Multi-Pollutant Sector Rule for the Pulp and Paper Industry

Response to Comments Received on

Proposed Information Collection Request

(Published on June 23, 2010)

U.S. Environmental Protection Agency

Office of Air Quality Planning and Standards

Sector Policies and Programs Division

Natural Resources and Commerce Group

Research Triangle Park, NC 27711

November 16, 2010

Table of Contents

Page

  TOC \o "2-5" \h \z \t "Heading 1,1"    HYPERLINK \l "_Toc276654608" 
1.0	Introduction	  PAGEREF _Toc276654608 \h  1  

  HYPERLINK \l "_Toc276654609"  2.0	Requirement to Collect Information	 
PAGEREF _Toc276654609 \h  2  

  HYPERLINK \l "_Toc276654610"  3.0	Burden of the Proposed ICR	  PAGEREF
_Toc276654610 \h  7  

  HYPERLINK \l "_Toc276654611"  3.1	Streamlining the ICR	  PAGEREF
_Toc276654611 \h  7  

  HYPERLINK \l "_Toc276654612"  3.2	Time to Complete the ICR	  PAGEREF
_Toc276654612 \h  9  

  HYPERLINK \l "_Toc276654613"  3.3	Timing of the ICR	  PAGEREF
_Toc276654613 \h  12  

  HYPERLINK \l "_Toc276654614"  3.4	Inventory Requirement	  PAGEREF
_Toc276654614 \h  13  

  HYPERLINK \l "_Toc276654615"  3.5	Process Flow Diagrams	  PAGEREF
_Toc276654615 \h  14  

  HYPERLINK \l "_Toc276654616"  3.6	HAP Additives	  PAGEREF
_Toc276654616 \h  15  

  HYPERLINK \l "_Toc276654617"  4.0	NCASI Emissions Information	 
PAGEREF _Toc276654617 \h  17  

  HYPERLINK \l "_Toc276654618"  5.0	Power Boiler Information	  PAGEREF
_Toc276654618 \h  18  

  HYPERLINK \l "_Toc276654619"  6.0	State-Only Permit Limits	  PAGEREF
_Toc276654619 \h  20  

  HYPERLINK \l "_Toc276654620"  7.0	Potential Changes	  PAGEREF
_Toc276654620 \h  22  

  HYPERLINK \l "_Toc276654621"  8.0	Technical Clarifications	  PAGEREF
_Toc276654621 \h  23  

  HYPERLINK \l "_Toc276654622"  8.1	Nominal Throughput Data	  PAGEREF
_Toc276654622 \h  23  

  HYPERLINK \l "_Toc276654623"  8.2	Emission Calculation Method Code	 
PAGEREF _Toc276654623 \h  23  

  HYPERLINK \l "_Toc276654625"  9.0	Other Specific Comments	  PAGEREF
_Toc276654625 \h  25  

  HYPERLINK \l "_Toc276654626"  9.1	Equip Detail Tab	  PAGEREF
_Toc276654626 \h  25  

  HYPERLINK \l "_Toc276654627"  9.2	Kraft Condensates Tab	  PAGEREF
_Toc276654627 \h  26  

  HYPERLINK \l "_Toc276654628"  9.3	PCC Tab	  PAGEREF _Toc276654628 \h 
26  

  HYPERLINK \l "_Toc276654629"  9.4	Emissions Test Data Tab	  PAGEREF
_Toc276654629 \h  26  

  HYPERLINK \l "_Toc276654630"  9.5	Segregation of CBI	  PAGEREF
_Toc276654630 \h  27  

  HYPERLINK \l "_Toc276654631"  10.0	CEMS Data	  PAGEREF _Toc276654631
\h  28  

 

List of Tables

Page

  TOC \h \z \t "Title_Table,1"    HYPERLINK \l "_Toc270691616"  Table 1.
 List of Commenters on Proposed ICR for Pulp and Paper Sector NSPS and
NESHAP Residual Risk and Technology Review	  PAGEREF _Toc270691616 \h  1
 

 

Acronyms and Abbreviations

ADTP

Air-dried ton(s) of pulp

AEI

Annual emissions inventory

AF&PA

American Forest & Paper Association

BACT

Best available control technology

BLS

Black liquor solids

Btu

British thermal unit(s)

CAA

Clean Air Act

CATs

Californians for Alternatives to Toxics

CBD

Center for Biological Diversity

CBI

Confidential business information

CEMS

Continuous emissions monitoring system

CISWI

Commercial/industrial solid waste incinerator(s)

CO

Carbon monoxide

dscf

Dry standard cubic foot (feet)

EPA 

U.S. Environmental Protection Agency

FR

Federal Register

GHG

Greenhouse gas

gpm

Gallon(s) per minute

gr

Grain(s)

HAP

Hazardous air pollutant(s)

hr

Hour(s)

ICR

Information collection request

lb

Pound(s)

MACT

Maximum achievable control technology

MM

Million

NAAQS

National ambient air quality standards

NCASI

National Council for Air and Stream Improvement, Inc.

NCG

Noncondensible gas

NEI

National Emissions Inventory

NESHAP

National emissions standards for hazardous air pollutant

NOI

Notice of intent

NOX

Nitrogen oxides

NSPS

New source performance standards

OSHA

U.S. Occupational Safety & Health Administration

PCC

Precipitated calcium carbonate

PM

Particulate matter

PSD

Prevention of Significant Deterioration

QA

Quality assure

RCRA

Resource Conservation and Recovery Act

RICE

Reciprocating internal combustion engine(s)

RTR

Residual risk and technology review

SARA

Superfund Amendments and Reauthorization Act

SDT

Smelt dissolving tank

SO2

Sulfur dioxide

SOG

Stripper off gas

TRS

Total reduced sulfur

1.0	Introduction

On June 23, 2010, the U.S. Environmental Protection Agency (EPA)
requested comment (75 FR 35792) on a proposed information collection
request (ICR) for the pulp and paper multi-pollutant sector rule (EPA
ICR No. 2393.01).  The sector rule includes both the new source
performance standards (NSPS) review and the national emission standards
for hazardous air pollutants (NESHAP) residual risk and technology
review (RTR).  The respondents affected by this action are
owners/operators of mills that are major sources of hazardous air
pollutant (HAP) emissions, as defined in 40 CFR 63.2.  These major
source mills include: 

Mills that perform chemical wood pulping (kraft, soda, sulfite, or
semichemical)

Mills that perform mechanical, groundwood, secondary fiber, and non-wood
pulping

Mills that perform bleaching

Mills that manufacture paper or paperboard products

A 60-day comment period (ending August 23, 2010) was provided for the
public to submit comments to EPA regarding the proposed new data
collection.  A total of three comments were received by EPA regarding
the proposed information collection request.  (See www.regulations.gov,
docket ID no. EPA-HQ-OAR-2007-0544, for the complete comments.) 
Table 11 lists the names of the commenters, the commenters’
affiliations, and the docket item number assigned to each of the
comments submitted to EPA.  Where one commenter (0011) endorsed the
comments by another commenter (0009), the docket item numbers associated
with both commenters were presented together in the comment summary.

Table 1.  List of Commenters on Proposed ICR for Pulp and Paper Sector
NSPS and NESHAP Residual Risk and Technology Review

Docket No.

EPA-HQ-OAR-2007-0544	Date Received	Commenter Name and Affiliation

0009	08/23/2010	Timothy G. Hunt

Senior Director, Air Quality Programs

American Forest & Paper Association (AF&PA)

0010	08/23/2010	Stephen E Woock

EHS&S Federal Regulatory Affairs Manager

Weyerhaeuser Company

0011	08/23/2010	Roy McAuley

Executive Director

Alabama Pulp & Paper Council



This document summarizes the substantive comments received by EPA on the
proposed ICR and presents EPA’s response to each of those comments. 
The comment summaries and responses are grouped by topic in the
following sections.2.0	Requirement to Collect Information

Comment:  Commenters 0009 and 0011 noted that in the Federal Register
(FR) notice of the proposed ICR, EPA asserts that the information being
solicited in the industry questionnaire is necessary in order to meet
Clean Air Act (CAA) requirements.  The two commenters argued that EPA is
substantially overstating what it is required to do under the CAA. 
According to the commenters, the questionnaire is directed, to a
significant extent, at obtaining information to support EPA rulemaking
that is not required and that, in the context of the large number of
additional CAA requirements being imposed on the industry, should not be
undertaken at this time.  Thus, the commenters argued that much of the
paperwork burden EPA proposes to impose through the proposed ICR is
unnecessary.  Commenter 0011 also stated that some of the information
requested by EPA is outside the scope of the CAA.

The commenters noted that the preamble for the proposed ICR (75 FR
35794) states that “Section 111(b)(1)(B) of the CAA mandates that EPA
review and, if appropriate, revise existing NSPS at least every eight
years….Another review of the kraft pulp mill NSPS is required under
the CAA.”  According to the commenters, this statement ignores the
fact that CAA section 111(b)(1)(B) also states that “the Administrator
need not review any such standard if the Administrator determines that
such review is not appropriate in light of readily available information
on the efficacy of such standard.”  The commenters noted that there
are no other qualifiers on the Administrator’s decision not to review
existing NSPS.  As a result, the commenters argued that EPA has almost
unlimited discretion to decide not to do a periodic review of the Kraft
Pulp Mill NSPS.  According to the commenters, stating that EPA is
“required” to review the kraft pulp mill NSPS is claiming a
justification for the ICR that does not really exist.

