercury

removal vary according to an EGU’s existing

configuration of pollution controls, the type of coal it

burns, and the efficiency of the mercury removal

required. For example, one study showed that a

relatively small EGU producing 100 megawatts of

power, burning low-sulfur bituminous coal, and

equipped with an ESP for controlling PM and ACI

and spray cooling for controlling elemental mercury,

would experience costs ranging from a low of 1.262

mills per kilowatt-hour if only 60 percent of the total

mercury is removed to a high of 2.810 mills per

kilowatt-hour if 90 percent of the total mercury is

removed, just 1.6 to 3.7 percent of the average retail

price of electricity in 2000. Jd. at 11 tbl.3 (showing

ESP-4). To give another example, in September 2000,

EPA looked at the effectiveness of removing mercury,

evaluating two possible regulatory scenarios and a

variety of scrubber/fabric filter/ESP/sorbent injection

configurations. Unsurprisingly, the regulatory

scenario that required larger reductions of mercury

(an 80 percent reduction) cost more than the version

that required lower reductions (60 percent).

Memorandum, Mercury Cost Calculations:

Assumptions, Approach, and Results (Sep. 2000),

available at http://www.epa.gov/ttn/atw/combust/

utiltox/hgmemo.pdf.

In other words, while EPA did not conduct a

formal cost-benefit analysis in connection with its

A-138-40; I-A-143, A-92-55, available at

http://www _.epa.gov/ttn/atw/combust/utiltox/eu_index-

master_121603.pdf.

26

2000 finding that regulation of EGU HAP emissions

was appropriate, it was neither blind nor indifferent

to the practical aspects of compliance with any rule

that it might later promulgate. It was aware of the

control technologies and strategies available, it was

aware of the factors driving costs of installation and

use of those technologies, and it was aware of

technological advances on the horizon that could be

adopted by EGUs in the near future. Regulation was

appropriate because it was practicable and

achievable by the industry without compromising

grid reliability or economic security.

Il. EPA REASONABLY AFFIRMED ITS 2000

FINDING THAT:., REGULATION WAS

APPROPRIATE BECAUSE COST-EFFECTIVE

EMISSION CONTROL TECHNOLOGIES FOR

MERCURY AND OTHER HAPS WERE IN WIDE

USE BY 2012.

By 2012 many EGUs were subject to mercury and

other HAP emission requirements through state

regulations. In the more than 11 years following

EPA’s finding that regulation was appropriate, EGUs

installed emission control technologies in order to

meet state mercury and HAP limits; limits that were

in some cases more stringent than those eventually

promulgated by EPA. In addition, other federal

regulations — such as the EPA rules directed at

interstate air pollution, ozone, and acid rain!9 —

19 The Clean Air Interstate Rule (CAIR) was issued by EPA in

2005. Although it was later invalidated by the D.C. Circuit

Court of Appeals in 2008, it remained in effect during litigation

27

compelled a number of EGUs to install control

technologies for other pollutants that have the co-

benefit of controlling mercury and other HAPs. In all

cases, the EGUs were able to meet regulatory

its 2000 finding that regulation of HAP emissions

A. Between 2000 And 2012, Many EGUs

Installed Mercury Emission Control

Equipment In Response’ To State

Regulations.

EPA was not the only governmental agency to

consider regulation of mercury and other HAPs in

over the Cross-State Air Pollution Rule (CSAPR). CSAPR is now

scheduled to replace CAIR in 2015. 79 Fed. Reg. 71,663. The

NOx SIP Call was issued in 1998 and was designed to contro]

interstate ozone pollution by reducing NO, emissions.

28

the late 1990s-early 2000s. A number of state air

pollution agencies*had been aware of the risks posed

by mercury to human health and the environment

and were looking at implementing state-wide rules.

When EPA made its finding that federal regulation

was appropriate and necessary, some _ states

continued to move forward with their own laws.

The earliest state mercury regulation was adopted

by New Hampshire in 2002, followed by Connecticut

in 2003 and New Jersey in 2004. CLEAN ENERGY

GROUP, ENSURING A CLEAN, MODERN ELECTRIC

GENERATING FLEET WHILE MAINTAINING ELECTRIC

SYSTEM RELIABILITY: SUMMER 2011 UPDATE, app. A

(June 2011). These regulations were based in part on

a policy decision that requiring mercury control

technology was both necessary to protect the public

health and was feasible to implement without risking

the reliability of the state’s electricity grid. See JAMES

E. MCCARTHY, CONG. RESEARCH SERV., RL33535,

MERCURY EMISSIONS FROM ELECTRIC POWER PLANTS:

STATES ARE SETTING STRICTER LIMITS 6 & nn. 9-10

(2006).

