-------
reading opacity In this manner and will
propose this revision to Method 9 as soon
as this analysis is completed. The Agency
•elicits comments and recommendations
on the need for this additional revision to
Method 9 and would welcome any sug-
gestions particularly from  air pollution
control agencies on how we might make
Method 9 more responsive to the needs of
these agencies.
  These actions are effective on Novem-
ber 12,1974. The Agency finds good cause
exists  for  not publishing these actions
as a notice of proposed rulemaklng and
for making them  effective Immediately
upon  publication  for  the  following
reasons:
   (1)  Only minor amendments are  be-
ing made to the opacity standards which
were remanded.
   (2)  The  T7JB. Court of Appeals  for
the District of Columbia Instructed EPA
to complete the remand proceeding with
respect to the Portland cement plant
standards by November S, 1974.
   (3) Because opacity standards are the
subject of other litigation, it is necessary
to reach a final determination with re-
spect to the basic issues involving opacity
at this time In order to properly respond
to this issue with  respect to such other
litigation.
  These regulations are Issued under the
authority of sections 111 and 114 of the
Clean  Air Act.  as amended (42 UJS.C.
1857C-4 and 9).
  Dated: November 1.1974.
                    JOHN  QtTARLES,
                Acting Administrator.


   HDftAl MWSTH. VOL -»*, NO. tl*-

     -niUOAY. «OVEMM* It,
     IULES  AND 1EGULATKM4S

   Title 40  Protection of Environment

     CHAPTER I—ENVIRONMENTAL
         PROTECTION AGENCY
             [FRL 392-7)

 PART 6O—STANDARDS OF  PERFORM-
ANCE FOR  NEW STATIONARY SOURCES

     Five Categories of Sources In the
      Phosphate Fertilizer Industry
          OPACTTT STANDARDS

  Many  commentators  challenged  the
proposed  opacity  standards  on  the
grounds that EPA had shown no correla-
tion between  fluoride  emissions  and
plume  opacity, and that no data were
presented which showed that a violation
of the  proposed opacity standard would
indicate  simultaneous violation  of  the
proposed  fluoride  standard.  For  the
opacity standard to be used as  an  en-
forcement tool to indicate  possible  vio-
lation  of the fluoride standard,  such  a
correlation  must  be established.  The
Agency has  reevaluated the opacity  test
data and determined that the correlation
is insufficient  to  support  a standard.
Therefore, standards for visible emissions
for diammonium phosphate plants, triple
superphosphate plants,  and  granular
triple  superphosphate storage facilities
have been deleted. This action, however.
is not  meant  to set a precedent re-
garding promulgation of visible emission
standards. The situation which necessi-
tates this decision relates only to  fluoride
emissions. In the future, the Agency will
continue  to set opacity standards for
affected facilities where such standards
are desirable and warranted based on
test data.
   In place of the opacity standard, a  pro-
vision  has been added which requires an
owner or operator to monitor the total
pressure drop across an affected facility's
scrubbing system. This requirement will
provide an  affected facility's scrubbing
system. This requirement will provide for
a record  of the operating  conditions  of
the control  system, and will serve as an
effective method for monitoring  compli-
ance with the  fluoride standards.
      MOMITOUMO RMtnanmcrs
  Several comments' were received with
regard to the sections requiring a flow
measuring device which has an accuracy
of ± 6 percent over its operating range.
The commentators  felt that  this accu-
racy could not  be met and that the
capital and operating  costs outweighed
anticipated utility. First of all, "weigh-
belts" are common devices in the phos-
phate fertilizer industry as raw material
feeds  are  routinely  measured.  EPA
felt there would be no economic Impact
resulting from this  requirement because
plants  would have  normally Installed
weighing  devices anyway. Second, con-
tacts with the Industry led EPA to be-
lieve that the ± 6 percent accuracy re-
quirement would be easily met, and a
search of pertinent literature showed
that weighing devices with i 1 percent
accuracy  are commercially available.
  Effective date. In accordance with sec-
tion 111 of the Act, these regulations pre-
scribing  standards of  performance for
the selected stationary sources are effec-
tive on  August 4. 1975, and apply to
sources at which construction or modifi-
cation commenced after October 22.1974.
                 RUSSELL E. Tunr,
                      Administrator.

  JOT.Y 25. 1975.


    ftDOAl UOISTIt, VOL 40, NO. 1M-
      -WEDNESDAY, AUOUn 6. 1t7S
                                                    III-102

-------
 PART 60—STANDARDS OF PERFORM-
ANCE FOR NEW STATIONARY SOURCES

Emotion  Monitoring Requirement*  and
  Revisions   to  fltrfortMi     ~
  Methods
•nee  Testing
  On September 11. 1974 (39 FR 32852).
the  Environmental  Protection  Agency
(EPA) proposed revisions to 40 CFR Part
60. Standards of Performance for New
Stationary Sources, to establish  specific
requirements pertaining to continuous
emission monitoring system performance
specifications, operating procedures, data
These requirements would apply to new
and modified facilities  covered  under
Part 60.  but would not apply  to existing
facilities.
  Simultaneously  (39  FR 32871).  the
Agency proposed revisions to 40 CFR
Part 51.  Requirements for the Prepara-
tion. Adoption,  and Submlttal of Imple-
mentation  Plans, which would  require
States to revise their State Implementa-
tion Plans (SIP's) to  include legal en-
forceable procedures requiring  certain
specified stationary  sources to monitor
emissions on a continuous basis. These
requirements would apply to existing fa-
cilities, which are not covered under Part
60.
  Interested parties  participated in the
rulemaking by sending comments to EPA
A  total of 105 comment letters were re-
ceived on the proposed revisions  to Part
60 from monitoring equipment manufac-
turers, data processing equipment manu-
facturers, industrial users of monitoring
equipment, air pollution control agencies
Including State, local, and EPA regional
offices, other Federal agencies, and con-
sultants. Copies of the comment letters
received and a summary of the issues and
EPA's responses are available for Inspec-
tion and copying at the U.S. Environ-
mental Protection Agency, Public Infor-
mation Reference Unit, Room 2922 (EPA
Library). 401 M Street. S.W., Washing-
ton. D.C. In addition, copies of the issue
summary and EPA responses may be ob-
tained upon written request  from  the
EPA  Public Information Center (PM-
2l5i,  401 M Street,  8.W..  Washington.
D.C.  20460  (specify Public  Comment
Summary: Emission Monitoring Require-
ments). The comments have been care-
fully considered, additional Information
has  been collected  and assessed, and
where determined by Ihe Administrator
to be appropriate,  changes have been
made to the proposed regulations. These
changes are Incorporated In the regula-
tions promulgated herein.

             BACKGROUND
  At the time the regulations were pro-
posed (September  11,  1974). EPA had
promulgated  12 standards of perform-
ance for new stationary sources under
section.Ill  of the  Clean Air Act,  as
amended, four of which required  the af-
fected facilities to  install  and operate
systems which continuously monitor the
levels of pollutant emissions,  where the
technical feasibility exists using cur-
rently available continuous monitoring
technology, and where  the cost of  the
     RULES  AND REGULATIONS

systems Is reasonable. When the four
standards that require monitoring sys-
tems were promulgated. EPA had limited
knowledge about the operation of such
systems because only a few systems had
been installed:  thus,  the requirements
were specified  in general terms. EPA
Initiated a program to develop perform-
ance specifications and obtain informa-
tion on the operation  of continuous
monitoring systems. The program was
designed to assess the systems' accuracy.
reliability, costs, and problems related
to installation,  operation, maintenance.
and data  handling. The proposed regu-
lations (39 FR 32852) were based on the
results of this program.
  The  purpose  of regulations promul-
gated herein is to establish minimum
performance  specifications for cdntinu-
ous monitoring systems, minimum data
reduction  requirements, operating pro-
cedures, and  reporting requirements for
those affected facilities  required  to In-
stall  continuous  monitoring systems.
The specifications and  procedures are
designed to assure that the data obtained
from continuous monitoring systems will
be accurate and reliable and provide the
necessary  Information  for determining
whether an owner or operator is follow-
ing proper operation and  maintenance
procedures.
  SIGNIFICANT COMMENTS AND CHANCES
    MADE  To PROPOSED REGULATIONS
  Many of the comment letters received
by  EPA contained multiple  comments.
The most  significant comments and the
differences between  the proposed and
final regulations are discussed below.
  (1)  Subpart  A—General Provisions.
The greatest number of comments re-
ceived pertained to the methodology and
expense of obtaining and reporting con-
tinuous monitoring  system  emission
data. Both air pollution control agencies
and affected  users of monitoring equip-
ment presented  the view  that the pro-
posed  regulations  requiring  that  all
emission data be reported were  exces-
sive, and  that  reports  of only  excess
emissions and retention of all the data for
two  years on  the affected facility's
premises Is sufficient. Twenty-five com-
mentators suggested that the effective-
ness of the operation and maintenance of
an affected facility and its air pollution
control  system  could be  determined  by
reporting only excess emissions. Fifteen
others recommended deleting the report-
Ing requirements entirely.
  EPA' has reviewed these comments and
has contacted vendors of monitoring and
data acquisition equipment  for  addi-
tional information to more fully  assess
the Impact of  the proposed reporting
requirements. Consideration was also
given to the resources that would  be re-
quired  of  EPA  to enforce the proposed
requirement,  the costs  that would  be
Incurred by an  affected source, and the
effectiveness  of  the proposed require-
ment In comparison with a requirement
to report  only  excess  emissions. EPA
concluded   that  reporting  only  excess
emissions would  assure proper operation
and maintenance  of the air pollution
 control equipment and would result  In
 lower costs to the source and allow more
 effective use of EPA resources by elimi-
 nating  the need for handling and stor-
 ing large  amounts of data  Therefore.
 the regulation promulgated herein re-
 quires owners or operators to report only
 excess  emissions  and  to  maintain  a
 permanent record  of all  emission data
 for a period of two years.
   In addition, the proposed specification
 of  minimum data reduction procedures
 has been changed. Rather than requiring
 Integrated averages as proposed, the reg-
 ulations promulgated herein also spec-
 ify a method by which a minimum num-
 ber of data points may be used to com-
 pute average emission rates. For exam-
 ple, average opacity emissions over a six-
 mlnute period may be calculated from a
 minimum  of  24  data  points  equally
 spaced over each six-minute period. Any
 number of equally spaced data points in
 excess  of 24 or continuously integrated
 data may  also be used to compute six-
 minute  averages.  This specification  of
 minimum   computation   requirements
 combined with the requirement to report
 only excess  emissions  provides  source
 owners  and operators  with maximum
 flexibility to select from a wide choice of
 optional  data  reduction  procedures.
 Sources which monitor only opacity and
 which  Infrequently  experience  excess
 emissions  may  choose  to  utilize. strip
 chart recorders, with or without contin-
 uous six-minute  Integrators;  whereas
 sources monitoring two or more pollut-
 ants plus other parameters necessary  to
 convert to units of the emission  stand-
 ard may choose to utilize existing com-
 puters  or  electronic data processes In-
 corporated with the monitoring system.
 All data must be retained for two years,
 but only excess emissions need  be re-
 duced to units of the standard. However.
 in order to report excess emissions, ade-
 quate procedures must be utilized to In-
 sure that excess emissions are identified.
 Here again, certain sources with minimal
 excess  emissions can determine  excess
 emissions by review of strip charts, while
'sources with varying emission and ex-
 cess air rates will most  likely need  to
 reduce all data to units of the standard to
 Identify any excess emissions. The regu-
 lations promulgated herein allow the use
 of extractive, gaseous monitoring systems
 on a time sharing basis by Installing sam-
 pling probes at several locations, provided
 the minimum  number of  data  points
 (four per hour) are obtained.
  Several commentators stated that the
 averaging periods for reduction of moni-
 toring data, especially opacity, were too
 short and  would result in an excessive
 amount of data that must be reduced and
 recorded. EPA evaluated these comments
 and concluded that to be useful to source
 owners and operators as well as enforce-
 ment agencies, the averaging time for the
 continuous  monitoring  data should  be
 reasonably consistent with  the averag-
 ing time for the reference methods used
 during  performance tests. The data re-
 duction  requirements for opacity have
 been substantially reduced because the
 averaging period was changed from one
                                                 III-103