The commenters stated that EPA has already been provided a great deal of
information about emissions from kraft pulping vents and kraft pulp mill
chemical recovery systems, as well as the emission controls used for
these sources, from the Subparts S and MM NESHAP rulemakings and from
compliance demonstrations and continuous monitoring required by those
regulations.  The commenters noted that EPA has also been considering
these emissions recently in its ongoing evaluation of residual risk for
pulping and chemical recovery sources.  The commenters argued that EPA
could, on the basis of “readily available information,” determine
that review of the kraft pulp mill NSPS, or a portion thereof, is not
necessary.

The commenters noted that the existing NSPS require, among other things,
collection and incineration of most sources of total reduced sulfur
(TRS) compound emissions at kraft pulp mills and, thus, already require
a high level of TRS reduction.  The commenters further noted that the
pulp and paper maximum achievable control technology (MACT) standards
also require incineration, or similarly highly effective means, of
controlling the organic gas streams from kraft pulp mills, which also
contain TRS.  Consequently, the commenters argued that EPA already has
sufficient information to conclude that there is no more effective
technology available for controlling pulp mill TRS emissions.

The commenters acknowledged that particulate matter (PM) control
technology has improved since the kraft pulp mill NSPS were promulgated,
but noted that EPA has since that time promulgated stringent PM emission
limits in the MACT standards for pulp mill combustion sources as a
surrogate for metal HAP.  The commenters noted that these MACT standards
are much lower than the NSPS limits--0.015 vs. 0.044 grain per dry
standard cubic foot (gr/dscf) for recovery furnaces, 0.12 vs. 0.2 pound
per ton of black liquor solids (lb/ton BLS) for smelt dissolving tanks
(SDTs), and 0.01 vs. 0.067 or 0.13 gr/dcsf for lime kilns. As a result,
the commenters argued that EPA should conclude that review of the NSPS
PM limits would not be productive.

The commenters further argued that EPA is under no obligation to add
pollutants emitted by kraft pulp mills but not covered by the kraft pulp
mill NSPS to those already regulated by the NSPS.  According to the
commenters, nothing in CAA section 111 requires that NSPS cover all
pollutants emitted by a source, and the commenters stated that EPA has
never interpreted it that way.  The commenters cited the following three
references as support for this position:  (1) 74 FR 51950, 51957 (Oct.
8, 2009) “The statutory scheme thus provides EPA with significant
discretion to determine which pollutant(s) should be regulated under the
NSPS.”); (2) National Lime Association vs. EPA, 627 F.2d 416, 426
(D.C. Cir. 1980) (observing that, while “lime plants were determined
to be sources of nitrogen oxides, carbon monoxide, and sulfur dioxide as
well as particulates, standards of performance were proposed and
ultimately promulgated only with respect to particulate matter.”); (3)
70 FR 9706, 9711 (Feb. 28, 2005) (where EPA declined to set nitrogen
oxide (NOX) limits for boilers below 100 million British thermal units
per hour (MMBtu/hr) heat input, based on current emission levels,
available technologies, and costs).  According to the commenters, before
expanding the kraft pulp mill NSPS to cover additional pollutants, EPA
must justify its departure from its prior decisions about which
pollutants the NSPS should cover.

The commenters argued that the FR preamble glosses over the fact that,
even if the CAA required EPA to review the kraft pulp mill NSPS, the
questionnaire addresses all pulp, paper, and paperboard mills, including
pulp mills that use processes other than the kraft process subject to
the kraft pulp mill NSPS, and mills that do not do any pulping or only
pulp wastepaper.  The commenters concluded that, for about 3/4 of the
mills that would be required to fill out the questionnaire, EPA’s
stated reason for collecting information to review NSPS is not
applicable.

The commenters argued that EPA is not under any legal obligation to
expand the current NSPS to cover other types of mills.  The commenters
stated that EPA has a great deal of discretion to choose source
categories to be regulated through NSPS and to define which emission
units within a source category will be subject to emission standards. 
According to the commenters, the CAA leaves to the EPA Administrator the
determination of whether, “in his judgment,” the source category
“causes, or contributes significantly to, air pollution which may
reasonably be anticipated to endanger public health or welfare.”  The
commenters stated that EPA has made the judgment in the past that other
types of pulp mills (such as those use the sulfite process or
thermo-mechanical pulp mills) and non-integrated paper and paperboard
and secondary fiber mills do not warrant development of NSPS.  The
commenters were not aware of any basis for changing that decision.  The
commenters noted that the emissions from those other types of pulp and
paper mills are lower now than they were in the past, because of MACT
standards and numerous other air pollution control regulations.  (The
commenters further noted that sulfite process pulp mills have largely
disappeared—the commenters believe there may only be five sulfite
mills still operating in the U.S.—which the commenters argued makes
them even less of a target for development of categorical NSPS
limitations.)

The commenters also argued that EPA’s claimed basis for needing
information to review and revise NESHAP is overstated.  The commenters
specifically pointed to the FR preamble statement that “recent case
law and legal petitions suggest the need to review the pulp and paper
NESHAP” and that EPA needed “to make the NESHAP consistent with CAA
precedent established in recent judicial rulings.” (75 FR 35,794) 
According to the commenters, these statements appeared to be in direct
conflict with the judicial review and emission standard review
provisions that Congress included in the CAA.  The commenters stated
that under CAA section 307, NESHAP must be challenged through a petition
for review filed in the courts of appeals within 60 days after
publication of the rule.  The commenters noted that neither the Subpart
S nor the Subpart MM NESHAP was challenged in that manner.  According to
the commenters, EPA is under no obligation to conform these regulations
to principles announced in some cases involving challenges to other
NESHAP.  The commenters further noted that Subparts S and MM have been
in effect for many years, and have been relied on by pulp and paper
mills in making their compliance expenditures and other capital
investments.  The commenters stated that Congress established a system
designed to provide certainty and clarity: anyone objecting to a rule
must challenge it within 60 days; otherwise it remains in effect. 
According to the commenters, there is no provision for continued
revision of the MACT standards in Subparts S and MM.  The commenters
stated that those standards are to remain in place unless EPA determines
more stringent limits are necessary to avoid unacceptable residual risk
under CAA section 112(f), or EPA determines that there have been
significant advances in hazardous air pollutant control technology,
pursuant to CAA section 112(d)(6).

The commenters argued that the preamble glosses over the fact that, even
if EPA were required to review existing pulp and paper NESHAP to address
prior case law, that would not provide any justification for seeking
information from the majority of the estimated 386 mills to which the
questionnaire would be directed, since most of those mills are not
currently subject to Subparts S or MM. The commenters indicated that, at
a minimum, the ICR needs to clearly state that the major paperwork
burdens on those mills are being imposed in order to support a further
expansion of NESHAP, if that is EPA’s intention.  According to the
commenters, a better resolution would be for EPA to exercise its
discretion not to expand the NESHAP affecting the industry at this time,
and, therefore, not to send the questionnaire to mills not currently
covered by Subparts S or MM, especially in light of all of the other
regulatory burdens facing the industry at this time (not to mention EPA
resource constraints).

Response:  Three federal emission standards are the subject of the pulp
and paper information collection:

Standard	Promulgation date

Standards of Performance for Kraft Pulp Mills (40 CFR part 60, subpart
BB)	1978 

(reviewed 1986)

National Emission Standards for Hazardous Air Pollutants from the Pulp
and Paper Industry (40 CFR part 63, subpart S)	1998

National Emission Standards for Hazardous Air Pollutants for Chemical
Recovery Combustion Sources at Kraft, Soda, Sulfite, and Stand-Alone
Semichemical Pulp Mills (40 CFR part 63, subpart MM	2001



Section 111(b)(1)(B) of the CAA mandates that EPA review and, if
appropriate, revise existing NSPS at least every 8 years. Section
112(f)(2) of the CAA directs EPA to conduct risk assessments on each
source category subject to maximum achievable control technology (MACT)
standards and determine if additional standards are needed to reduce
residual risks.  Section 112(d)(6) of the CAA requires EPA to review and
revise the MACT standards, as necessary, taking into account
developments in practices, processes, and control technologies.  The
section 112(f)(2) residual risk and section 112(d)(6) technology reviews
are to be done 8 years after NESHAP promulgation.  Thus, all three of
the emission standards listed above are due for review according to
dates set forth in the CAA.  While we understand that the pulp and paper
industry is currently facing a large number of additional CAA
requirements, EPA cannot base its decisions (whether about issuing a
survey or developing a rulemaking) on what other rulemakings are
underway for a particular industry, but on what is required under the
CAA.  