New Jersey officials observed, for example, that

municipal solid waste incinerators using fabric filter

control and ACI had achieved 99 percent mercury

control over the last decade. They further stated:

“The USDOE cost analyses indicate that retrofitting

the coal-fired boilers with activated carbon injection

(ACI) and [fabric filters] ... can achieve 90 percent

mercury emission reduction. ACI has a low capitol

(sic) cost. It also has low operating costs if [fabric

filter] technology is used.” Jd. at n.10 (quoting New

29

Jersey Department of Environmental Protection,

Summary of Public Comments and Agency

Responses, Control and Prohibition of Mercury

Emissions, December 6, 2004 New Jersey Register,

pp. 83-84, available at http://www.nj.gov/

dep/rules/adoptions/mercury_rule7-27.pdf).

In 2005, EPA backtracked from its 2000 finding

that regulation of HAP emissions at EGUs was

appropriate and necessary and attempted to renfove

EGUs from the list of regulated sources in Section

112(c) of the Clean Air Act.2° This decision sparked

another round of interest in regulation at the state

level, with ten additional states adopting mercury

regulations in 2006 and 2007.2! After that, another

five states adopted regulations, bringing the total to

eighteen states by the summer of 2011.22 These

eighteen states represent 40 percent of states with

20 Revision of December 2000 Regulatory Finding on the

Emissions of Hazardous Air Pollutants From Electric Utility

Steam Generating Units and the Removal of Coal- and Oil-Fired

Electric Utility Steam Generating Units From the Section

112(c) List, 70 Fed. Reg. 15,994 (Mar. 29, 2005). EPA then

attempted to regulate mercury through section 111 instead.

Standards of Performance for New and Existing Stationary

Sources: Electric Utility Steam Generating Units, Final Rule,

70 Fed. Reg. 28,606 (May 18, 2005). Both regulations were

eventually vacated by the D.C. Circuit.

21 Those states were Delaware, Maryland, Illinois, North

Carolina, Montana, Minnesota, Massachusetts, New York,

Colorado, and Georgia. Clean Energy Group, supra, at App. A.

22 Wisconsin, South Carolina, Michigan, Oregon, and Virginia.

Id.

30

coal-fired EGUs.?3 Collectively, they have established

mercury limits on approximately 96 GW of coal-fired

capacity ~— approximately 30 percent of total

capacity.”4

In the regulated states, EGUs have been able to

achieve mercury emission reductions of 90 percent on

average, using the older technologies for controlling

PM and SO: that also reduce particle-bound and

ionized mercury, as well as making use of the newer

technologies for controlling elemental mercury that

became commercially available after 2000. U 3. GOVT

ACCOUNTABILITY OFFICE, GAQO-10-47, MERCURY

CONTROL TECHNOLOGIES AT COAL-FIRED POWER

PLANTS HAVE ACHIEVED SUBSTANTIAL EMISSIONS

REDUCTION 7 (2009). In particular, ACI, which was in

the pilot-test stage at the end of 2000, matured as a

technology in the first decade of the century. Field

testing from 2001-2002 demonstrated that ACI used

in conjunction with an ESP was capable of capturing

up to 94 percent of the elemental mercury released

during the combustion of bituminous coals and

approximately 65 percent of the mercury released

during the combustion of subbituminous coals. When

used in conjunction with a fabric filter, removal

efficiencies in excess of 90 percent were achieved

while using less sorbent than required to achieve

23. Alaska, Hawaii, Idaho, Rhode Island, and Vermont do not

have any coal-fired EGUs. EPA, National Electric Energy Data

System (NEEDS) Database v.4.10, http://www.epa.gov/

airmarkets/documents/ipm/NEEDSv410.zip (last visited Mar. 2,

2015) (NEEDS v.4.10).

24 Jd.

31

similar efficiencies with an ESP NORTHEAST STATES

FOR COORDINATED AIR USE MANAGEMENT, MERCURY

EMISSIONS FROM COAL-FIRED POWER PLANTS 4-5

fig.4.2 (2003).

In 2002, Salem Harbor Station in Salem,

Massachusetts became the first coal-fired power

plant in the U.S. to utilize ACI, installing it on three

units. NEEDS v.4.10. By 2010, an additional 65 units

had installed or planned to install ACI to meet

current or anticipated mercury emission limits.

At the same time, the cost of ACI fell considerably

relative to earlier estimates. Before 2000, the

estimated costs had relatively large uncertainties

and varied considerably, ranging from $10 million to

$31 million for a large EGU unit.25 Mercury RTC,

Vol. VIII 3-11 tbl.3-5 (Dec. 1997). By 2007, the costs

ranged from $800,000 to $10.3 million while

achieving very high removal efficiencies. U.S.

DEPARTMENT OF ENERGY, DOE/NETL’S PHASE II

MERCURY CONTROL TECHNOLOGY FIELD TESTING

PROGRAM: UPDATED ECONOMIC ANALYSIS OF

ACTIVATED CARBON INJECTION 32 tbl.11, 35 tbl.14, 38°

tbl.17 (May 2007). Such reductions in costs and/or

improvements in efficiency are usually expected as a

new technology takes hold.

26 A large unit has a capacity of 975MW, and a smaller unit

100MW.