-------
                                             RULES AND  REGULATIONS
minute, which was proposed, to six min-
utes to be consistent with revisions made
to Method 9 (39 FR 39872).
  Numerous comments were received on
proposed ( 60.13 which resulted In several
changes. The proposed section has been
reorganized and  revised In several  re-
spects to accommodate the comments
and provide ch-.ity.  to more specifically
delineate the equipment subject to Per-
formance Specifications in Appendix B.
and to more specifically define require-
ments for equipment purchased  prior to
September 11,  1974.  The provisions In
I 60.13 are not intended to prevent  the
use of any equipment that can be demon-
strated  to  be  reliable  and  accurate;
therefore, the performance  of monitor-
Ing systems i£ specified in general terms
with minimal references to specific equip-
ment types. The  provisions  in § 60.13(1)
are included to  allow owners or operators
and equipment vendors to apply to  the
Administrator for approval  to use alter-
native equipment or procedures  when
equipment capable of producing accurate
results may not. be commercially avail-
able (e.g. condensed water  vapor Inter-
feres with  measurement  of  opacity),
when unusual circumstances may justify
less costly procedures, or when the owner
or  operator or  equipment  vendor may
•Imply prefer to use other equipment or
procedures that are consistent with his
current practices.
  Several  paragraphs  In  i 60.13  have
been changed on the basis  of the com-
ments received. In response to comments
that the monitor operating frequency re-
quirements did not consider periods when
the monitor Is inoperative  or undergo-
ing maintenance, calibration, and adjust-
ment, the operating frequency  require-
ments have been changed. Also  the fre-
quency of cycling requirement for opacity
monitors has been changed to  be con-
sistent with the  response time  require-
ment In  Performance  Specification   1,
which reflects the capability of commer-
cially available equipment.
  A second area  that received comment
concerns maintenance  performed upon
continuous  monitoring  systems.   Six
commentators noted that the proposed
regulation  requiring extensive retestlng
of continuous monitoring systems for all
minor failures  would discourage proper
maintenance of the  systems. Two  other
commentators noted the difficulty of de-
termining a general list of critical com-
ponents, the replacement of  which would
automatically require a retest of the sys-
tem.  Nevertheless,  it is EPA's  opinion
that some control must be  exercised to
Insure that a suitable monitoring system
is not rendered unsuitable by substantial
alteration or a lack of needed  mainte-
nance. Accordingly, the regulations pro-
mulgated herein  require that owners or
operators submit with the quarterly re-
port information on any repairs or modi-
fications made to the system during the
reporting period. Based  upon this Infor-
mation, the Administrator  may review
the status of the monitoring system with
the owner or operator and, if determined
to be necessary, require retesting of  the
continuous monitoring system (•).
   Several commentators noted that the
 proposed reporting requirements are un-
 necessary for affected facilities not re-
 quired to Install continuous  monitoring
^ystems. Consequently, the  regulations
 promulgated herein do not contain the
 requirements.
   Numerous comments  were received
 which indicated that some  monitoring
 systems may not be compatible with the
 proposed  test  procedures  and require-
 ments.  The comments were evaluated
 and, where appropriate, the  proposed
 test procedures and  requirements  were
 changed.  The  procedures  and require-
 ments promulgated herein are applicable
 to the majority  of acceptable systems:
 however, EPA recognizes that there may
 be  some  acceptable  systems  available
 now or in  the  future which could not
 meet the  requirements. Because of this,
 the regulations promulgated herein In-
 clude a provision which allows the Ad-
 ministrator to approve alternative testing
 procedures. Eleven commentators noted
 that adjustment of the monitoring in-
 struments may not be necessary as  a re-
 sult of dally zero and span checks. Ac-
 cordingly, the  regulations promulgated
 herein require  adjustments  only when
 applicable 24-hour drift limits are ex-
 ceeded. Pour commentators  stated that
 it Is not necessary to introduce calibra-
 tion gases near the probe tips. EPA has
 demonstrated in  fleld evaluations that
 this requirement is necessary in order to
 assure accurate  results; therefore, the
 requirement has been retained. The re-
 quirement enables detection of any dilu-
 tion or absorption of pollutant gas by the
 plumbing and conditioning systems prior
 to the  pollutant  gas entering the gas
 analyzer.
   Provisions have been  added to these
 regulations to require that the gas  mix-
 tures used for the daily calibration check
 of extractive continuous monitoring sys-
 tems be traceable to National Bureau of
 Standards  (NBS) reference gases.  Cali-
 bration  gases  used to conduct  system
 evaluations  under  Appendix  B  must
 either be  analyzed prior to use or shown
 to be traceable to NBS materials. This
 traceablllty requirement will assure the
 accuracy  of the calibration gas mixtures
 and the comparability of data from sys-
 tems at all locations. These  traceability
 requirements will not be applied when-
 ever the NBS materials are not available.
 A list of available NBS Standard Refer-
 ence Materials  may be obtained from the
 Office of Standard Reference Materials,
 Room B311. Chemistry Building,  Na-
 tional Bureau of Standards, Washington,
 D.C. 20234.
   Recertincatlon  of  the continued ac-
 curacy of the calibration gas mixtures Is
 also necessary  and should be performed
 at Intervals recommended by  the  cali-
 bration gas mixture manufacturer. The
 .NBS materials and calibration gas  mix-
 tures traceable to these materials should
 not be used after expiration of  their
 stated shelf-life. Manufacturers of cali-
 bration gas mixtures generally use NBS
 materials   for   traceability   purposes,
 therefore, these amendments to the reg-
ulations will  not Impose additional re-
quirements upon most manufacturers.
  (2)  Subpart  - D—Fossil-Fuel  Fired
Steam Generators. Eighteen commenta-
tors  had questions or remarks concern-
Ing the proposed revisions dealing with
fuel  analysis. The  evaluation of these
comments and discussions with coal sup-
pliers and electric utility companies led
the  Agency to conclude that the pro-
posed -provisions for fuel analysis are not
adequate or consistent with the  current
fuel  situation. An attempt was made to
revise the proposed provisions; however,
it became apparent  that  an in-depth
study would be necessary before mean-
ingful provisions could be developed. The
Agency has decided to promulgate all of
the regulations except those dealing with
fuel  analysis. The  fuel analysis provi-
sions of Subpart D have been reserved
in the regulations promulgated  herein.
The Agency has initiated a study to ob-
tain the necessary Information  on the
variability of sulfur content in fuels, and
the capability of fossil fuel fired steam
generators to  use  fuel analysis and
blending to prevent excess sulfur dioxide
emissions. The results of this study will
be used to determine whether fuel anal-
ysis  should  be allowed as a means of
measuring  excess emissions,  and if al-
lowed, what procedure  should  be re-
quired.  It  should be pointed out that
this  action does not afiect facilities which
use  flue gas desulfurizatlon as a means
of complying with  the  sulfur  dioxide
standard; these  facilities are still re-
quired  to  Install continuous emission
monitoring systems for  sulfur  dioxide.
Facilities which  use low sulfur fuel as a
means of complying with the sulfur di-
oxide standard  may use a  continuous
sulfur dioxide monitor or fuel analysis.
For  facilities that elect to use fuel anal-
ysis  procedures, fuels  are not required
to be sampled or analyzed  for prepara-
tion of reports of excess emissions until
the Agency finalizes the procedures and
requirements.
  Three   commentators  recommended
that carbon dioxide continuous monitor-
ing systems be allowed as an alternative
for oxygen monitoring for measurement
of the amount of diluents in flue gases
from steam   generators.  The  Agency
agrees with this recommendation  and has
Included a provision which allows the use
of carbon  dioxide monitors. This -pro-
vision allows  the use of pollutant moni-
tors that produce data on a wet  basis
without requiring additional equipment
or procedures for correction of data to a
dry basis. Where CO, or O, data  are not
collected on a consistent basis  (wet or
dry)  with the pollutant data, or where
oxygen is measured  on a wet basis, al-
ternative procedures to provide  correc-
tions for stack moisture and excess air
must be approved by the Administrator,
Similarly, use of a carbon  dioxide con-
tinuous monitoring system downstream
of a flue gas desulf urization system is not
permitted without  the Administrator's
prior approval due to  the potential for
absorption of CO, within the  control
device. It should be noted that when any
fuel  is fired directly in the stack gases
                                                    III-104