It should be noted that EPA has entered into a proposed Consent Decree
(to address a lawsuit filed by Sierra Club) that would require proposal
of the subpart S RTR by June 15, 2011 and final action by December 16,
2011. In addition, EPA has received a notice of intent (NOI) to sue from
Californians for Alternatives to Toxics (CATs) and the Center for
Biological Diversity (CBD), contending that EPA has failed to review the
NSPS for kraft pulp mills within the statutory deadline under CAA
section 111(b)(1)(B).  The EPA received a petition for rulemaking in
January 2009 requesting that EPA revise various NESHAP, including the
NESHAP for chemical recovery combustion sources at pulp mills, to make
the NESHAP consistent with CAA precedent established in recent judicial
rulings.  Also, in December 2008, the U.S. Court of Appeals for the D.C.
Circuit vacated the startup, shutdown, and malfunction provisions
contained in the NESHAP General Provisions that apply to pulp and paper
mills.  To the extent that these legal actions need to be addressed in
the pulp and paper NESHAP, EPA intends to collect the information needed
to investigate potential rule revisions at the same time as collecting
information for the CAA-required statutory reviews.  

In the context of NSPS review, we acknowledge that CAA section
111(b)(1)(B) states that “the Administrator need not review any such
standard if the Administrator determines that such review is not
appropriate in light of readily available information on the efficacy of
such standard.” (emphasis added)   However, up-to-date and readily
available information are lacking.  As discussed below, much of the
information EPA has available for the pulp and paper industry is nearly
20 years old.  At present, EPA has no complete data base reflecting the
post-MACT and post-effluent guidelines configurations of pulp and paper
emission units and air pollution control systems.  The EPA is seeking
additional updated information to use in the residual risk and
technology review (RTR) regulatory analyses required under CAA sections
112(d) and 112(f)(2).  There is considerable overlap in the emission
units regulated under NSPS subpart BB and the two NESHAP for the pulp
and paper industry.  As a matter of administrative efficiency, EPA
intends to use a single collection of information to seek updated
information for use in the RTR for both NESHAP and in reviewing the NSPS
(as required under CAA section 111(b)). 

We disagree that the industry has already provided sufficient
information to EPA to perform the required regulatory reviews.  We
acknowledge that the industry provided volumes of useful information for
purposes of NESHAP development.  However, this information used to
support the subpart S and MM rulemakings is now outdated and may no
longer be representative of the latest post-MACT developments in
practices, processes, and control technologies for the pulp and paper
industry.  (The survey collecting pre-MACT information was sent out in
February 1992.)  Further, EPA’s Office of Air Quality Planning and
Standards (OAQPS) does not receive the results from most compliance
demonstrations and continuous monitoring being requested in the ICR
(e.g., because these data, if submitted at all, are typically submitted
to State agencies delegated authority to administer federal standards). 
The preliminary information that EPA has received from some mills
recently for its evaluation of residual risk for the subpart S RTR is
only a small fraction of the information needed to assess all of the
required regulatory reviews.  Thus, the information EPA has in hand is
not sufficient for performing thorough regulatory reviews.   

The EPA believes that the most appropriate and efficient way to assess
the latest technological developments for the pulp and paper industry is
to gather updated information through a CAA section 114 ICR.  The survey
would also reveal if there have been no advances in technologies (as the
commenters assert).  Information collected directly from pulp and paper
mills will have the greatest practical utility for purposes of
performing the RTR and NSPS reviews as information from the affected
industry will contain the most up-to-date, accurate, and reliable
equipment and operational data for each mill.  CAA section 114(a) states
that the Administrator may require any owner or operator subject to any
requirement of this Act to:  (A) Establish and maintain such records;
(B) make such reports; (C) install, use, and maintain such monitoring
equipment, and use such audit procedures, or methods; (D) sample such
emissions (in accordance with such procedures or methods, at such
locations, at such intervals, during such periods, and in such manner as
the Administrator shall prescribe); (E) keep records on control
equipment parameters, production variables or other indirect data when
direct monitoring of emissions is impractical; (F) submit compliance
certifications in accordance with section 114(a)(3); and (G) provide
such other information as the Administrator may reasonably require.

We do not wish to prejudge the outcome of our regulatory reviews prior
to collecting information upon which to base these reviews.  Without
more information, we cannot at present time conclude that a review of
the NSPS emission limits is unnecessary, nor can we conclude that the
emission limits should be changed.  The aforementioned NOI asserted that
EPA must provide standards for all air pollutants that endanger public
health and welfare when reviewing and revising the NSPS for kraft pulp
mills.  We will not know whether if is prudent to extend the NSPS to
additional pollutants until we have collected and reviewed updated
information to make definitive determinations.  

Mills that do not manufacture kraft pulp are included in the pulp and
paper survey because they are part of the subpart S source category for
which MACT standards were considered, and some of these mills are also
part of the subpart MM source category.  These mills are included in the
survey for RTR purposes.  Some of these mills may be true area sources
of HAP emissions, and therefore exempt from the pulp and paper survey. 
Criteria for exemption from the pulp and paper sector survey are as
follows: 

The mill is not a “major source” or “synthetic area source” of
HAP emissions, or

The mill was not operational in 2009 (and remains closed), or 

The mill does not produce pulp, perform bleaching, or serve as a primary
manufacturer of paper or paperboard products.

3.0	Burden of the Proposed ICR

3.1	Streamlining the ICR

Comment:  Three commenters (0009, 0010, and 0011) recommended that EPA
significantly streamline the survey.  One commenter (0010) stated that
EPA is seeking far more information—and overly detailed
information—than is necessary, given the time period and the timing
constraints that EPA feels it must address.  Commenter 0010 suggested
that EPA focus on the emissions data and emission units information, at
least as a first pass at the survey.  After EPA has completed evaluating
those data, commenter 0010 indicated EPA could then target any
additional specific mill processing and other information needed to
focus the new regulations.  Commenter 0010 noted that each of the
regulations addressed by the ICR will be a follow-up regulatory review
by EPA and are not original rulemakings in response to a new statutory
requirement.  As a result, commenter 0010 argued the additional
information EPA is seeking is largely unrelated, or at least
unnecessary, to carry out the regulatory reviews that EPA is authorized
to conduct.  According to commenter 0010, much of the data and
information (e.g., production and production information and capacities,
detailed process diagrams and data, potential future changes) that the
commenter thinks EPA should forego at least for the first round of the
survey also tend to be confidential business information (CBI).

For those mills covered by Subparts S and MM, two commenters (0009 and
0011) recommended that EPA focus on requesting information as it
pertains to emissions subject to Subpart S.  Commenters 0009 and 0011
noted that EPA currently has proposed to enter into a consent decree
with a deadline of June 15, 2011 to sign a RTR proposal for Subpart S. 
Based on conversations with EPA representatives, the two commenters
expected that the survey will be issued to facilities in late 2010 or
early 2011.  Commenters 0009 and 0011 estimated it will take at least 2
months for EPA to compile and analyze the responses and do the final
round of dispersion modeling with updated emissions inventory
information.  According to the two commenters, this would leave little
time for the new information to be gathered by this survey to be
meaningfully considered by the technical staff in their proposal and
still allow for adequate review by EPA management.  Therefore,
commenters 0009 and 0011 recommended that EPA issue a more limited
survey targeted toward Subpart S to try to meet its June 15 deadline,
postponing any future information collection activities until after the
Subpart S review is completed.

The two commenters noted that this approach would exclude data based on
NSPS requirements and other information not related to the collection
and control of HAP from processes covered under Subpart S.  If
facilities are only required to submit data applicable to Subpart S,
commenters 0009 and 0011 indicated that chemical recovery combustion
equipment, power boilers, and woodyard operations would be eliminated
from the survey (based on the list of emission units listed in
Attachment 1 of the survey instructions).  According to the two
commenters, this would reduce the burden of the ICR and allow EPA to
focus on the data necessary for the Subpart S RTR it is trying to
complete by June 15, 2011.  To the extent that EPA has specific data
needs for data related to equipment not addressed directly by Subpart S
(e.g., availability of power boilers used to incinerate organic HAP),
commenters 0009 and 0011 indicated that the industry would be willing to
work with EPA to identify and provide the additional information, as
they have in the past.

If EPA believes it cannot avoid conducting a review of the kraft pulp
mill NSPS, on the basis of “readily available information,”
commenters 0009 and 0011 recommended that EPA limit just to kraft pulp
mills the distribution of those portions of the questionnaire relevant
to NSPS review.  The two commenters argued that sending the
questionnaire to the hundreds of other mills would impose a major
reporting burden.