26 These figures represent annualized costs based on tests

done over a shorter period.

32

With controls for elemental mercury added to the

existing arsenal of controls, EGUs of differing design,

location, and size, that utilize different coals, found it

possible to achieve state-established mercury

limits.27 In fact, by the trme EPA proposed MATS in

2011, the EGUs in six states were already in

compliance with mercury limits stricter than the

proposed MATS standard. CLEAN ENERGY GROUP,

ENSURING A CLEAN, MODERN ELECTRIC GENERATING

FLEET WHILE MAINTAINING ELECTRIC SYSTEM

RELIABILITY: SUMMER 2011 UPDATE at 3 (June 2011).

The experience of the states that have implemented

mercury rules demonstrates that control of mercury

emissions is possible with available technology and

can be accomplished on a cost-effective basis and

without compromising reliability. See, e.g.,

NORTHEAST STATES FOR COORDINATED AIR USE

MANAGEMENT, COMMENTS ON Docket ID No. EPA—

HQ—OAR-2009-0234 at 5-8 (AUG. 2, 2011)

(discussing experience of NESCAUM _ states);

NATIONAL ASSOCIATION OF CLEAN AIR AGENCIES,

COMMENTS ON DOCKET ID No. EPA—-HQ—OAR-2003-—

0234 (Aug. 4, 2011) (discussing experiences in

regulated states).

27, U.S. GOov’T ACCOUNTABILITY OFFICE, GAO-10-47, MERCURY

CONTROL TECHNOLOGIES AT COAL-FIRED POWER PLANTS HAVE

ACHIEVED SUBSTANTIAL EMISSIONS REDUCTION 8 (2009) (finding

that sorbent injection can be used to reduce mercury emissions

on boiler configurations present at nearly three-fourths of coal-

fired EGUs).

33

B. Between 2000 And 2012, Many EGUs

Installed Control Technologies That Will

Reduce Mercury And Other HAP

Emissions In Response To Federal

Regulation Of Other Pollutants And Were

Able To Meet The MATS Requirements

Before They Took Effect.

MATS was not the only federal regulatory

initiative that affected mercury and HAP emissions

from EGUs. CAIR required the EGUs in 28 states to

limit their own emissions of NOx and SOz if those

emissions would contribute to nonattainment of

NAAQS in another state. The NO, SIP call required

26 states, mostly in the eastern U.S., to submit state

implementation plans that would address the

interstate transport of NOx and the formation of

ozone. The Acid Rain Program, part of the 1990 CAA

amendments, had also created incentives for EGUs to

control SOe, although not to the extent hoped for at

its enactment. While none of these rules addressed

mercury or HAP emissions per se, EGUs installed a

variety of control technologies — including scrubbers

and other FGD technologies — in order to comply with

their requirements.

As a result of the limitations on SO2 and NO,,

between 2000 and 2011, EGUs representing 80 GW

28 The Title IV Acid Rain Program (CAA §§ 401-416) differed

from other EPA regulations because it allowed for SO»

emissions trading, allowing some EGUs to lower their emissions

and obtain allowances or credits which they could then sell to

EGUs that did not lower their emissions to the same degree.

34

of generating capacity (approximately 25 percent of

the U.S. total) installed FGD technology, and EGUs

representing 96 GW _ of generating capacity

(approximately 30 percent of the U.S. total) installed

SCR systems. 2011 NESCAUM Report at 26. These

control technologies had the practical co-benefit of

limiting mercury and other HAPs, in addition to their

primary duties of reducing criteria pollutants.

EGU operators have come to recognize that

mercury and other HAP emission control is

technologically, operationally, and _ financially

feasible. Perhaps the most telling statistic is this: “At

the end of 2012, 64.3 percent of the U.S. coal

generating capacity in the electric power sector

already had the appropriate environmental control

equipment to comply with the MATS” — more than

two years before MATS was scheduled to take effect.

U.S. ENERGY INFO. ADMIN., Coal-fired power plant

operators consider emissions compliance strategies

(Mar. 28, 2014), http://www.eia.gov/

todayinenergy/detail.cfm?id=15611. Factoring in the

additional capacity that already planned to add

control equipment before the compliance deadline,

nearly 70 percent of total coal-fired capacity was

either in compliance with the MATS or already had

plans in place to achieve compliance at the end of

2012. Only a minority still needed to determine how

to comply with MATS — and it had over two years to

do so. Id. This is on-the-ground confirmation that

regulation of HAP emissions at EGUs_ was

appropriate and that EPA’s decision was and

remains reasonable.

35

CONCLUSION

For the foregoing reasons, this Court should

affirm the lower court’s decision.

Respectfully submitted,

Elizabeth J. Hubertz *

Interdisciplinary Environmental Clinic

Washington University School of Law

One Brookings Drive, Campus Box 1120

St. Louis, Missouri 63130

314.935.8760 e ejhubertz@wulaw.wustl.edu

* Counsel of record