-------
                                             «ULES AND IEOULATIONS
for reheating, the T and F,  factors
promulgated herein must be prorated
based upon the total  heat Input of the
fuels fired within the facility regardless
of the locations of fuel firing. Therefore.
any facility using a flue gas desulfuriza-
Uon system may be limited to dry basis
monitoring  Instrumentation due to the
restrictions on use of a CO diluent moni-
tor unless water vapor Is also measured
subject to the Administrator's approval.
  Two commentators  requested  that an
additional factor (F •) be developed for
use with oxygen continuous monitoring
systems that measure flue gas diluents on
a wet basis. A factor of this  type was
evaluated by EPA. but is not being pro-
mulgated with  the regulations  herein.
The error in the accuracy of the factor
may exceed  ±5 percent without  addi-
tional measurements to correct  for va-
riations in flue gas moisture content due
to fluctuations In ambient humidity or
fuel moisture content. However. EPA will
approve installation of wet basis oxygen
systems on  a case-by-case basis If the
owner or operator will proposed use of
additional measurements and procedures
to control the accuracy of the P.. factor
within acceptable limits. Applications for
approval of such systems should include
the frequency  and type  of  additional
measurements proposed and the resulting
accuracy of the  F« factor under the ex-
tremes    of    operating    conditions
anticipated.
 •r One commentator stated that the pro-
posed requirements for recording heat
input are superfluous because this infor-
mation Is not needed to convert monitor-
Ing data to units of the applicable stand-
ard. EPA has reevaluated  this require-
ment and has determined that the con-
version of excess emissions into  units of
the standards  will be based  upon the
P factors and that measurement of the
rates of fuel firing will not be needed ex-
cept when combinations of fuels are fired.
Accordingly, the regulations promulgated
herein require such measurements only
when multiple fuels are fired.
  Thirteen commentators questioned the
rationale for the proposed Increased op-
erating temperature  of  the Method 5
sampling train for fossil-fuel-fired steam
generator  particulate  testing and the
basis  for raising rather than lowering
the temperature. A brief discussion of the
rationale behind this revision was pro-
vided In the preamble to  the proposed
regulations, and a more detailed discus-
sion is provided  here. Several factors are
of primary importance In developing the
data base for a standard of performance
and in specifying the reference  method
for use in conducting a performance test.
Including:
  a. The method used for data gathering
to  establish a  standard  must  be the
aame as. or must have a known relation-
ship to. the method subsequently estab-
lished as the reference method.
  b. The method should measure pollut-
ant emissions indicative of  the perform-
ance of the best systems of emission re-
duction. A method meeting this criterion
will not  necessarily measure  emissions
as they would exist after  dilution and
 cooling to ambient temperature and pres-
 sure, as would occur upon release to the
 atmosphere. As such, an emission factor
 obtained through use of such a method
 would, for example, not necessarily be of
. use in an ambient dispersion model. This
 seeming inconsistency results from the
 fact that standards of performance are
 intended to result In Installation of sys-
 tems of emission  reduction which  are
 consistent with best demonstrated tech-
 nology, considering cost. The Adminis-
 trator, in establishing such standards, is
 required to  Identify best  demonstrated
 technology and to develop  standards
 which reflect such technology. In order
 for these  standards  to  be meaningful.
 and for the required control technology
 to be predictable, the compliance meth-
'ods must  measure emissions which are
 indicative  of the performance  of such
 systems.
   c. The method should include sufficient
 detail as needed  to produce  consistent
 and reliable test results.
   EPA relies primarily upon Method 5
 for gathering a consistent data base  for
 particulate matter standards. Method 5
 meets the above criteria by providing de-
 tailed  sampling methodology  and  in-
 cludes an out-of-stack filter to facilitate
 temperature control. The latter is needed
 to define particulate matter on a com-
 mon basis since it is a function of tem-
 perature and is not an absolute quantity.
'If temperature is not controlled, and/or
 if the effect of temperature upon particu-
 late formation is unknown, the effect on
 an emission control limitation for partic-
 ulate matter may be  variable and un-
 predictable.
   Although selection of temperature can
 be varied from industry to industry, EPA
 specifies a nominal sampling tempera-
 ture of 120" C for most source categories
 subject to standards of  performance.
 Reasons for selection of 120° C  Include
 the following:
   a.  Filter temperature  must  be held
 above 100° C at sources where moist gas
 streams are present. Below 100° C, con-
 densation can occur with resultant plug-
 ging of filters and possible gas/liquid  re-
 actions. A temperature of 120" C allows
 for  expected  temperature  variation
 within the train, without dropping below
 100° C.
   b. Matter existing in particulate form
 at 120° C is indicative'of the perform-
 ance of the best particulate emission  re-
 duction systems for most industrial proc-
 esses. These Include systems of emission
 reduction that may involve-not only the
 final control device, but also the process
 and stack  gas conditioning systems.
   c. Adherence to one established tem-
 perature  (even though  some variation
 may be needed  for some source categor-
 ies) allows comparison of emissions from
 source category to source category. This
 limited standardization used in the de-
 velopment of standards of performance
 is a benefit to equipment vendors and to
 source owners by providing a consistent
 basis for comparing test results and pre-
 dicting control system performance. In
 comparison,   In-stack  filtration  takes
 place at stack temperature, which usually
 Is not constant from one source to the
 next. Since the temperature varies, in-
 stack filtration does not necessarily pro-
 vide a consistent definition of particulate
 matter and does not allow for compari-
 son of  various  systems of control. On
 these bases. Method 5  with a sampling
'filter temperature controlled at approxi-
 mately 120* C was promulgated as the
 applicable test method for new fossil-fuel
 fired steam generators.
   Subsequent to the promulgation of the
 standards  of  performance for steam
 generators, data became available  indi-
 cating that certain combustion products
 which do not exist as particulate matter
 at the elevated temperatures existing in
 steam generator stacks may be collected
 by Method 5 at lower temperatures (be-
 low 160°  C). Such material, existing In
 gaseous  form  at  stack   temperature,
 would not be controllable by emission re-
 duction systems involving  electrostatic
 preclpltators    (ESP).   Consequently,
 measurement of such condensible matter
 would not be indicative of the  control
 system performance. Studies conducted
 in the past two years have confirmed that
 such condensation can  occur. At sources
 where fuels containing 0.3 to 0.85 percent
 sulfur were burned, the incremental in-
 crease in particulate matter concentra-
 tion resulting from sampling at 120' C
 as compared to about 150°  C was found
 to be variable, ranging  from 0.001 to
 0.008 gr/scf. The variability Is not neces-
 sarily predictable, since total sulfur  oxide
 concentration, boiler design and opera-
 tion, and  fuel additives each appear to
 have a potential effect.  Based upon  these
 data, it Is concluded that the potential
 increase In particulate  concentration at
 sources meeting the standard of per-
 formance  for sulfur oxides is not a seri-
 ous problem in comparison with the par-
 ticulate standard which is approximately
 0.07 gr/scf. Nevertheless, to insure that
 an unusual case will not occur where a
 high concentration of  condensible  mat-
 ter, not controllable with an ESP. would
 prevent attainment of the particulate
 standard,  the sampling temperature al-
 lowed at fossil-fuel fired steam boilers is
 being raised to 160° C. Since  this  tem-
 perature is attainable at new steam gen-
 erator stacks, sampling at temperatures
 above 160° C would not yield results nec-
 essarily representative of the capabilities
 of the best systems of emission reduction.
 .  In evaluating  particulate  sampling
 techniques and the effect  of sampling
 temperature, particular  attention has
 also been given  to the possibility that
 SO, may  react In the front half of the
 Method 5 train to form particulate mat-
 ter.- Based upon a series of comprehen-
 sive tests involving both source and con-
 trolled environments, EPA has developed
 data that show such reactions do not oc-
 cur to a significant degree.
   Several control agencies commented on
 the increase in sampling  temperature
 and suggested that the need Is for  sam-
 pling at lower, not higher, temperatures.
 This is a  relevant comment and is one
 which must be considered in terms of the
 basis upon which  standards are estab-
 lished.
                                                      III-105

-------
  For existing boilers which are not sub-
ject to  this standard, the existence  of
higher stack  temperatures and/or the
use of higher sulfur fuels may result  in
significant condensation  and resultant
high  indicated  participate  concentra-
tions when sampling  is  conducted  at
120° C. At one r.oal fired steam generator
burning coal  containing  approximately
three percent sulfur, EPA  measurements
at 120° C showed an Increase of 0.05 gr/
dscf over an average of seven runs com-
pared to samples collected at approxi-
mately 150° C. It is believed that this In-
crease  resulted,  in  large part.  If not
totally,  from  SO]  condensation  which
would occur also when the stack emis-
sions are released into the atmosphere.
Therefore, where standards are based
upon emission reduction to achieve am-
bient air quality standards rather than
on  control technology (as is  the  case
with the standards promulgated herein),
a lower sampling temperature may  be
appropriate.
  Seven commentators questioned the
need for  traversing for  oxygen at  12
points within a duct during performance
tests. This requirement, which is being
revised  to  apply only  when  partlculate
sampling is performed  (no more than  12
points are required) is Included  to in-
sure that potential stratification  result-
Ing  from  air in-leakage  will not ad-
versely  affect  the accuracy of  the
paniculate test.
  Eight commentators stated  that the
requirement for  continuous  monitoring
of nitrogen oxides should  be deleted be-
cause only two air quality control re-
gions have ambient levels of nitrogen
dioxide that exceed the national ambient
air quality standard for nitrogen dioxide.
Standards of  performance Issued under-
section 111 of the Act are designed to re-
quire affected  facilities to  design and in-
stall the best systems of emission reduc-
tion (taking into account the cost of such
reduction). Continuous emission mon-
itoring  systems are required to  insure
that the  emission control systems are
operated and  maintained properly. Be-
cause of this, the Agency does not 'feel
that it is  appropriate to delete the  con-
tinuous emission monitoring system re-
quirements for nitrogen oxides; however.
In evaluating these comments the Agency
found  that some situations may exist
where the nitrogen oxides monitor is not
necessary  to  insure  proper operation
and maintenance. The quantity of nitro-
gen oxides emitted from certain types Of
furnaces Is considerably below the nitro-
gen oxides emission limitation. The low
emission level is  achieved through the
design of the  furnace  and does not re-
quire specific  operating  procedures  or
maintenance  on a continuous basis  to
keep the nitrogen oxides emissions below
the  applicable standard.  Therefore,  In
this situation,  a  continuous  emission
monitoring system for nitrogen oxides is
unnecessary.  The  regulations promul-
gated herein do  not require continuous
emission monitoring systems for nitrogen
oxides on facilities whose emissions are
30 percent or  more below the applicable
standard.
     AULES AND KEGULATIONS

  Three  commentators  requested  that
owners or operators of steam generators
be permitted to use NO, continuous mon-
itoring systems  capable of measuring
only nitric oxide (NO) since the amount
of nitrogen  dioxide  (NO:) in the flue
gases is comparatively small. The reg-
ulations proposed and those promulgated
herein allow use of such systems or any .
system meeting all of the requirements
of Performance  Specification 2  of Ap-
pendix B. A system that measures only
nitric oxide (NO) may meet these specifi-
cations including the relative accuracy
requirement (relative to the reference
method tests which measure NO + NO,)
without modification. However,  in the
Interests of maximizing  the accuracy of
the system and creating conditions favor-
able to acceptance of such systems (the
cost of  systems  measuring only NO is
less). the owner or operator may deter-
mine the proportion of  NO: relative to
NO in the flue gases and use a factor to
adjust the continuous monitoring system
emission data (e.g.  1.03 x  NO = NO,)
provided that  the factor is applied not
only to the performance  evaluation data,
but also applied consistently to all data
generated by the continuous monitoring
system thereafter. This procedure Is lim-
ited to facilities that have less than 10
percent NO,  (greater than 90 percent
NO) in order to not seriously Impair the
accuracy of the system due to NOi to NO
proportion fluctuations.
  Section 60.45(g) (1) has been reserved
for the future  specification of the excess
emissions for opacity that must be re-
ported. On November 12,  1974  (39 FR
39872).  the Administrator  promulgated
revisions to Subpart A, General Provi-
sions, pertaining to the opacity provi-
sions and to Reference Method 9. Visual
Determination of the Opacity of Emis-
sions  from  Stationary  Sources.  On
April 22,1975 (40 PR 17778), the Agency
issued a notice soliciting comments on
the  opacity provisions  and Reference
Method 9. The Agency Intends to  eval-
uate  the comments  received and make
any  appropriate revision to the  opacity
provisions and Reference Method  9. In
addition,  the Agency is evaluating the
opacity standards  for  fossil-fuel  fired
steam generators under -i 60.42(a) (2) to
determine if changes are needed because
of the new Reference Method 9. The pro-
visions on excess emissions for  opacity
will be issued after the Agency completes
Its evaluation of the opacity standard.
   (3)  Subpart O—Nitric  Acid  Plants.
Two commentators questioned the long-
term validity of the proposed conversion'
procedures for reducing data to units of
the standard.  They  suggested that the
conversion could  be accomplished  by
monitoring the flue gas  volumetric rate.
EPA reevaluated the proposed procedures
and  found that monitoring the flue gas
vplume would be the most direct  method
and would also be an accurate method of
converting monitoring data,  but would
require the Installation of an additional
continuous monitoring system. Although
this option is available and would be ac-
ceptable  subject to the Administrator's
approval, EPA does not believe that the
additional expense this method  (moni-
toring volumetric rate) would entail is
warranted. Since nitric acid plants, for
economic  and technical reasons, typi-
cally  operate  within  a fairly  narrow
range of  conversion efficiencies (90-96
percent) and tail gas diluents (2-5 per-
cent oxygen), the flue gas volumetric
rates are reasonably  proportional to the
acid production  rate.  The error that
would be Introduced  into the data from
the maximum variation of these param-
eters  is  approximately 15  percent  and
would usually be much less. It is expected
that the tail gas oxygen concentration
(an indication of the degree of tail gas
dilution) will be rigidly controlled at fa-
cilities using catalytic  converter control
equipment.  Accordingly, the  proposed
procedures for data conversion have been
retained due to the small  benefit that
would result from requiring additional
monitoring equipment.  Other procedures
may be approved by the Administrator
under 160.13(1).
   (4) Subpart H—Sulfurlc Acid Plants.
Two commentators stated that the pro-
posed procedure for conversion of moni-
toring data  to  units  of the standard
would  result  in large data  reduction
errors. EPA has evaluated  more closely
the operations of sulf uric acid plants and
agrees that the proposed procedure is in-
adequate. The proposed conversion pro-
cedure assumes  that the operating con-
ditions  of the affected facility  will re-
main approximately the same as during
the continuous monitoring system eval-
uation tests. For sulfuric acid plants this
assumption is  invalid. A  sulfuric  acid
plant is typically designed  to operate at
a  constant   volumetric   throughput
(scfm). Acid production rates are altered
by by-passing portions of the process air
around the furnace or  combustor to vary
the concentration of  the  gas entering
the converter. This  procedure produces
widely varying amounts of  tail gas dilu-
tion relative to the production rate. Ac-
cordingly, EPA  has  developed new con-
version procedures whereby the appro-
priate  conversion  factor  is  computed
from en analysis of the SO: concentra-
tion entering the converter. Air injection
plants must make additional corrections
for the diluent  air added.  Measurement
of the inlet 8O: is a  normal quality con-
trol procedure used by  most sulfuric acid
plants and does not represent an addi-
tional  cost  burden.  The Reich test or
other suitable procedures may be used.
   (5) Subpart J—Petroleum Refineries.
One commentator stated  that the re-
quirements for Installation  of continuous
monitoring systems for oxygen and fire-
box temperature  are  unnecessary  and
that installation of a flame detection de-
vice would be superior for process con-
trol purposes. Also, EPA  has obtained
data  which show no  identifiable rela-
tionship between furnace  temperature,
percent oxygen in the  flue  gas, and car-
bon monoxide emissions when the facil-
ity is operated'in compliance with the
applicable standard. Since firebox tem-»
perature and oxygen measurements may
not be  preferred by source owners and
operators for process control,  and no
                                                     III-106