Response:  As explained above, there is considerable overlap in the
emission units covered under subpart S, subpart MM, and NSPS subpart BB;
there are statutory obligations under the CAA that EPA must fulfill for
all three subparts; and it is more efficient for EPA to use one
information collection to gather all of the information useful for
performing all of the regulatory reviews.  We understand the concerns
that the commenters have regarding the large amount of information
requested of them in such a short time.  We are constrained by the
consent decree deadline for subpart S RTR in setting the timeline for
this survey.  We have taken a number of steps to address the commenters
concern and to streamline the survey.  Several simplifying changes to
the survey questions are noted throughout this document.  More
importantly, we have divided the survey into three parts based on
equipment type (i.e., pulp and paper production units versus chemical
recovery combustion units) with different submittal dates as follows: 

Part I - Mill Overview and Subpart S Data

a. One-page notification of true area source, non-operational, or
non-applicable status (due in 20 days)

	b. Completion of Part I data submittal (due in 30 days)

Part II - NEI Update (due in 100 days)

Part III - Chemical Recovery Combustion Sources (180 days)

Each part of the streamlined survey has a dedicated instruction
document.  In addition, we developed a survey overview document that
contains information and attachments useful for all parts of the survey.
The survey overview document includes an attached one-page notification
form that mills may submit to indicate if the survey does not apply for
their mill (e.g., because they are true area sources, were not
operational in 2009, or do not manufacture primary pulp and paper
products).  Survey Parts I and II apply for all mills that are not
exempt from the survey.  Part III applies only for pulp mills that
operate chemical recovery combustion sources.

Part I of the survey includes data useful for the subpart S technology
review (general mill information and flow diagrams; processes, products;
equipment, controls, and compliance options in use for pulp and paper
production equipment; permit limits; and emissions test data for pulp
and paper processing equipment).  Part III of the streamlined pulp and
paper survey requests information for chemical recovery combustion
units.  The data request under Part III is similar in scope to the data
request under Part I (except that all flow diagrams are requested in
Part I).  The information requested in Parts I and III will be also
useful for mills with applicable equipment under the NSPS review.

Part II of the survey is the NEI update useful for both the subpart S
and subpart MM residual risk reviews.  As noted elsewhere by commenters,
the NEI update is one of the most time consuming portions of the pulp
and paper survey.  We are requesting one NEI update (as opposed to an
update for subpart S emission units followed by a later update for
subpart MM emission units) because one update: (1) is more informative
for residual risk purposes (e.g., consideration of facility-wide risk),
and (2) is far more efficient for both EPA and respondents (e.g., avoids
multiple repetitive data handling/manipulation operations).  Also, in
some cases emission units covered under subpart MM (e.g., a lime kiln)
serves as the control devices for emission units covered under subpart S
(e.g., non-condensable gases [NCGs] or stripper off gases [SOGs]).  In
addition to providing more time for completion of the NEI update, we
streamlined the NEI update as follows: 

Made any NEI updates on TRS and CAP emissions that respondents would
provide optional.

Added a separate “NEI blank.xls” spreadsheet template for mills with
no prior NEI data (to minimize confusion), and clarified instructions
for these mills.

Revamped to NEI update instructions to provide more detail on some of
the data columns of particular importance for residual risk purposes.

Given the consent decree schedule limitations, providing more time for
industry respondents to supply their NEI updates reduces the amount of
time that EPA has to process the NEI data received and model it for
residual risk purposes.  Therefore, EPA anticipates performing some
preliminary risk modeling (based on NEI data in hand) while awaiting the
NEI revisions from mills, understanding that data supplied by mills will
provide a far more robust data set for further RTR risk modeling
efforts. 

3.2	Time to Complete the ICR

Comment:  Three commenters (0009, 0010, and 0011) stated that EPA
underestimated the industry burden for responding to the questionnaire
and allowed too little time (60 days) to respond to a very lengthy
questionnaire.  Commenter 0011 stated that the scope of the proposed
questionnaire and the time provided for responses were unreasonable and
unnecessary.  Commenter 0011 stated that EPA’s estimates of the time
and cost to complete the survey are particularly low for the mills in
their state, given that those mills are larger than average.

Commenters 0009 and 0011 stated that, if EPA goes forward with the
survey content as proposed, industry would need at least 120 days to
review the ICR, get clarification on any necessary items that are not
understood, collect the information requested (assuming all of the
information requested is available), arrange it in a format for
inclusion in the spreadsheets, quality assure (QA) the information after
entry into the spreadsheets, and get management review of the
information to be submitted.

Commenter 0010 estimated that 3 to 15 work months (90 to 450 work days)
for one person will be necessary to complete the survey, based on their
staff’s review of the survey elements.  According to commenter 0010,
their burden estimates will result in higher costs than EPA is
estimating, depending on the labor rate.  Commenter 0010 argued that if
EPA does not streamline the survey, EPA may inadvertently encourage
facilities to respond to the survey with less thought and fewer updates
to the emissions database that EPA plans to use to pre-populate each
facility’s survey.

Commenters 0009 and 0011 noted that most of the time to prepare and
complete the ICR is dedicated to updating the National Emissions
Inventory (NEI) data for each facility for 2009.  The two commenters
also noted that EPA estimated only 16 to 40 hours for facilities to
update their NEI inventory if previously submitted.  According to the
commenters, for member facilities of the National Council of the Paper
Industry for Air and Stream Improvement (NCASI), almost all of the air
toxic emission factors have been updated in the past year, and it will
require a significant effort for a facility simply to update the
calculations based on the revised emission factors.  The commenters
argued that EPA’s estimate of 16 hours is likely not even enough to QA
the data entered by EPA for a large integrated pulp and paper facility. 
The commenters noted that EPA is also requesting additional columns of
information that mills have not previously provided (e.g., maximum
hourly emissions).

Commenters 0009 and 0011 provided the following table outlining the
differences in burden hours and cost where the commenters believe
EPA’s estimates are not accurate.  The basis of the commenters’
burden estimates is discussed further in the paragraphs below.

Respondent Activity Number	Respondent Activity	EPA Hour Estimate
Commenter’s Hour Estimate

4.B.a. 	NEI Update: Mills that previously updated their NEI for RTR 	16
40 if mills submitted a 2009 state AEI with Air Toxics 

120 if mills submitted a 2009 state AEI without Air Toxics 

4.B.a. 	NEI Update: Mills that have not previously updated their NEI for
RTR	40	40 if mills submitted a 2009 state AEI with Air Toxics 

120 if mills submitted a 2009 state AEI without Air Toxics 

4.B.a. 	NEI Update: Mills with no prior NEI data	220	AF&PA agrees that
220 is a reasonable estimate in this case 

4.B.b 	Detailed Flow Diagrams	8	40 

4.B.c. 	Inventory	32	40 – Facility with NEI 

60 – Facility with no prior NEI 

4.B.c. 	HAP additives	8	40 

4.B.e. 	CEMS spreadsheet	12	40 



For states that have air toxics included in the annual emissions
inventory, the two commenters recommended that EPA obtain data from the
state agencies to update the database for review prior to issuing the
ICR.  The two commenters stated that this would significantly decrease
the time for these mills to complete the survey since they would only
need to verify, rather than update, this information.

According to commenters 0009 and 0011, the February 2010 updates to
NCASI’s air toxics technical bulletin showed some emission rates
decreasing and others increasing.  As a result, the two commenters
indicated that a mill that previously submitted an NEI revision may want
to update emissions to provide a more accurate estimate (regardless of
whether the mill was modified since initial submittal).  For this
reason, the commenters did not see a difference in time for these mills
to update their inventory.  The commenters noted that additional time
will be necessary for mills that are in states that do not require
reporting of air toxics in each annual emissions inventory (AEI).  The
two commenters provided the following subcategories and time estimates
for completing the NEI portion of the survey:

Facility previously submitted NEI for RTR and reports toxics in their
AEI:  40 hours for preparation and 40 hours for data entry

Facility previously submitted NEI for RTR and does not report toxics in
their AEI:  120 hours for preparation and 40 hours for data entry

Facility that did not submit an NEI for RTR and reports toxics in their
AEI:  180 hours for preparation and 60 hours for data entry

Facility that did not submit an NEI for RTR and does not report toxics
in their AEI:  220 hours for preparation and 60 hours for data entry

Response:  The time constraints imposed by the consent decree for
subpart S RTR limit the amount of time that can be provided for industry
to respond to the pulp and paper survey.  As noted in Section 3.1 above,
we have taken steps to partition and streamline the survey where
possible to provide respondents more time to complete the full survey. 
Part I (Mill Overview and Subpart S Data) is due after 30 days, Part II
(NEI Update) is due after 100 days, and Part III (Chemical Recovery
Combustion Sources) is due after 180 days.  

We recognize that the NEI update is likely to be the most time consuming
part of the survey, and therefore, have provided additional time beyond
the proposed 60-days for completion of the NEI update.  We have also
reduced the proposed scope of the NEI update (e.g., by making
revisions/additions of non-HAP data optional).        

The survey has been further revised to request only off-the-shelf
process flow diagrams; simplify the request for permit limits; eliminate
the questions about potential future changes; clarify the scope of power
boiler questions; only request information on additives with
“measurable” amounts of HAP; and to clarify various other questions
in response to industry comments.