-------
                                            tULES  AND REGULATIONS
known method Is available for transla-
tion of these measurements into quanti-
tative reports of excess carbon monoxide
emissions, this requirement appears to
be of little use to the affected  facilities
or to EPA. Accordingly, requirements for
Installation  of continuous  monitoring
systems  for  measurements  of firebox
temperature and oxygen are deleted from
the regulations.
  Since EPA has  not yet developed per-
formance specifications for carbon mon-
oxide or  hydrogen  sulflde  continuous
monitoring systems, the  type of equip-
ment that may be Installed by an owner
or operator in compliance with EPA re-
quirements Is  undefined.  Without con-
ducting performance evaluations of such
equipment, little reliance can be placed
upon the value of any data such systems
would generate. Therefore, the sections
of the regulation requiring these systems
are being reserved  until  EPA  proposes
performance specifications applicable to
HrS and CO  monitoring systems.  The
provisions of 5 60.l05(a) (3) do not apply
to an owner or operator electing to moni-
tor H;S. In that case, an H;S monitor
should not be installed until specific HrS
monitoring  requirements are  promul-
gated. At the time specifications are pro-
posed, all owners or operators who have
not entered into binding contractual ob-
ligations to purchase continuous moni-
toring equipment by [date of publication!
will be  required  to Install  a carbon
monoxide continuous monitoring system
and a hydrogen sulflde continuous moni-
toring system  (unless a  sulfur dioxide
continuous monitoring system has been
Installed) as applicable.
   Section 60.105(a)(2). which specifies
the excess emissions for capacity  that
must be reported, has been reserved for
the same reasons discussed under  fossil
fuel-fired steam generators.
   (6) Appendix B—Performance Speci-
fications. A large number of comments
were  received  in reference to specific
technical and  editorial changes needed
in the specifications. Each of these com-
ments  has  been reviewed  and several
changes in format and procedures have
been  made. These include adding  align-
ment  procedures for opacity  monitors
and more specific instructions for select-
ing a location for installing the monitor-
Ing equipment. Span requirements have
been specified so that commercially pro-
duced equipment may be standardized
where possible. The format of the  speci-
fications was simplified by redefining the
requirements in terms of percent opacity.
or oxygen, or carbon dioxide, or percent
of span. The proposed requirements were
in terms  of  percent of the emission
standard which is less convenient  or too
 vague  since reference to the emission
 standards would have  represented  a
 range  of pollutant  concentrations de-
 pending upon the amount of diluents (i.e.
 excess  air and water vapor)  that are
 present in the effluent. In order to cali-
 brate  gaseous monitors  In  terms of a
'specific concentration, the requirements
 were revised  to  delete reference to the
 emission standards.
   Four commentators noted that the ref-
•erence methods used to evaluate con-
tinuous monitoring system performance
may be less accurate than the systems
themselves.  Five  other  commentators
questioned the need for 27 nitrogen ox-
ides reference  method tests. The ac-
curacy specification for gaseous monitor-
ing systems was specified at 20 percent, a
value in excess  of the actual accuracy
of monitoring systems that provides tol-
erance for reference method Inaccuracy.
Commercially   available   monitoring
equipment has been evaluated using these
procedures and the combined errors (i.e.
relative accuracy)  in the reference meth-
ods  and the monitoring  systems have
been shown not to exceed 20 percent after
the  data  are averaged by the specified
procedures.
  Twenty commentators noted that the
cost, estimates contained in the proposal
did  not  fully reflect installation costs.
data reduction and recording costs, and
the  costs of evaluating the  continuous
monitoring  systems. As  a result, EPA
reevaluated the  cost analysis. For opac-
ity  monitoring  alone.  Investment costs
including data reduction equipment and
performance tests  are   approximately
$20.000. and annual operating costs are
approximately $8.500. The same location
on the stack used for conducting per-
formance tests with Reference Method 5
(participate) may  be used by Installing
a separate set of ports for the monitoring
system so that no additional expense for
access is required. For power plants that
are  required to  install opacity, nitrogen
oxides, sulfur dioxide, and diluent  (Or
or CO,)  monitoring systems, the  Invest-
ment cost is approximately $55,000, and
the operating cost is approximately $30,-
000. These are significant costs but are
not unreasonable in comparison to the
approximately  seven million dollar In-
vestment cost  for the smallest steam
-generation facility affected by these regu-
lations.
   Effective  date.  These regulations are
 promulgated under the authority of sec-
 tions  111. 114 and 301 (a) of the Clean
 Air Act as amended [42 U.S.C. 1857c-fl",
 1857c-9, and 1857g(a) ] and become ef-
 fective October  6. 1975.
   Dated: September 23,1975.
                    JOHN OVARIES.
                Acting Administrator.
    FEDHAl KOtSTR. VOL 40. NO. 1*4-

        -MONDAY, OCTOMI «,  <*7S
                                                          III-107

-------
                                              RULES AND REGULATIONS
    Title 40—Protection of Environment
      CHAPTER  I—ENVIRONMENTAL
          PROTECTION AGENCY
       SUBCHAPTER C—AIR  PROGRAMS
               |FRL 423-6)

 JART  51—REQUIREMENTS  FOR  THE
   PREPARATION.  ADOPTION  AND SUB
   MITTAL OF  IMPLEMENTATION  PLANS
 Cmission Monitoring of Stationary Sources
   On September  11,  1974.  the  Environ-
 mental Protection  Agency  (EPA)  pro-
 posed revisions to 40 CFR Part 51. Re-
 quirements for the  Preparation. Adop-
 tion,  and Subinittal of  Implementation
 Plans. EPA proposed to expand  5 51.19 to
 require  Stales to revise their State Im-
 plementation  Plans  (SIP's)  to include
 legally enforceable  procedures  requiring
 certain specified  categories of  existing
 Stationary sources to monitor emissions
 on a continuous basis. Revised SIP's sub-
 mitted by States in response to the pro-
 posed revisions to 40 CFR 51.19 would
• have  (1)  required  owners or operators
 of  specified   categories  of  stationary
 sources to install emission monitoring
 equipment within one year of  plan ap-
 proval.  (2)  specified the categories  of
 sources subject to the requirements. <3)
 identified for* each category of sources
 the pollutant(s)  which  must be moni-
 tored. (4) set forth performance specifi-
 cations for contfnuous emission monitor-
 ing instruments.  (5>  required that such
 instruments meet performance specifi-
 cations through on-site  testing by the
 owner or operator, and (6)  required that
 data  derived from  such monitoring  be
 summarized and  made available to the
 State on a quarterly basis.
   As  a minimum.  EPA  proposed  that
 States must adopt and implement legally
 enforceable procedures to require moni-
 toring of emissions for existing sources
 in the following  source  categories (but
 only for sources required to limit emis-
 sions to comply with an adopted regula-
 tion of the State Implementation Plan):
   (a) Coal-fired  steam  Generators  of
 more than 250 million BTU per hour heat
 input (opacity, sulfur dioxide, oxides  of
 nitrogen and oxygen);
   (b) Oil-fired steam generators of more
 than 250 million BTU per hour heat In-
 put (sulfur dioxide, oxides of  nitrogen
 and oxygen). An opacity monitor was re-
 quired only if an emission control device
 is needed to  meet  partlculate  emission
 regulations,  or if violations of  visible
 emission regulations are noted;
   (c)  Nitric  acid  plants  (oxides   of
 nitrogen);
   (d) Sulfuric  acid plants (sulfur di-
 oxide) ;  and
   (e)  Petroleum refineries' fluid catalytic
 cracking  unit  catalyst  regenerators
 (opacity).
   Simultaneously, the Agency  proposed
 similar continuous emission monitoring
 requirements for new sources for each  of
 the previously identified source categor-
 ies, subject to the provisions of federal
 New Source Performance Standards set
 forth in 40 CFR Part 60. Since many  of
 the technical aspects of the two proposals
 were similar, if not the  same,  the pro-
po^ed regulation* for Part 51 (|.c.._those
relating to SIP's and existing  sources i
included by .reloronrc many specific tech-
nical details set lorth In 40 CFR Part 60.
(39 FR 32852).
  At the time of the proposal of the con-
tinuous emission monitoring regulations
in the  FEDERAL REGISTER, the Agency  In-
vited comments on the  proposed rule-
making action. Many interested parties
submitted comments. Of the 76 comments
received. 35 were from  electric  utility
companies. 26 were from oil  refineries or
other industrial companies. 12 were from
governmental agencies, and 3 were from
manufacturers and/or suppliers of emis-
sion monitors.  No  comments  were  re-
ceived  from environmental groups. Fur-
ther, prior to the proposal of the regula-
tions in the FEDERAL REGISTER, the Agency
sought comments from various State and
local air pollution control agencies and
instrument  manufacturers. Copies  of
each of these  comments are  available
for public inspection at the EPA Freedom
of  Information  Center,  401 M Street,
S.W.,  Washington.  D.C.  20460.  These
comments have been considered, addi-
tional information collected and  assessed,
and where  determined by the Adminis-
trator  to be appropriate, revisions and
amendments have  been made in for-
mulating these regulations promulgated
herein.
  General Discussion  of Comments. In
general, the comments  received by  the
Agency tended to raise various objections
with specific portions of the regulations.
Some misinterpreted the proposed reg-
ulations, not  realizing  that   emission
monitoring under the proposal  was not
required unless a source was required to
comply with an adopted emission limita-
tion or sulfur in fuel limitation that was
part of an approved or promulgated State
Implementation Plan. Many questioned
the Agency's authority and  the need to
require sources to use continuous emis-
sion monitors.  Others stated  that the
proposed regulations were inflationary.
and by themselves could not reduce emis-
sions to the atmosphere  nor could they
improve air quality. A relatively  common
comment was that the benefits to be de-
rived from  the proposed emission moni-
toring  program were not commensurate
with the costs associated with the pur-
chase, installation, and operation of such
monitors. Many'stated that the proposed
regulations were not cost-effectively ap-
plied and they objected to all sources
within  an identified source category be-
ing required to monitor emissions, with-
out regard for other considerations. For
instance, some suggested that it was un-
necessary  to monitor emissions  from
steam generating plants  that may soon
be retired from operation, or steam gen-
erating boilers that are infrequently used
(such as for peaking and cycling opera-
tions)  or  for  those sources located in
areas of the nation which presently have
ambient concentrations better than na-
tional ambient air quality standards. This
latter comment was especially prevalent
in relation  to  the need  for continuous
emission monitors designed  to  measure
emissions of oxides of nitrogen. Further,
commentors generally  suggested that
state  and local control agencies, rather
than  EPA  should be  responsible  for
determining which sources should moni-
tor emissions. In this regard, the  com-
mcntors suggested that a determination
of the sources which should install con-
tinuous monitors should be made on a
case-by-case basjs. Almost all objected to
the data reporting requirements stating
that  the proposed requirement of sub-
mission of all collected data was excessive
ami burdensome. Comments from state
and local air pollution control agencies in
general were similar to those from  the
utility and industrial groups, but in addi-
tion, some indicated that the manpower
needed  to implement the programs  re-
quired by the proposed regulations was
not available.
  Rationale  for  Emission  Monitoring
Regulation. Presently, the Agency's reg-
ulations setting forth the requirements
for approvable SIP's  require  States to
have  legal  authority to require  owners
or operators of stationary sources to  in-
stall,  maintain, and use emission  moni-
toring  devices and  to  make periodic
reports  of  emission data to the  State
(40 CFR 5l.ll(a)(6». This requirement
was designed to partially implement the
requirements of Sections 110(a) (2) (F)
Mi) and (HI)  of the Clean Air Act. which
state  that  implementation  plans  must
provide  "requirements for  installation
of equipment by owners or operators of
stationary sources to monitor emissions
from  such sources", and "for periodic
reports  on  the nature and  amounts of
such  emissions".  However, the original
Implementation plan  requirements did
not require SIP's to contain legally en-
forceable procedures mandating contin-
uous emission monitoring and recording.
At  the  time the original requirements
were published, the Agency had accumu-
lated  little data on the availability and
reliability of  continuous monitoring de-
vices. The  Agency  believed  that the
state-of-the-art was such  that  it was
not prudent to require existing sources
to install such devices.
  Since that time, much work has been
done  by the Agency and others to field
test and  compare various  continuous
emission monitors. As a result of this
work, the Agency now believes that for
certain  sources, performance  specifica-
tions  for accuracy, reliability and dura-
bility can be established  for continuous
emission monitors of oxygen,  carbon
dioxide,  sulfur dioxide,  and  oxides  of
nitrogen and for the continuous meas-
urement of opacity. Accordingly,  it is
the Administrator's judgment that Sec-
tions  110(a)(2)(F) (ID and  (ill)  should
now be more fully implemented.
  The  Administrator believes that  a
sound program of continuous emission
monitoring and reporting wUl play  an
Important role in the effort  to attain
and maintain national standards. At the
present time, control agencies rely  upon
infrequent  manual  source   tests and
periodic  field  Inspections   to provide
much of the  enforcement information
necessary to  ascertain  compliance  of
sources  with adopted regulations. Man-
ual source tests are generally performed
on a relatively infrequent basis, such as
                               FtDHAt UOISTII. VOL. 40. NO.  1»4—MONDAY, OCTOKI 4, WS
                                                        III-108

-------
                                              RULES AND REGULATIONS
 once per year, and In some case*, affected
 sources probably have never been tested.
 Manual stark  tests are generally  per-
 formed under  optimum  operating  con-
 ditions, and as such, do not reflect the
 lull-time  emission   conditions from  a
 source.  Emissions continually vary  with
 fuel firing rates, process material  feed
 rates and  various other operating condi-
 tions. Since manual stack tests are  only
 conducted for a relatively short  period
 of time (e.g.. one to three hours*,  they
 cannot be representative of all operating
 conditions.  Further,  frequent manual
 stack  tests (such  as  conducted on  a
 quarterly  or more  frequent  basis'   are
 costly and may be more expensive than
 continuous monitors that provide much
 more  information.   State  Agency   en-
 forcement  by  field Inspection  is  also
 sporadic, with only occasional Inspection
 of certain sources, mainly  for  visible
 emission enforcement.
   Continuous emission  monitoring  and
 recording  systems,  on the other hand,
 can provide a continuous record of emis-
 sions under all operating conditions. The
 continuous emission monitor  Is a good
 indicator  of whether a source Is using
 good operating and maintenance prac-
 tices  to minimize emissions  to the  at-
 mosphere  and can  also provide a valu-
 able record to Indicate the performance
 of a source in complying with applicable
 emission control  regulations.  Addition-
 ally, under certain  instances, the data
 from continuous monitors  may be  suf-
 ficient evidence to issue a notice of  vio-
 lation.  The continuous  emission  record
 can also be utilized to  signal a plant
 upset or equipment  malfunction so  that
 the plant  operator  can take  corrective
 action to reduce emissions.  Use of emis-
 sion monitors can therefore provide  val-
 uable Information to-mtnimize emissions
 to the  atmosphere  and  to assure that
 full-time control  efforts, such as good
 maintenance  and operating conditions,
 are being utilized by  source operators.
   The,Agency believes that It is necessary
 to establish national minimum require-
 ments for emission monitors for specified
 sources rather than  allow States to de-
 termine  on a case-by-case basis the spe-
 cific sources which need to  continuously
 monitor emissions. The categories  speci-
 fied in the regulations represent very  sig-
 nificant  sources of emissions to the at-
 mosphere.  States in  developing  SIP's
 have generally adopted  control regula-
 tions to  minimize emissions from these
sources. Where such regulations exist, the
 Agency believes that  continuous emission
monitors are necessary to provide Infor-
mation that may be used to provide an
indication of source compliance. Further.
 It  Is believed that  if the  selection  of
 •ources on  a case-by-case basis were  left
 to the States,  that  some States would
 probably  not undertake an  adequate
 emission  monitoring  program.   Some
 State Agencies who  commented on  the
 proposed  regulations  questioned   the
 •tete-of-the-art of emission monitoring
 ww stated their opinion that the pro-
 P«««d   requirement* were   premature
 TlwrtloTft.  K  u the  Administrator's
           (bat. In order to assure  an
 adequate  nationwide  emission  monl-
 torini: procnirn. minimum emission mon-
 itoring requirements must be established.
   Thr source categories  affected by the
 regulations  were selected because  they
 are significant sources of emissions and
 because the  Agency's work at the time of
 the proposal of these regulations In the
 field of continuous omission  monitoring
 evaluation focused almost exclusively on
 these source categories.  The Agency Is
 continuing to develop data on monitoring
 devices  for additional source categories.
 It is EPA's Intent to expand the minimum
 continuous emission monitoring require-
 ment', from  time to time when the eco-
 nomic and  technological feasibility  of
 continuous  monitoring  equipment  is
 demonstrated and where such monitor-
 ing is deemed appropriate for other sig-
 nificant source categories.
   Discussion o/ Major Comments. Many
 eommentors discussed  the  various cost
 aspects of the proposed regulations, spe-
 cifically stating  that  the costs  of con-
 tinuous monitors were excessive  and In-
 flationary. A total of 47 commentors ex-
 pressed concern for the cost and/or cost
 effectiveness  of  continuous  monitors.
 Further, the Agency's cost estimates for
 purchasing  and  installing  monitoring
 systems and the costs for data reduction
 and reporting were questioned. In many
 cases, sources provided cost estimates for
 installation and operation of  continuous
 monitors considerably  in excess of  the
 cost estimates provided by the Agency.
   In response to these comments, a fur-
 ther review was undertaken by the Agen-
 cy  to assess the cost impact of the regu-
 lations. Three conclusions resulted from
 this review. First. It was determined that
 the cost ranges of the various emission
 monitoring  systems provided  by  the
 Agency are generally accurate  for  new
 sources.  Discussions  with   equipment
 manufacturers and suppliers confirmed
 this cost information. Approximate  in-
 vestment costs, which  include the  cost
 of the emission monitor. Installation cost
 at a new facility, recorder, performance
 testing, data reporting systems and asso-
 ciated engineering costs are as follows:
 for  opacity.  $20.000; for  sulfur  dioxide
 and oxygen  or  oxides  of nitrogen  and
 oxygen,  $30.000:  and for a source that
 monitors opacity, oxides of nitrogen, sul-
 fur dioxide and oxygen. $55.000  Annual
 operating costs, which  include data re-
 duction and  report preparation,  system
 operation,  maintenance, utilities, taxes,
 insurance and annualized capital  costs
 at 10? :or 8 years arc: $8.500: $16.000:
 and $30.000  respectively  for  the cases
 described above.(l)
  Secondly,  the  cost  review Indicated
 that the cost of Installation of emission
 monitors for existing  sources could  be
 considerably higher than for new sources
 because of the difficulties in providing
 access to a sampling location that ran
 provide a representative sample of emis-
sions. The cost estimates provided by the
Agency In the proposal were specifically
 developed  for  new  sources  whose  in-
stallation costs are relatively stable since
provisions for monitoring equipment can
be incorporated at the time of plant de-
sign. This feature is not available for ex-
 isting sources, hence higher costs gei
 rrally result.  Actual costs of Installalir
 at existing sources may  vary from 01
 to five times the cost of normal installs
 tion at  new sources, and in some casi
 even higher costs can result. For exam
 pic. discussions with instrument suppli
 ers indicate that a typical cost of instal
 latlon of an opacity monitor on an exist
 Ing source may be two to three times tlv
 purchase price of the monitor. Difficul
 tics also exist for Installation of gaseou
 monitors nt existing sources.
   It should be noted that these instal In
 tion costs Include material costs for seal
 folding, ladders,  sampling  ports  an'
 other items necessary to provide acces
 to a lo:atlon  where source emissions cai
 be measured. It  is the Agency's oplnio:
 that such  costs cannot be solely attrib
 uted to these continuous emission moni
 toring  regulations.  Access  to samplini
 locations Is generally necessary to dc
 termlne compliance with applicable stat>
 or local emission limitations  by routln<
 manual  stack testing methods. There-
 fore, costs of providing access to a rep-
 resentative sampling location  are more
 directly  attributed to the cost of com-
 pliance  with, adopted  emission  limita-
 tions, than with  these continuous emis-
 sion monitoring  regulations.
   Lastly, the  review of cost Information
 Indicated that a  numb;r of commentors
 misinterpreted the  extent  of  the pro-
 posed regulations, thereby providing cost
 estimates for continuous monitors which
 were not required.  Specifically, all com-
 mentors did not  recognize that the pro-
 posed regulations required emission mon-
 itoring for a source only If an applicable
 State or local emission limitation of an
 approved SIP affected such a source. For
 example, if the  approved SIP did not
 contain an adopted control regulation to
 limit oxides  of  nitrogen from  steam-
 generating, fossil fuel-fired boilers of n
 capacity In excess of 250 million BTU per
 hour heat  Input, then such source need
 not monitor  oxides  of  nitrogen emis-
 sions. Further, some utility industry com-
 mentorn included the costs of continuous
 emission monitors for sulfur dioxide. The
 proposed regulations, however, generally
 allowed the use of fuel analysis by speci-
 fied ASTM procedures as an alternative
 which, in most cases. Is  less expensive
 than continuous monitoring. Finally, the
 proposed regulations required  the con-
 tinuous  monitoring of oxygen  in  the
 exhaust  gas  only  if  the source must
 otherwise continuously monitor oxides of
 nitrogen or sulfur  dioxide.  Oxygen In-
 formation is used solely to provide a cor-
 rection for  excess air when converting
 the measurements of gaseous pollutants
 concentrations in the exhaust gas stream
 to units  of an applicable  emission limi-
 tation. Some commentors did not recog-
 nize this point 'which WAS not specifical-
 ly stated in the  proposed regulations)
 and provided  cost estimates for  oxygen
 monitors when thev were not required by
 the proposed regulations.
  While  not all commentors' cost esti-
 mates were  correct, for various reasons
noted above. It is clear that the costs
 associated   with   implementing  these
emission monitoring regulations are sig-
                              HDfl^l If Omit, VOL  40, NO. 1*4—MONDAY. OCTOIII *,  WS
                                                 III-109