We believe the burden estimates provided by the commenters for the
respondent activities listed above (NEI updates, inventory, and the CEMS
spreadsheet) are reasonable, and we are increasing these burden
estimates, accordingly.  We have also adjusted the burden estimates to
reflect the 3-part streamlined survey.  Because we are requesting only
off-the-shelf process flow diagrams, with respondents marking updates on
the diagrams as necessary, we increased the burden estimate for this
respondent activity from a proposed 8 hours to 20 hours (instead of 40
hours as suggested by the commenters).  Similarly, because we are only
requesting information on additives with “measurable” amounts of
HAP,” with “measurable” defined in a way that should significantly
reduce the scope of the information request, we increased the burden
estimate for this respondent activity from a proposed 8 hours to 20
hours (instead of 40 hours as suggested by the commenters).

3.3	Timing of the ICR

Comment:  Three commenters (0009, 0010, and 0011) stated that the timing
of the survey is problematic and provides further justification for
reducing the scope of the ICR and providing more than 60 days for
responding.

Commenter 0010 noted that EPA has been issuing rulemakings at a
significant pace over the past year, with several of these rules coming
out around the time the EPA would distribute its survey.  Given their
limited staff and resources, commenter 0010 stated that it was important
that EPA consider the timing of the survey relative to the burden.  As
an example, commenter 0010 noted that over the next 8 months, their pulp
mills expect to review proposals and final rules and implement actions
to comply with the following air program initiatives:

Revisions to, and first ever reports under, the Greenhouse Gas (GHG)
Mandatory Reporting rule;

First-time inclusion of GHGs on projects they have in progress or are
planning under EPA’s GHG Tailoring Rule for Title V and Prevention of
Significant Deterioration (PSD)

Understanding and implementing as appropriate Industry Sector Guidance
on application of Best Available Control Technologies (BACT) for GHGs;

Updates and additions to the Reciprocating Internal Combustion Engines
(RICE) MACT for stationary engines;

Potential industrial boiler treatment under the proposed Clean Air
Transport Rule;

Understanding and implementing new modeling requirements for sulfur
dioxide (SO2), NOX, and PM under the tighter national ambient air
quality standards (NAAQS) and EPA’s new NAAQS modeling initiatives;
and

Final solid waste definition rules under the Resource Conservation and
Recovery Act (RCRA) non-hazardous secondary materials solid waste
definition for energy production, and simultaneously understanding and
implementing compliance strategy plans for the final Boiler MACT and/or
Commercial/Industrial Solid Waste Incinerator (CISWI) rule for each
boiler and process heater onsite.

Commenters 0009 and 0011 also noted that companies may be responding to
the ICR during the first part of 2011, which is a time when mill and
company personnel who would be involved in responding to the ICR also
are responsible for preparing numerous other reports to state and
federal authorities (e.g. state annual emission inventories, Title V
permit annual compliance reports, Title III reports for the Superfund
Amendments and Reauthorization Act (SARA), etc.).

Response:  We understand the commenters’ concerns regarding the timing
of the survey.  The deadlines in the consent decree for subpart S RTR
dictate the timing for the subpart S portion of the survey (Part I) and
the NEI update (Part II).  As noted in Section 3.1 above, we have taken
additional steps to partition and streamline the survey, which should
alleviate some of the commenters’ concerns regarding the timing of the
survey.  While Part I of the survey (Mill Overview and Subpart S Data)
would be due after 30 days, we are providing additional time for the
industry to complete the remaining parts of the survey:  Part II - NEI
update (100 days) and Part III - chemical recovery combustion unit data
(180 days).

3.4	Inventory Requirement

Comment:  Two commenters (0009 and 0011) noted that in the Inventory tab
of the survey, EPA is requesting information not previously requested in
the NEI for RTR (e.g., maximum hourly emissions; and startup, shutdown,
and malfunction [SSM] emissions).  The commenters’ reaction to the
additional inventory requirements are discussed below:

Maximum hourly emission rates: The commenters agreed that the maximum
hourly emissions are a necessary piece of information and having this
information is an improved approach over calculating the maximum
emissions as ten times the average annual rate.

Startup, shutdown, and malfunction emissions: The commenters agreed that
emissions can be different, and usually higher, during periods of
startup, shutdown, and malfunction.  However, the commenters stated that
the data for emissions during startup, shutdown, and malfunction are not
readily available in the format requested in the survey, i.e.,
facilities only keep track of the number of minutes for each such period
(as required under 40 CFR Part 60 Subpart BB and 40 CFR Part 63 Subpart
S and MM).  According to the commenters, most facilities do not have
test data during startup, shutdown, and malfunction events, because the
unit must be under “normal operating conditions” for a valid
performance test.  Therefore, the commenters recommended that each
facility document the number of minutes each unit subject to these
regulations experienced startup, shutdown, and malfunction periods
during 2009.

Response:   Part II of the revised survey requests the following.

Field	Description

Routine Emissions (TPY)	Revised routine emissions

Routine Emissions Max Hourly Rate (lbs per hour)	Maximum hourly emission
rate for routine emissions.

Startup Emissions (TPY)	Emissions that occurred during startup periods.

Startup Emissions Max Hourly Rate (lbs per hour)	Maximum hourly emission
rate for startup period emissions.

Shutdown Emissions (TPY)	Emissions that occurred during shutdown
periods.

Shutdown Emissions Max Hourly Rate (lbs per hour)	Maximum hourly
emission rate for shutdown period emissions.



These data fields are requested because EPA is modeling both short- and
long-term risks.  Long-term risks would be based on the annual routine
emissions (tpy).  The routine emissions (tpy) can be used in determining
an average hourly emission rate for the year (e.g., tpy divided by
operating hours) for purposes of risk modeling.  Short-term risks would
be based on the maximum hourly emission rate(s) (e.g., lb/hr data).

For routine emissions, the average hourly emission rate (e.g., tpy
divided by operating hours) can differ from the maximum hourly emission
rate due to fluctuations in production rates or other changes in other
routine operating conditions.  In the absence of site- or
process-specific information, it has been EPA’s practice to
conservatively assume that maximum hourly emissions are 10 times the
routine hourly emissions for purposes of risk modeling.  We would prefer
to have better site-, process-, and/or pollutant-specific data (where
such data can be reasonably estimated) than to apply this default
factor-of-10 approach.  We believe individual facilities are in the best
position to provide such specific data.  Estimates are acceptable.  If
these data are unavailable and are not provided, then we’ll apply our
default approach.  We intend to work with pulp and paper industry
representatives through AF&PA to review the factor-of-10 default
approach and its applicability for the different types of pulp and paper
processes. 

Startup and shutdown are conditions of operation that can occur
throughout the year (with the number of startups and shutdowns depending
on the type of process and equipment involved).  Malfunctions are more
sudden and unpredictable in nature than are startups/shutdowns.  We
understand that emissions may differ (e.g., be higher or lower than
routine emissions) during startup, shutdown, and malfunction periods,
and that mills are unlikely to have measured emissions during such
periods.  We have changed the instructions for the startup and shutdown
fields to make responses “optional,” and eliminated the malfunction
fields.  We retained these now-optional questions in the survey in order
for us to gain a better understanding of how emissions change during
startup and shutdown for any mills that may have previously measured
(e.g., with CEMS) or estimated such emissions.  Again, we feel that
individual mills would be in the best position to provide the specific
annual (tpy) and maximum hourly (lb/hr) emissions data for startup and
shutdown periods.  

Additional questions relating to the duration of startup and shutdown
events are included in the Part I and III Equip detail and Controls
tabs. The EPA is considering standards that could apply during startup
and shutdown events (or whether the current standards developed for
normal operation should apply) in light of the December 2008 vacatur of
the NESHAP startup, shutdown, and malfunction exemption in 40 CFR Part
63 subpart A.  Questions pertaining to emission unit and control system
startup and shutdown are asked in order to provide EPA with an
understanding of the duration, emissions potential, work practices, and
control mechanisms of startup and shutdown events for the wide variety
of equipment used at pulp and paper mills.  To the extent that this
information is supplied to the Agency, it will inform our rulemaking
efforts and help to tailor new requirements for startup and shutdown
periods for the pulp and paper industry.  

It is helpful for EPA to know when certain data or information are
simply not known.  Therefore, throughout the survey, we request that
mills enter “unknown” where mills do not have the requested
information, cannot obtain the information without extraordinary effort,
and cannot provide a reasonable estimate.

3.5	Process Flow Diagrams

Comment:  Commenters 0009 and 0011 noted that the ICR requests that each
facility submit process flow diagrams, which EPA estimates will only
take 8 hours.  According to the two commenters, some facilities prepared
these types of diagrams for Title V permit applications and renewals,
but many states do not require the facility to submit updated drawings
with each permit renewal.  As a result, noted the commenters, the
process flow diagrams that most facilities have will be outdated, will
not contain all of the information and identifiers EPA is requesting,
and many times will be in software that is no longer used. The
commenters also indicated that these diagrams do not have the same
equipment ID numbers as the NEI database.  Therefore, according to the
commenters, the time to prepare these documents has been underestimated.
 The commenters estimated that it will take each facility at least 40
hours to update these diagrams in order to provide all of the
information EPA is requesting.  The commenters argued that it is
unreasonable for EPA to specify in great detail the information that
needs to be included in process flow diagrams and then assume that mills
already have the process flow diagrams and will only need a few hours to
assemble and provide them.  The commenter’s understanding was that
this ICR would not require mills to generate new information, which the
commenters said would not be the case for most mills in order to provide
the process flow diagrams that EPA is requesting.