-------
                                             RULES AND REGULATIONS
nlflcant  The  Administrator,  however.
believes that the benefits to be derived
from emission monitoring, are such that
the costs are not unreasonable. The Ad-
ministrator does,  however, agree with
many commentors that the proposed reg-
ulations, in somf cases, were not applied
cost-effectively and. as such, the regula-
tions promulgated  herein  have  been
modified to provide  exemptions to cer-
tain sources from these minimum  re-
quirements.
  One comment from another Federal
Agency  concerned the time period that
emissions are to be  averaged when re-
porting excess emissions Specifically, the
commentor assumed  that  the emission
control   regulations   that  have  been
adopted by State and local agencies were
generally designed to attain annual am-
bient air quality standards. As such, the
commentor pointed out that short-term
emission levels in excess of the adopted
emission standard should be acceptable
for reasonable periods of time.
  The Administrator does not agree with
this rationale for the following reasons.
First, it is  not universally true that  an-
nual Ambient standards were the design
basis of emission control regulations. In
many cases,  reductions to  attain short-
term standards  require  more control
than do annual standards.  Even if  the
regulations were  based  upon annual
standards,  allowing  excess emissions of
the adopted  emission control  regulation
on  a short-term basis could cause non-
compliance with annual standards. More
importantly, however, a policy of legally
allowing excesses of adopted control reg-
ulations would in effect make the current
emission limitation unenforceable. If the
suggestion  were implemented, a question
would arise as to what Is the maximum
emission level that would not be consid-
ered an excess to the adopted regulation.
The purpose of the adopted emission lim-
itation  was to establish the acceptable
emission level. Allowing emissions in ex-
cess of  that adopted level would  cause
confusion,  ambiguity, and in many cases
could result  in  an unenforceable situa-
tion. 'Hence the Administrator does not
concur with the commentor's suggestion.
  Modifications to the  Proposed Regu-
lations.  The modification  to  the  regu-
lations  which has the  most significant
impact involves the  monitoring require-
ments for  oxides  of nitrogen at fossil
fuel-fired steam generating boilers  and
at nitric acid plants. Many commentors
correctly noted  that the Agency in  the
past (June 8, 1973. 38 FR 15174) had in-
dicated  that the need  for many  emis-
sion control regulations for  oxides of
nitrogen were  based  upon  erroneous
data. Such a statement was made after
• detailed laboratory analysis of the ref-
erence  ambient measurement method
for nitrogen dioxide revealed the method
to   give   false   measurements.   The
sampling technique  generally  indicated
concentrations  of   nitrogen   dioxide
higher  than  actually  existed in  the
atmosphere.  Since many control  agen-
cies prior  to that  announcement had
adopted emission regulations  that were
determined to  be needed based  upon
these erroneous data, and since new data.
collected by  other  measurement tech-
niques, indicated  that in most areas of
the nation such control regulations were
not necessary to satisfy the requirements
of the SIP. the  Agency suggested that
States  consider   the   withdrawal  of
adopted control regulations for the con-
trol of oxides of nitrogen from their SIP's
(May 8.  1974. 39 FR 16344). In many
States, control agencies  have not taken
action to remove these regulations from
the SIP. Hence, the commentors pointed
out that the proposed regulations to re-
quire continuous  emission  monitors  on
sources  affected  by such regulations is
generally unnecessary.
  Because  of  the  unique  situation in-
volving oxides of nitrogen control regu-
lations,  the  Administrator  has deter-
mined that the proposed regulations to
continuously monitor oxides of nitrogen
emissions may place an undue burden on
source operators, at least from a stand-
point of EPA specifying minimum moni-
toring requirements.  The  continuous
emission monitoring  requirements for
such sources therefore have been modi-
fied. The  final regulations  require the
continuous   emission   monitoring  of
oxides of nitrogen only for those sources
in Air Quality Control Regions < AQCR's)
where the Administrator has specifically
determined  that a  control strategy for
nitrogen dioxide  is necessary.  At the
present  time such control  strategies are
required only  for the Metropolitan Los
Anceles Intrastatc  and  the  Metropoli-
tan Chicago Interstate AQCR's.
  It should be noted that a recent com-
pilation of  valid nitrogen  dioxide air
quality data suggests that approximately
14 of the other 245 AQCR's in the nation
may need to develop a  control strategy
for nitrogen dioxide. These AQCR's are
presently being evaluated by the Agency.
If any additional AQCR's  are identified
as needing a control strategy for nitro-
gen  dioxide at that time,  or any time
subsequent to this promulgation, then
States in  which those AQCR's  are lo-
cated must also  revise their SIP's to
require continuous emission monitoring
for  oxides  of nitrogen  for specified
sources. Further, it should be noted that
the regulations promulgated today are
minimum  requirements, so that States.
if they believe the  control of oxides of
nitrogen from sources is necessary may,
as they deem appropriate,  expand the
continuous emission monitoring require-
ments to apply to additional sources not
affected by these minimum requirements.
  Other modifications  to  the proposed
regulation  resulted from  various com-
ments. A number of commentors noted
that the proposed regulations included
some sources whose emission impact on
air quality was relatively minor. Specifi-
cally, they noted  that  fossil fuel-fired
steam generating units  that were used
solely for  peaking and cycling purposes
should be exempt from the proposed reg-
ulations. Similarly, some suggested that
smaller  sized units, particularly steam -
generating unite less than 2.500 million
BTU per hour heat input,  should also
be exempted.  Others  pointed out that
units soon to be retired from operation
should not be  required to install con-
tinuous monitoring  devices  and that
sources located In areas of the  nation
that already have air quality better than
the national standards should be relieved
of the required monitoring and reporting
requirements. The Agency has considered
these comments and has made the fol-
lowing Judgments.
  In relation to  fossil fuel-fired  steam
generating units,  the Agency has  deter-
mined  that such units that have an an-
nual boiler capacity factor of 30%  or less
as currently defined by the Federal Power
Commission shall be exempt  from the
minimum requirements for monitoring
and reporting. Industrial boilers used at
less than 307r of their annual capacity,
upon demonstration  to the State, may
also be granted an exemption from these
monitoring requirements. The rationale
for this exemption is  based upon the fact
that all generating units do not produce
power at their full capacity at all times.
There  are three major classifications of
power  plants  based  on the degree to
which  their  rated capacity is utilized on
an annual basis.  Baseload units are de-
signed to run at near full capacity  almost
continuously. Peaking unite are operated
to supply electricity during  periods of
maximum system demand. Unite  which
are  operated  for intermediate service
between the extremes of baseload and
peaking are termed cycling unite.
  Generally accepted  definitions term
units generating  60  percent or more of
their annual capacity as baseload. those
generating less than  20 percent as peak-
ing and those between 20 and 60 percent
as cycling. In general, peaking unite are
older,  smaller, of lower efficiency, and
more costly to operate than base load or
cycling unite. Cycling unite are also gen-
erally  older, smaller and  less efficient
than base load unite. Since the expected
life of peaking unite is relatively short
and total emissions from such unite are
small,  the benefits gained  by  installing
monitoring  instruments are  small in'
comparison  to the cost of such  equip-
ment. For cycling units, the question of
cost-effectiveness is more difficult to as-
certain. The unite at the upper  end of
the capacity factor range  Hence, the
final regulations do not affect any boiler
that has an annual boiler capacity factor
of less  than 30%. Monitoring require-
ments  will thus be more cost effectively
applied to the newer, larger, and more
efficient  units  that  burn  a  relatively
larner portion of the  total fuel supply.
   Some commentors noted that the age
of the facility should be considered in
relation to whether a source need com-
                              HDIIAl IEOISTEI, VOt. 40, NO. It4—MONDAY, OCTOBER 6, WS
                                                   III-110

-------
                                              RULES AND  REGULATIONS
 ply  with  the proposed regulations. For
 fossil fuel-fired steflm generating units.
 the  exemption  relating to  the  annual
 boiler capacity  fnctor  previously  dis-
 cussed should generally provide relief for
 older units. It is  appropriate, however.
 that the  age of the facility  be  consid-
 ered for other categories of sources af-
 fected by the proposed regulations. As
 such, the  final regulations allow thnt any
 source that  Is  scheduled  to be  retired
 within five years of the Inclusion of mon-
 itoring requirements for the source in
 Appendix P  need not comply with the
 minimum emission  monitoring  require-
 ments promulgated herein.  In the Ad-
 ministrator's  Judgment,  the selection of
 five  years as the  allowable  period for
 this exemption  provides reasonable re-
 lief  for those units that will shortly be
 retired. However. It maintains full re-
 quirements on many older units with a.
 number of years of service  remaining.
 In general, older units operate less effi-
 ciently and are  less well controlled than
 newer units so that emission monitoring
 Is generally useful.  The exemption pro-
 vided in the  final regulations effectively
 allows such retirees slightly more than a
 two-year  period of relief, since the sched-
 ule of implementation of the regulations
 would generally require the installation
 of  emission   monitors  by  early  1978.
 States must  submit, for EPA approval.
 the  procedures  they will implement to
 use  this  provision. States are  advised
 that such exemptions should only be pro-
 vided where  a bona fide intent to rease
 operations has been clearly  established.
 In cases  where such sources postpone
 retirement. States shall have established
 procedures to require  such sources to
 monitor and  report emissions. In  this re-
 gard, it  should be  noted  that  Section
 113 f2) of  the Act provides that any
 person who falsifies or misrepresents a
 record, report or other document filed or
 required under the Act shall, upon con-
 viction, be subject to fine or imprison-
 ment, or  both.
  A  further modification to the proposed
 regulations affects the minimum size of
 the units within each of the source cate-
 gories to which emission monitoring and
 reporting  shall be required. As suggested
 by many commentors. the Agency has in-
 vestigated the cost  effectiveness of re-
 quiring all units  within the identified
 source categories to install emission mon-
 itors. Each  pollutant for  each  source
 category identified in the proposed reg-
 ulations was evaluated.  For fossil  fuel-
 fired steam generating units, the  pro-
 posal required compliance for all boilers
 with 250 million BTU per hour heat In-
put, or greater. For opacity, the proposed
regulationsvr.equlred emission monitoring
 for all coal-fired units, while only those
 oil-fired units that had been observed as
violators of visible emission  regulations
 or must use an emission control device to
 meet particular matter regulations were
 required  to install  such devices. Gas-
 fired units were exempted  by the  pro-
 posed regulations.
  After Investigating  the  particulate
emission potential of these sources. It has
 been determined that no modification In
 the size limitation for boilers In relation
 to opacity is warranted. The  rationale
 for this  Judgment is  that the smallcr-
 sizod units affected by the proposed reg-
 ulation tend to be less efficiently oper-
 ated or controlled for particulate matter
 than are the larger-sized units. In fact.
 smaller units generally tend to emit more
 particulate  emissions  on an equivalent
 fuel basis than  lamer-sized units.  '2>
 Because  of the potential of opacity regu-
 lation violations, no modifications have
 been made  to  the regulations as to the
 size of  steam generating boilers that
 must measure  opacity.
  Emissions of oxides of nitrogen from
 boilers are a function of the temperature
 in the combustion chamber and the cool-
 Ing of the  combustion products.  Emis-
 sions vary considerably with  the si7.e and
 the type  of  unit.  In general, the larner
 units produce  more oxides  of nitroRcn
 emissions. The Agency  therefore finds
 that the minimum size of a unit affected
 by the final regulatioas can be Increased
 from 250 to  1.000 million BTU per hour
 heat  input,  without significantly reduc-
 ing the total emissions of oxides of nitro-
 gen that would be affected by monitoring
 and reporting requirements. Such a mod-
 ification would,  have the effect of exempt-
 ing  approximately 567r  of  the  boilers
 over 250 million BTU per hour heat input
 capacity, on a national basis, while main-
 taining emission monitoring  and report-
 ing requirements for approximately 78r.r
 of the potential oxides of nitroccn emis-
 sions from such sources.'2 >  Further. In
 the 2 AQCR's  where the Administrator
 has  specifically  called  for a  control
 strategy  for nitrogen dioxide, the boilers
 affected by the  regulation constitute 50T
 of the steam generators greater than 250
 million  BTU per  hour heat Input,  yet
 they  emit 80 <5 of the nitrogen oxides
 from such  steam  generators  in these
 2 AQCR's.(2)
  Also, certain types of boilers or burn-
 ers, due  to  their design characteristics.
 may on a regular basis  attain emission
 levels of  oxides of nitrogen  well  below
 the emission limitations of the applica-
 ble plan. The regulations have been re-
 vised to  allow  exemption  from  the
 requirements  for  installing  emission
 monitoring and recording equipment for
 oxides of nitrogen when a facility is
 shown during  performance tests to  op-
 erate with oxides of nitrogen  emission
 levels 30% or  more below the emission
 limitation of  the  applicable  plan.  It
 should be noted that this provision  ap-
 plies solely  to  oxides of nitrogen emis-
 sions  rather  than other pollutant emis-
 sions, since oxides of nitrogen emissions
are more directly related to boiler  de-
sign  characteristics   than   are   other
 pollutants.
  Similar evaluations were  made  for
nitric acid plants, sulfuric  acid  plants
 and catalytic cracking unit catalyst re-
generators at petroleum refineries. For
each of these Industries it was found that
modifications to the proposed  regulations
could be  made  to increase the minimum
size of the units affected by the proposed
regulations   without   significantly  de-
creasing  the total emissions of various
 pollutants  that  would  be affected  by
 these monltorinc and rcportinq require-
 ments. Specifically, for nitric arid plants
 It was found that by modifying the pro-
 posed  regulations to affect only those
 plants that have a total daily production
 capacity of 300 tons or more of nitric acid
 (rather  than  affecting  all facilities as
 proposed)   that approximately 797*  of
 the nitric  acid production on a national
 basis would be affected by the provisions
 of these  monitoring and reporting re-
 quirements. On the other hand, such a
 modification  reduces  the number  of
 monitors required for compliance with
 these regulations by approximately 467.-.
 (2)  At the present time, only nitric acid
 plants in AQCR's where the Administra-
 tor  has specifically  called for a control
 strategy for nitrogen dioxide will be can-
 didates for continuous emission monitor-
 ing  requirements for the  reasons men-
 tioned previously. In the 2 AQCR's where
 such a control strategy has been called
 for. there  is only one known nitric acid
 plant and  that is reported to be less than
 300  tons per day  production capacity—
 hence no nitric acid plants at the present
 time will be affected by these monitoring
 requirements.
   Similarly, evaluations of sulfuric acid
 plants and catalytic cracking catalyst re-
 generators  at  petroleum  refineries re-
 sulted in the  conclusion that minimum
 size limitations of 300 tons per day pro-
 duction rate at sulfuric acid plants, and
 20.000 barrels  per day of fresh  feed to
 any catalytic cracking unit at petroleum
 refineries  could  be reasonably estab-
 lished. Such modifications exempt ap-
 proximately 377' and  39% respectively
 of such plants on a  national basis from
 these emission  monitoring and reporting
 rcauirements.  while  allowing about 9%
 of the sulfur dioxide emissions from sul-
 furic acid  plants and  12% of the par-
 ticulate matter emissions from catalytic
 cracking units to be emitted to  the at-
 mosphere  without being  measured and
 reported.  '2)  The Agency believe that
 such modifications provide a reasonable
 balance  between  the  costs  associated
 with emission monitoring and reporting.
 and the need to obtain such information.
  A number of commentors  suggested
 that sources be exempt  from  the pro-
 posed emission monitoring regulations if
 such sources are located  within areas of
 the  nation that  are already attaining .
 national  standards.  The  Administrator
 does not believe that such an approach
 would be consistent  with Section 110 of
 the Clean  Air  Act. which  requires con-
 tinued maintenance of  ambient  stand-
ards after attainment. In many areas.
 the  standards  are being  attained  only
 through   effective  Implementation   of
emission limitations.  Under the Clean Air
Act. continued compliance with  emis-
sion limitations in these areas is  Just  as
important  as compliance In areas which
have not attained the standards.
  Another  major  comment concerned
the  proposed   data  reporting  require-
ments. Thirty-four (34) commentors ex-
pressed concern at the amount of data
which the  proposed regulations required
to be recorded, summarized, and submit-
                              FIOIIAl U6ISTII, VOL.  40. NO. 194—MONDAY. OCTOIfl 6. 1t75
                                                          Ill-Ill

-------
                                               RULES AND  REGULATIONS
  ted  to  the State. It was generally indi-
  cated by the commentors that the datn
  reportinp  requirements were excessive.
  Commentors  questioned the purpose of
  reporting all  measured data while some
  State agencies indicated they hnve lim-
  ited resources t"1 handle such informa-
  tion. EPA believes that, in some cases.
  the  commentors misconstrued the data
  reporting  reouirements  for  existing
  sources. In light of each of these com-
  ments, the final regulations, with respect
  to the  data  reporting  requirements for
  gaseous pollutants  and opacity,  have
  been modified
   For gaseous emissions, the proposed
  regulations required the reporting of all
  one-hour averages obtained by the emis-
  sion monitor. Because of the comments
  on this provision, the Agency has reex-
  atnined the proposed data reporting re-
  quirements. As a result, the Agency has
  determined that only information con-
  cerning emissions in excess of emission
  limitations of the applicable plan is nec-
  essary to satisfy the intent of these reg-
  ulations.  Therefore, the data reporting
  requirements  for  gaseous  pollutants
  have been modified. The final regulations
  require that States adopt procedures that
  would require sources  to report to the
  State on emission levels in excess of the
  applicable emission limitations 'i.e., ex-
  cess emissions) for the time period spec-
  ified in the regulation with which com-
  pliance is determined. In other words, if
  an  applicable emission  limitation  re-
  quired no more than 1.0 pounds per .hour
  SO., to be emitted for any two-hour aver-
  aging period,  the data to be reported by
  the source should identify the emission
  level  (i.e.. emissions stated in pounds per
  hour) averaged  over  a two-hour time
  period,  for periods only when  this emis-
  sion level was in excess of the  1.0 pounds
  per  hour emission limitation. Further.
  sources shall  be  required to maintain a
  record of all  continuous monitoring ob-
  servations for gaseous pollutants  'and
  opacity measurements)  for a period of
  two years and to make such data avail-
  able to  the State upon request. The final
  regulations have also been amended to
  add a provision  to require sources to re-
  port to  the State on the apparent reason
  for oil noted violations of applicable reg-
  ulations.
   The proposed data reporting require-
  ments for opacity have also been modi-
  fied.  Upon reconsideration of the extent
 of the data needed to satisfy the intent
 of these regulations. It is the Adminis-
  trator's judgment that for opacity States
 must obtain   excess emission  measure-
 ments during each  hour of operation.
 However,   before  determining   excess
 emissions, the number of minutes gen-
 erally exempted by State opacity regu-
 lations  should be considered.  For ex-
 ample,  where a  regulation allows two
 minutes of  opacity  measurements  in
 excess  of the   standard,  the   State
 need  only require the  source  to  re-
.  port all opacity measurements in excess
 of the standard  during any one hour.
  minus the two-minute  exemption. The
 excess measurements shall be reported
 In actual  per  cent opacity averaged for
 one clock minute or such other time pe-
 riod deemed appropriate  by the State.
 Averages may be calculated  cither by
 arithmetically avcraping a minimum  of
 4 equally spaced data points per minute
 or by integration of the monitor output.
  Some  commentors  raised  questions
 concerning the provisions in the proposed
 regulations  which allow the use of fuel
 analysis for computing emissions of sul-
 fur dioxide  in lieu of Installing a con-
 tinuous monitoring device for  this pol-
 lutant. Of primary concern with the fuel
 analysis  approach  among  the  com-
 mentors was the frequency of the analy-
 sis to determine the sulfur content of the
 fuel.  However, upon  inspection of the
 comments by  the Agency, a more sig-
 nificant  issue  has been uncovered. The
 issue involves the determination of what
 constitutes excess emissions when a fuel
 analysis is used as the method to measure
 source emissions. For example, the sulfur
 content varies significantly within a load
 of  coal, i.e., while the average sulfur
 content of a total load of coal may be
 within acceptable limits in relation to a
 control  regulation  which  restricts the
 sulfur content of coal, it is probable that
 portions of  the coal may  have a sulfur
 content above the allowable level. Simi-
 larly, when  fuel oils of different specific
 gravities  are  stored  within a  common
 tank, such fuel oils tend to stratify and
 may  not be  a  homogeneous  mixture.
 Thus, at times, fuel oil in excess of allow-
 able limits may be combusted. The ques-
 tion which arises is whether the combus-
 tion of this  higher sulfur coal or oil is a
 violation of  an applicable sulfur content
 regulation. Initial  investigations of this
 issue  have indicated  a  relative lack of
 specificity on the subject.
  The Agency is confronted  with this
 problem not only in relation to specifying
 procedures for the emission reporting re-
 quirements for existing sources but also
 in relation to enforcement considerations
 for new sources affected by New Source
 Performance Standards. At this time, a
 more thorough investigation of the situ-
 ation in necessary prior to promulgation
 of procedures  dealing with fuel analysis
 for both oil and coal. At the conclusion
 of this investigation, the Agency will set
 forth  its findings and provide  guidance
 to State and  local control agencies on
 this issue. In the  meantime, the portion
 of the proposed regulations dealing with
 fuel analysis is being withheld from pro-
 mulgation at this time. As such, States
 shall not be required to adopt provisions
 dealing  with emission monitoring or re-
 porting of sulfur dioxide emissions from
 those  sources where the States  may
 choose to allow the option of fuel anal-
 ysis as an alternative to sulfur dioxide
 monitoring.  However, since  the   fuel
 analysis alternative may not be utilized
 by a source  that has installed sulfur di-
oxide  control equipment  (scrubbers).
 States shall  set forth legally enforceable
 procedures which require emission moni-
tors on such sources,  where these emis-
 sion monitoring  regulations otherwise
require their installation.
  Other Modifications to Proposed  Reg-
ulations. In addition to  reducing the
number of monitors required under the
proposed regulations, a number of modi-
fications to  various procedures in the
proposed  regulations  have  been  con-
sidered and  are included in the final
regulations. One modification which has
been made is the deletion of the require-
ment to install continuous monitors  at
"the most  representative"  location. The
final  regulations require the placement
of an'emission monitor at "a representa-
tive" location in the exhaust gas system.
In many cases "the most representative"
location may be difficult to locate and
may be inaccessible without new plat-
forms, ladders, etc.. being iastalled. Fur-
ther,  other representative  locations can
provide adequate information on pollut-
ant emissions  if minimum criteria for
selection of monitoring locations are ob-
served. Guidance in determining a repre-
sentative sampling' location is contained
within  the  Performance  Specification
for each pollutant monitor in the  emis-
sion  monitoring regulations  for  New
Source Performance Standards (Appen-
dix B, Part 60 of this Chapter). While
these  criteria   are  designed  for  new
sources, they are also useful in deter-
mining representative locations for ex-
isting sources.
  A further modification to the proposed
regulation  is the deletion of the require-
ment for new  performance  tests  when
continuous emission monitoring equip-
ment is modified or repaired. As pro-
posed,  the  regulation would have re-
quired a new performance test whenever
any  part  of  the continuous  emission
monitoring  system  was  replaced. This
requirement was originally incorporated
in the regulations to assure the use  of
a well-calibrated, finely  tuned monitor.
Commentors pointed out  that the re-
quirement  of conducting new perform-
ance tests  whenever any part of an in-
strument is changed or replaced is costly
and in many cases  not  required.  Upon
evaluation  of this comment, the Admin-
istrator concurs that performance tests
are not required after each repair or re-
placement  to  the  system. Appropriate
changes have been  made to the regula-
tions to delete the requirements for new
performance tests.  However, the  final
regulations require  the reporting of the
various repairs made to  the  emission
monitoring system  durine  each quarter
to the  State. Further,  the State  must
have procedures to require  sources to re-
port to the State on a quarterly basis in-
formation on the amount of time and the
reason why the continuous monitor was
not in  operation.  Also  the State  must
have  legally enforceable procedures  to
reouire a source to conduct a new per-
formance test whenever,  on the basis  of
available information, the -State deems
such test is necessary.
  The  time period proposed for the In-
stillation  of the  reauired monltorine
system, i.e.. one vpar after plan apnroval.
wns thought, hv 21 commentors to be too
hrjpf. orlmarilv because of lack  of avail-
able instruments, the lack of trained ner-
snnnrl and  the  time available for Instal-
lation of the required monitors. Eauip-
ment  suoolicrs  were  contacted by the
Agency and thev confirmed the avail-
ability  of emission  monitors. However.
                               FIDMAl MCISTEI. VOL. 40, NO. 1»4—MONDAY, OCTOUI », 197S
                                                         III-112