Response:  We acknowledge the commenters’ concerns regarding the level
of detail specified for the process flow diagrams.  Our intent was not
to prescribe what information must be included in the diagrams, but to
give survey respondents an idea of the type of information that EPA was
looking for in the diagrams.  We are only seeking off-the-shelf diagrams
(not new drawings).  We believe the diagrams will help answer a lot of
questions we may have (thereby eliminating the need for follow-up calls
to many of the mills).  Respondents can use whatever diagrams they have,
and hand mark any changes/notes if needed.  Respondents can also forego
the labeling of each emission unit with IDs.  We have revised the survey
instructions to clarify these points.

3.6	HAP Additives

Comment:  Commenters 0009 and 0011 noted that the ICR requests that each
facility submit information on the HAP additives used at the paper
machines.  The two commenters stated that many facilities are constantly
changing additives and trying new additives to improve product quality
and develop new and improved products to adapt to changing market
conditions and product demand, so any information provided will
represent a snapshot in time.

The commenters also stated that suppliers are not always willing to
share their formulations because it is proprietary information, so mills
would need time to contact the many additive suppliers and request
information on the amounts of the six HAP in each additive.  The
commenters noted that pulp and paper companies are also very reluctant
to reveal the combination and quantities of additives used in the
manufacturing of their products. Therefore, the commenters concluded
that this information would, at a minimum, likely be deemed CBI.

The commenters noted that the survey requests that a facility report any
additive used in a “measurable” amount.  According to the
commenters, the U.S. Occupational Safety & Health Administration (OSHA)
requires that chemical suppliers provide specific constituent amounts in
Material Safety Data Sheets (MSDS) for any carcinogens that are present
in amounts greater than 0.1 percent and any non-carcinogens that are
present in amounts greater than 1 percent.  The commenters indicated
that this information will be available from suppliers, and the
commenters recommended that “measurable” amounts be based on these
criteria.  The commenters also requested that additives not be linked to
a specific paper machine but reported as total usage, in order to
preserve the confidentiality of the papermaking process.

Response:  We acknowledge that numerous HAP and/or non-HAP additives are
used at the paper machines.  The questions related to HAP additives are
limited to only those HAPs of potential concern for residual risk
purposes in order to reduce the scope of the information request.  We
are not asking for suppliers to reveal specific additive formulations,
nor are we asking for information beyond what is reported on MSDS
sheets.  Respondents to the survey would only need to include additives
with measurable quantities of HAP as reported on the MSDS for the
additive.  Based on the commenters’ suggestion, we have revised the
survey instructions to include a definition for the term
“measurable” that would be consistent with OSHA requirements.  We
have also added a sentence to clarify that we are not asking for
companies to go back to their suppliers to seek information beyond what
is reported on MSDS.  

To address the concern regarding confidential information, we have
removed the paper machine designation from the survey spreadsheet.  We
have also suggested in the pull-down menu text that annual resin use
might be provided (e.g., should a mill be more comfortable providing an
annual usage as opposed to daily or per ton of product).

As noted in the survey instructions, although it is unclear whether
paper machine additives contribute significantly to paper machine
emissions (e.g., because additives may partition to the paper product or
white water, and not necessarily to the air), the HAP additives
information would be used to evaluate work practices related to
additives.  Therefore, a question has been added to the survey asking
mills if they are aware of any non-HAP substitutes for the HAP additives
included in their response.  Mills are also asked to indicate if any of
the additive substitutes they identify contain other compounds that may
not be desirable from a worker safety perspective. 

4.0	NCASI Emissions Information

Comment:  Two commenters (0009 and 0011) offered the following responses
to EPA’s statement (on page 8 of the Supporting Statement) that
NCASI’s emissions test data may be out-of-date and only represent
NCASI members.

First, the commenters noted that NCASI issued updated Technical Bulletin
No. 973 in February 2010 for Air Toxics emissions for Pulp and Paper
Mill Sources.  According to the commenters, the update included new
site-specific data and eliminated data that were old or not
representative of the industry.  In addition, the commenters stated that
NCASI used the latest statistical techniques to address both non-detect
values and outliers.

Second, the commenters argued that, while the survey would request
emissions inventory data from other non-NCASI members, it has been the
experience of pulp and paper industry consultants that these facilities
either (1) do not estimate or include all HAP that NCASI members do in
their emissions inventories due to lack of data, (2) use older NCASI
emission factors if they were previous members of the association, or
(3) obtain updated NCASI data through other unapproved sources. 
According to the commenters, in all but the last case, these emission
estimates are not accurate and will skew the data set.

The commenters expected the majority of HAP emissions calculations (and
even much of the criteria pollutant emissions) will be based on NCASI
factors, and stated that it is important to note that these data have
been recently updated.  The commenters noted that the NCASI database
contains data means, medians, and detailed source information, so some
facilities may use a mean emission factor, some facilities may use
median emission factors, and some facilities may select the emission
factor from the equipment that best represents their mill.  According to
the commenters, the emissions estimates for non-NCASI members may be
based on older factors and/or omit compounds detected by NCASI.  The
commenters noted that their industry relies heavily on published
emission factor data and can use different treatments of the same
dataset, so estimated emissions from mills with similar configurations
may be very different.  The commenters argued that, as EPA examines the
data and performs the risk analyses, it is important to understand these
differences in order to ensure the proper use of data in the modeling
effort.

Response:  We acknowledge that, for the pulp and paper industry,
NCASI’s technical bulletins are an authoritative source for emission
factors.  The survey instructions cite them as a resource.  We also note
that the emissions data in the latest technical bulletins have been
updated—although as late as NCASI Technical Bulletin No. 973, the
majority of the emissions data are still 1990s (pre-MACT) data, based on
our examination of the bulletin’s appendices.  It was not our
intention in the Supporting Statement to devalue NCASI’s emission
factors but simply to explain why we believed that new emissions test
data were needed.

We have no control over how mills obtain the data, and we recognize that
inventories for NCASI member mills will likely be based on updated
factors, while other mill inventories may not.  We also recognize that
different mills may use different emission factors from the same NCASI
data set.  We are not prescribing how mills should estimate their
emissions.  We are, however, requesting that survey respondents provide
their emissions data source in the Part II NEI update.

5.0	Power Boiler Information

Comment:  Commenters 0009 and 0011 noted that EPA requests information
throughout the survey on power boilers and other incineration devices,
as these units can be used to treat non-condensable gas vents regulated
under 40 CFR Part 60 Subpart BB and 40 CFR Part 63 Subpart S.  While the
two commenters acknowledged that these units can be part of the
compliance approach for these regulations, the commenters did not
believe that EPA should collect detailed information related to the
fuels and capacities of power boilers under this ICR.

The commenters argued that any add-on control information for these
units is not necessary since the noncondensible gases (NCGs) are
presumed to be destroyed by the boiler (as indicated in the survey
instructions).  In addition, the commenters noted that pulp and paper
facilities have recently completed an ICR for the Boiler MACT
regulations, and stated that the Boiler MACT ICR database contains
detailed information for each boiler.  Therefore, the commenters
recommended that power boiler information should not be requested again
in the pulp and paper ICR.  According to the commenters, the information
requested should be limited to the type of unit used to control NCGs and
stripper offgases (SOGs) (boiler, oxidizer, kiln, etc.), if the facility
has a back-up control device, and how often (e.g., number of minutes)
the mill has bypassed these control devices due to their unavailability.

Response:  We have reduced the scope of the questions related to power
boiler fuel use as a result of these comments.  We have removed
questions pertaining to actual 2009 fuel use and replaced them with a
less-detailed request for an approximate percent of annual heat input
capacity (MMBtu/yr) supplied by each fuel.  We have also made the
questions related to heating values optional.  This information,
although less detailed, should provide EPA with enough information to
calculate emissions and perform impacts analyses for fuel-fired
equipment.  

The power boiler questions remaining in the pulp and paper survey appear
in Part I (Mill Overview and Subpart S Data), and Part II (NEI Update). 
The EPA is interested in pulp and paper mill boilers because boilers
contribute to the facility-wide risk at pulp and paper mills. Boilers
may also be used as a control device for pulp mill NCGs or SOGs. 
Further, boilers are one of the larger sources of air pollutants at pulp
and paper mills, and information on boilers would aid in understanding
mill-wide emissions (and in putting pulp and paper process-related
emissions into perspective).

The scope of the pulp and paper information request for boiler
information is more limited than the information requested in the Boiler
MACT survey.  It would be far less time-consuming from an Agency
resource perspective to seek limited, updated boiler information in the
pulp and paper survey than to try and piece together information from
the prior Boiler MACT survey with the information received in the pulp
and paper survey.  As proposed, mills are not required to include air
pollution control device details for boilers in the Part I Controls tab
since NCGs combusted by boilers have been demonstrated to be destroyed
in the boiler and are subsequently unaffected by the boiler’s APCD. 
Also as proposed, we are not collecting existing or new emissions test
data for boilers.  We are collecting limited data for pulp and paper
mill boilers instead of relying on the boiler MACT data set for
following reasons:  

The boiler data collected through the pulp and paper survey will be for
the 2009 base year consistent with the remainder of the pulp and paper
data collected (including the NEI data), as opposed to an earlier base
year for the Boiler MACT data set.

 

The limited fuel type and use (percent of MMBtu/hr capacity) data
requested in the pulp and paper survey would simplify emission
calculations (e.g., whereas the boiler MACT data included fuel use
ranges).

It would be a time-consuming exercise for EPA to attempt to match the
Boiler MACT equipment IDs with the NEI emission unit IDs provided in the
pulp and paper survey’s NEI update (and past experience with this type
of exercise suggests that it would be nearly impossible to match all of
the boilers). The pulp and paper survey should provide the matching
codes for pulp and paper boilers.

6.0	State-Only Permit Limits

Comment:  Commenters 0009 and 0011 noted that the Inventory tab of the
survey requests permit limits for each unit affected by the ICR.  The
two commenters agreed that each facility should report federally
enforceable limits, but recommended that state-only limits not be
included.  The commenters argued that state-only limits on criteria
pollutants and toxics derived from modeling analyses often do not
represent actual emissions (or in some cases even potential emissions)
and do not allow for “apples to apples” comparison of permit limits
across all sources.  As an example, the commenters stated that many
facilities “optimize” emissions of air toxics in their modeling
analyses up to a certain percentage of the state’s allowable ambient
levels, in order to gain a compliance margin and reduce the likelihood
of future modeling when emission factors change slightly.  The
commenters indicated that modeled rates included in permits as
state-only limits are often much higher than actual or potential
emissions and would not give EPA any meaningful information.  According
to the commenters, limits are often on a source-by-source basis and take
up several pages in the permit, so exclusion of these limits will reduce
the burden on facilities.

Response:  Permit limits are requested in Parts I and III of the
streamlined survey for the different emission units covered by each part
of the survey.  Mills may provide all permit limits including limits for
chemical recovery combustion sources in Part I if desired (and,
therefore, would not need to include these limits in Part III).  To
streamline the permit limit request, the permit limit columns were
removed from the proposed Inventory tab (which would have forced permit
limits to be entered in a format that lined up the NEI data) and placed
in new Part I and Part III Permit limits tabs.  

The EPA is not requesting that every single permit limit be included in
the survey response.  The instruction document for the proposed survey
narrowed the request for permit limits to concentration limits (ppmdv,
gr/dscf, etc.), percent reduction, mass per unit production (lb/ton of
material throughput, etc.) because these limits can be compared from
facility to facility independent of emission unit or mill capacity. 
Mass emission rate limits (lb/hr or tpy), which are typical of the
optimized state permit limits mentioned by the commenters, are not being
requested unless this is the only way in which limits are specified in
the permit.  The streamlined survey instruction documents for Parts I
and III contain this same language.  

In addition, we added a statement in the revised survey spreadsheets
stating that mills do not have to provide every limit for pollutants
with multiple limits in different units of measure.  Instead, mills are
referred to the survey instructions for a list of suggested units of
measure for permit limits.  The Part I and Part III survey instruction
documents (like the proposed instruction document) include a table of
suggested units of measure for permit limits (see Table 2 of each
instruction document).  The suggested units of measure included in Table
2 of the survey instructions were selected based on the units of measure
for Federal limits.  If respondents provide limits in the suggested
terms, then EPA can readily compare permit limits with the federal
limits in order to easily identify mill equipment with emission limits
more stringent than existing Federal standards.  The survey is further
set up to prevent rekeying of Federal emission limits that are restated
in the permit.  The survey instructions specify the following:

“Your permit may simply restate the emission limits contained in
Federal rules such as NESHAP subparts S and MM and NSPS subparts BB or
Db.  Some permits may list all of the compliance options contained in
the Federal rules as permit limits.  If this is the case, you do not
need to enter permit limits that are exactly the same as Federal NESHAP
or NSPS.  Instead, specify the applicable Federal rule (not the specific
numeric Federal limits) in Column F and skip the remaining permit
questions/columns in the table.”  

The instructions go on to specifically state, “Please do not enter
State lb/hr or tpy emission limits in addition to the Federal limits.”

Nevertheless, EPA is sensitive to the burden associated with completing
the request for permit limits.  In order to reduce this burden, we
included survey instructions to allow mills to submit a copy of their
permit with their survey response instead of completing the permit limit
section of the survey.  A mill-specific summary of permit limits or
compliance matrix would be acceptable as well. 

7.0	Potential Changes

Comment:  Commenters 0009 and 0011 noted that EPA is requesting
information on “potential changes.”  The two commenters argued that
this information is not appropriate and recommended that the tab be
removed entirely, for three reasons.

First the commenters noted that facilities are in a continuous cycle to
determine how to improve their operations and evaluate the impact of
potential regulations.  The commenters stated that, due to the
ever-changing market and regulatory climate, many of these potential
projects are either never seen to completion or are modified due to
changes in regulatory requirements.  The commenters argued that the
information requested is constantly changing, would be out of date by
the time EPA was finished analyzing the information, and would change
again based on the outcome of EPA’s analysis.

Second, according to the commenters, the request for information on any
potential changes within the next eight years aggravates the problems
described previously.  The commenters argued that it is mere speculation
to talk about what changes might take place over such a long time,
during which regulatory requirements likely will change several times,
let alone customer demand and requirements, process technology, and
products.  The commenters also noted that the proposed spreadsheet asks
for potential changes “in the next 8 years (e.g., between 2009 and
2017).”  According to the commenters, there is no apparent reason for
including changes in 2009 and 2010.  Additionally, the commenters noted
that not all of the questions contain a parenthetical qualifying “in
the next 8 years,” begging the question whether those portions are
supposed to address changes from 2011 to 2019.

Third, the commenters noted that the pulp and paper industry competes
globally with all companies searching for innovative solutions to
maintain and increase their place in the market. Therefore, the
commenters indicated that concepts to create new or improved products,
improve process efficacy and efficiency, and comply with regulations are
typically CBI, and companies generally would not be willing to share
such “potential projects” in a public forum.

Response:  We acknowledge the issues noted by the commenters with
respect to supplying information on potential changes.  The Potential
changes tab was originally included in the survey to provide information
useful for projecting new or modified source impacts.  However, in
response to the comment above, we have removed the tab from the survey.

Another method for projecting the number of new sources and
modifications would be to look at the number of new sources and
modifications that have occurred over the past 5 years and project
forward the same number of new sources and modifications for the
upcoming 5 years (e.g., the period over which new source impacts would
be calculated).  Other portions of the survey contain questions (e.g.,
equipment installation dates, reason for NSPS applicability, etc) that
will inform this type of projection. 

8.0	Technical Clarifications

8.1	Nominal Throughput Data

Comment:  Two commenters (0009 and 0011) indicated that the survey needs
to be consistent in its terminology. According to the two commenters,
data requests are inconsistent in requesting either design, potential or
actual data.  The commenters recommended that EPA request nominal
capacity data.

Response: In response to the comment, we have made the following
specific changes to the survey to improve the consistency in
terminology:

In the Part I Mill tab, noted in the instruction row that nominal
production capacity will suffice for market pulp and paper/paperboard
for 2009.  [Note:  The revised survey’s Mill tab was formerly the
information to the right of the orange line in Facility tab of the
proposed survey.]

In the Part I Equip detail tab, replaced the word “throughput” with
“capacity” for the causticizing area to request “Causticizing area
capacity (use lime kiln tons CaO/yr)” and noted in the instruction row
that lime kiln capacity is of interest.

In the Part I Paper prod tab, added the words “nominal production”
for the paper/paperboard production process to request “Nominal
production capacity, ADT/d” and noted in the instruction row that
nominal production capacity is of interest.

In the Part I survey instructions, we included example calculations to
aid in understanding of the capacity utilization questions.

In the Part III Equip detail tab:

Replaced the words “feed rate” with “capacity” for recovery
furnaces to request “Spent liquor solids capacity, million lb BLS/yr
(or million lb RLS/yr for sulfite)” and noted in the instruction row
that BLS/RLS feed rate capacity is of interest.

Replaced the words “feed rate” with “capacity” for semichemical
combustion units to request “Spent liquor solids capacity, million
lb/yr.”

Replaced the word “rate” with “capacity” for lime kilns to
request “Lime production capacity, ton/d.”

8.2	Emission Calculation Method Code

Comment:  Commenters 0009 and 0011 noted that in the Inventory tab, the
facility is asked to supply an emission calculation method code.  The
two commenters recommended the addition of another code for facilities
that utilize an emission factor from a similar source at another
location within the company or a published emission factor from a
similar source, called “Similar Source Emission Factor.”

Response:  There is no specific code applicable to the situation the
commenter describes.  However, there is an emissions comment column
available within the NEI framework.  Consistent with that, we have added
an “Emissions Comment” column in the NEI update next to the
“Emission Calculation Method Code for Revised Emissions” column to
allow mills to describe the emission factor source or emissions
calculation method when none of the Emission Calculation Method Codes
apply.  

9.0	Other Specific Comments

9.1	Equip Detail Tab

Comment:  Two commenters (0009 and 0011) offered the following comments
on the production line, equipment capacity, and process tank questions
included in the Equip Detail tab of the survey.

Regarding production lines, the two commenters noted that EPA is
requesting that the facility designate each pulping line and then
correlate the ancillary equipment to each line.  The commenters were
concerned that this approach will lead to confusion.  According to the
commenters, while some mills do have straight pulping lines with
dedicated evaporators, bleaching lines, and chemical recovery units,
these facilities are not the norm.  The commenters stated that equipment
at most mills is interrelated, and, except for the digesters, the
products from these units can then travel a variety of ways in the mill.
 For this reason, the commenters recommended that Column C in the Equip
Detail tab be eliminated.

Regarding equipment capacity units, the two commenters noted that, for
several units, EPA requests capacity or throughput data in air-dried
tons of pulp (ADTP).  The commenters indicated that for some units, like
evaporators, these units are not appropriate.  The commenters suggested
that EPA request data in units that are more representative (i.e.,
gallons per minute [gpm] of weak black liquor for evaporators).

The two commenters noted that EPA also requests information for process
tanks for various materials (pulp, black liquor, green liquor, white
liquor). According to the commenters, most facilities will track total
generation of these materials, but will not track the throughput of
individual tanks.  The commenters stated that many facilities do not use
production-based emission factors for process tanks, but instead use
lb/hr/tank emission factors published in NCASI technical bulletins. The
commenters also stated that some tanks can be used for different
materials. Therefore, the commenters suggested that EPA request only the
number of tanks that store each material.

Response:  Regarding the comment on production lines, we agree with the
commenters that some lines will not have dedicated equipment. 
Consequently, we have omitted the column referenced by the commenters
everywhere it appeared in the draft survey.

Regarding the comment on equipment capacity units, the units were
originally selected based on the units for emission factors available in
the literature (i.e., NCASI technical bulletins) for those equipment. 
However, in response to this comment, we have revised the Equip detail
tab (in Part I) to ask for evaporator capacity in units of “gal/min of
weak black liquor.”

In response to the comment on process tanks, we have revised the Equip
detail tab (in Part I) to eliminate the request for annual tank
throughput and are just requesting information on tank capacity.  We
need tank capacity (in gallons) to distinguish relative tank size (e.g.,
in case we develop subcategories) for any tanks not subject to subpart S
that are determined to have uncontrolled emissions of interest for RTR. 
The survey instructions for tank capacity limit the number of tanks to
be included.  We are most interested in identifying any tanks that are
not already covered by subpart S.

9.2	Kraft Condensates Tab

Comment:  Commenters 0009 and 0011 noted that the survey is constructed
so a mill can indicate that it treats its condensates for Subpart S
either in a steam stripper or in the biological treatment system.  The
two commenters noted that some facilities have both systems and use them
in combination to demonstrate compliance or have one as a back-up system
(e.g., when an operational problem with the stripper causes a decrease
in collection and/or treatment).  The commenters recommended that EPA
include additional columns to note if the facility has both systems and
if they are used in combination or as a back-up system.

Response:  We agree with the commenters’ recommendation and have
revised the Kraft Condensates tab to include a column for a description
if multiple methods are used to treat a condensate.  We also relabeled
the Table headers (formerly A, B, C, etc) with Roman numerals (I, II,
III, etc) to minimize confusion with spreadsheet column numbering.

9.3	PCC Tab

Comment:  Commenters 0009 and 0011 noted that EPA is requesting
information on facilities that route process exhaust to a PCC plant. 
The two commenters were unclear why this information is necessary since
these facilities are typically independently owned and operated under a
separate air permit.  The commenters stated that the PCC plant is not
part of the pulp and paper source category and requested that
information on these plants not be included in the survey.

Response:  The PCC tab is included in Part III of the revised pulp and
paper survey.  EPA is seeking to understand the interrelation of lime
kiln exhaust (or other process exhaust flow) to PCC plants and confirm
the permitting relationship between the two facilities.  We are aware of
at least one pulping facility with a PCC plant that appears to be
included in the same air permit.  As we evaluate lime kilns under the
pulp and paper subpart BB NSPS review and subpart MM RTR, it is
pertinent to know how the lime kiln exhaust is being used and whether
the exhaust being transferred to another facility is controlled or
uncontrolled prior to transfer.  We want to understand this arrangement
so that we can account for it in our rulemaking (e.g., with clarifying
rule language).  Nonetheless, we have limited the PCC questions in the
survey (e.g., removed requests for the address and latitude/longitude of
the PCC plant) to try to minimize the burden on survey respondents.

9.4	Emissions Test Data Tab

Comment:  Two commenters (0009 and 0011) agreed with EPA’s request
that each facility document the site-specific data available and include
a copy of the most recent test reports for review and entry into a
database.  The two commenters stated that this approach will allow for
EPA to extract the data for their use in a consistent manner and ensure
that the data submitted are reviewed by EPA.  The commenters hoped that
this will lead to a consistent handling of non-detects as well as a
review of the test methods, calculations, etc. to ensure that data are
properly documented and utilized.

Response:  We acknowledge the commenters’ statement.  We are
requesting copies of the most recent test reports for each mill to
ensure consistency in both the calculation of the emissions data and the
treatment of non-detect emissions data and to reduce the industry burden
of responding to this survey.  Parts I and III of the revised survey
request emissions test data for the different equipment and pollutants
listed in attachments to each of the instruction documents.  We added
acetaldehyde and formaldehyde to the list of HAPs for which emission
test data are requested (along with methanol which has been used as a
surrogate for these HAP) because acetaldehyde and formaldehyde may be of
particular interest for residual risk purposes.   

9.5	Segregation of CBI

Comment:  According to commenters 0009 and 0011, the legitimate need to
protect CBI from disclosure can often considerably complicate the
analysis of survey information, especially by interested members of the
public.  The two commenters urged EPA to segregate those portions of the
survey that are likely to contain information that will be claimed by
some or most companies as CBI from other portions of the survey, and to
have them reported in separate spreadsheets, to facilitate analysis of
information that likely will not be claimed as CBI by most or all mills.
 As an example, the commenters indicated that the tabs in the proposed
survey spreadsheet on pulp production, paper production, byproducts,
additives, and potential changes appear to be the portions most likely
to contain information that many or most companies will claim as CBI.

Response:  We believe the survey already provides sufficient mechanisms
for segregation of CBI-containing data.  Specifically, mills are
requested to segregate their CBI by shading red the CBI-containing cells
in the spreadsheets and submitting the CBI data separately, as specified
in section C2 of each survey instruction document.  Creating additional
separate spreadsheets within Parts I and III of the revised survey to
include the information cited by the commenters might lead to confusion
or encourage respondents to claim the entire spreadsheet as CBI, when
only a few columns may contain confidential information.  Separate
spreadsheets are already provided (as proposed) for submittal of cost
information which EPA recognizes has the potential to be considered
confidential by many mills.  Respondents are encouraged to consider
carefully which information they report is truly CBI that is protected
under the Clean Air Act.  See, e.g, 42 USC 7414(c) and 40 CFR 2.301.  It
should be noted that some capacity information is already reported
publicly (e.g., through data available for purchase from RISI). 
However, we understand that capacity utilization information may be
considered confidential.  The EPA will adhere to strict procedures in
place for protecting the confidentiality of information submitted with a
confidentiality claim.

10.0	CEMS Data

Comment:  Commenters 0009 and 0011 noted that, in the ICR, EPA requests
only 30 days of CEMS data.  The two commenters argued that this was not
a sufficient period of time to fully assess the variability of emissions
during startup and shutdown periods, malfunction periods and normal
operating periods.  The commenters suggested that EPA request a full
year of data to get a full representation of variability.  The
commenters stated that this is consistent with the approach that the
commenters and NCASI are proposing for the Boiler MACT carbon monoxide
(CO) standard.

Response:  Based on the comment, we have revised the survey to request a
full year of CEMS data, if available.  Data are requested for periods of
operation during calendar year 2009. 

In addition, we have made the following revisions to the CEMS data
request based on conversations with industry representatives:

Added a box to the survey instructions to provide guidance on how to
handle situations when CEMS data are not available in a form that can be
transferred into EPA’s P&P CEMS spreadsheet.xls (or when production
data are unavailable to be coupled with the CEMS data).  Alternative
formats for the CEMS data will suffice in these cases.

Made the hourly average opacity column “optional” since mills
typically demonstrate compliance based on 6-minute averages.  The column
was retained in case any mills have data in hourly average form (but not
6-minute averages).

Clarified (in the survey instructions) that mills are not required to
supply CEMS data for power boilers unless the boiler has a CEMS
installed to show compliance with a NESHAP subpart S (or subpart MM) or
NSPS subpart BB emission limit.     







	  PAGE   \* MERGEFORMAT  14 

	