-------
                                             RULES  AND REGULATIONS
 the  Administrator has determined that
 the  time necessary for purchase. Inslal-
 lation and  performance tr.Miiu: of .such
 monitors may require morr than  one
 year for rcrtnin  installations, especially
 where gaseous monitors nre required. In
 order to provide sources with n in pic time.
 the Agency has modified the final reputa-
 tions to allow States to adopt, procedures
 that will provide sources 18 months after
 the  approval or promulgation of the re-
 vised SIP to satisfy the installation and
 performance testing procedures  required
 by these continuous monitoring regula-
 tions. A provision is also included  to al-
 low, on a case-by-cnse bnsis. additional
 extensions for sources where good faith
 efforts have been undertaken to purchase
 and  install  equipment, but where such
 Installation  cannot   be   accomplished
 within  the time  period  prescribed  by
 the  regulations.
   A  number of State and local  agencies
 also commented on the lack of time pro-
 vided sources to Install the monitors re-
 quired  by  the  proposed  regulations.
 These agencies also indicated that they
 must Acquire sufficient skilled manpower
 to Implement  the  regulations,  such  as
 personnel to provide guidance to sources.
 to monitor  performance  tests and  to
 analyze the emission data that are to be
 submitted by  the sources. Further, some
 State agencies Indicated that more than
 six months was  needed to develop  the
 necessary  plan  revisions.  Most  State
 agencies who commented stated that one
 year should be provided to allow States
 to revise their SIP's. The Administrator
 Is aware of the various priorities which
 confront State and local agencies at this
 time 'e.g.. compliance schedules, enforce-
 ment actions, litigation proceedings, re-
 evaluation of adequacy of SIP's to attain
 and  maintain national standards, etc.)
 and. as such, believes  that a six-month
 postponement in the submittnl of plan
 revisions to require emission monitoring
 and  reporting  is  Justified  and prudent.
 Hence. States must submit plan revisions
 to satisfy the requirements of  this sec-
 tion  within  one year of promulgation of
 these regulations in the FEDERAL REGIS-
 TER.  However.  States  are  advised  that
 such plan revisions may be submitted
 any  time prior to the final  date, and are
 encouraged to do so where possible.
  The proposed regulations provided the
 States with the option of allowing sources
 to continue to use emission monitoring
 equipment that does not meet perform-
 ance specifications set forth In the regu-
 lations for up to five years from the date
 of approval of the State regulations or
 EPA  promulgation. Some  commenters
 asked that  this  provision be extended
 Indefinitely. In some cases they indicated
 they  had recently purchased and had
 already  installed  monitoring  systems
 which were only  marginally away from
meeting the applicable performance spec-
 ifications.  The  Agency believes,  how-
 ever, that such a modification to the pro-
 posed regulations should not be  allowed.
 It is believed that such a provision would
 result in Inadequate monitoring systems
 being maintained after their useful  life
 has ended. Though some monitoring sys-
tems will'probably last longer than five
years, it is believed Unit this lime period
will  provide aclcciuntr  time to  amortize
the  cosl  of such equipment.  In  cases
where  existing  emission monitors are
known  not  to provide  reasonable esti-
mates of emissions.  States should con-
sider more stringent procedures to pro-
vide a  more speedy  retirement  of such
emission monitoring  systems.
  Some commentors raised the question
of whether existing  oxygen monitors
which are installed in most  fossil fuel-
fired steam generating boilers to monitor
excess oxygen  for the purposes of com-
bustion control could be used to satisfy
the requirement  for monitoring oxygen
under the proposal. Upon investigation.
it has  been determined that.  In  some
cases, such oxygen monitors may be used
provided that they are located  so that
there is no influx of dilution air between
the oxygen monitor and the  continuous
pollutant monitor. In some cases, it may
be  possible to  install  the continuous
monitoring device at the same location
as the  existing   oxygen monitor.  Care
should be taken, however, to assure that
a representative  sample is obtained. Be-
cause  of  the  various  possibilities that
may arise concerning the usefulness  of
existing  oxygen  monitors,  the  State
should  determine, after a  case-by-case
review,  the acceptability of  existing oxy-
gen monitors.
  Another technical  issue which  was
raised suggested  that continuous emis-
sion   monitors  which   provide  direct
measurements of pollutants in units com-
parable to the emission limitations and
other devices not specifically identified
in the  proposed  regulations are avail-
able for purchase and installation. The
Agency is aware that  various monitor-
ing systems  exist but has not as yet de-
termined specific performance specifica-
tions for these monitoring  systems that
are  directly  applicable  to the source
categories covered by these regulations.
However,  it  is not EPA's intent to deny
the use of any  equipment that can  be
demonstrated to be reliable and accurate.
If monitors can be demonstrated to pro-
vide the same relative degree of accuracy
and  durability as provided by  the per-
formance specifications in Appendix  B
of Part 60.  they  shall generally be ac-
ceptable to satisfy the requirements  of
these regulations under Section 3.9  of
Appendix P. Further, where alternative
procedures  (e.g.. alternate  procedures
for conversion of data  to units of appli-
cable regulations) can be shown by the
State to be equivalent to the procedures
set forth in Appendix P of  these regula-
tions,  then  such alternate  procedures
may  be submitted by the State for ap-
proval by EPA. Section  3.9 of Appendix P
identifies certain examples where alter-
native emission  monitoring systems  or
alternative procedures  will generally  be
considered by  the Agency for approval.
  It  should be noted that some sources
may be unable to comply with the regu-
lations  because of technical  difficulties,
(e.g..  the presence of  condensed water
vapor in the flue gas), physical limita-
tions of accessibility at the plant facility.
or.  In  other  cases, because of extreme
economic,  hardship.  Stales should use
their  judgment  In Implementing  these
reciulicmcnts in such cases. Section 6 of
Appendix P of this Part provides various
examples where  the Installation of con-
tinuous emission monitors  would not be
feasible or reasonable. In such  cases
alternate emission monitoring  (and re-
porting' by more routine methods, such
ns  manual  stack  testing,  must be re-
quired. States in preparing their revised
SIP must set forth and describe the cri-
teria they  will use to identify such un-
usual cases, and must further describe
the alternative procedures  they will Im-
plement to  otherwise satisfy the intent of
these regulations. Stales are advised that
this provision .is intended for unusual
cases, and.  as such, should  not be widely
applied.
  It was   pointed out  by  some  com-
mentors that  carbon dioxide  monitors
could probably be used in lieu of oxygen
monitors to provide information to con-
vert emission  data to the  units of the
applicable  State  regulation.  Detailed
discussion  of  the  technical merits and
limitations, of  this approach is discussed
in the  Preamble to the Part 60 Regula-
tions. As pointed out in that  Preamble.
such monitors may be  used in certain '
situations.  Modifications have therefore
been made to  the Part 51 regulations to
allow the use of such monitors which in-
clude references  to technical specifica-
tions contained in Part 60 for carbon di-
oxide monitors. Also, the cycling time for
oxygen monitors has been changed from
one hour to 15 minutes to correspond to
the specification in Part 60. The differ-
ence between  cycling times in the two
proposals was an oversight. The cycling
time for carbon dioxide monitors will
also be 15  minutes as in Part 60.
  A number of other miscellaneous tech-
nical comments were also received. Com-
mentors indicated that the  proposed ex-
emption for opacity monitoring require-
ments  that may be granted to oil-fired
and gas-fired  steam generators should
also apply  to units burning a combina-
tion of these  fuels. The Administrator
concurs with this suggestion and an ex-
emption for such sources burning oil and
gas has ben provided in the final  regu-
lations subject to  the same restrictions
as  are  imposed  on  oil-fired  steam
generators.
  As previously  indicated, the regula-
tions for emission monitoring for exist-
ing sources refer in  many  cases to the
specific performance  specifications set
forth in the emission monitoring regula-
tions for new sources n fleeted by Part 60.
Many of the comments received on the
proposed regulations in effect pointed to
issues affecting both proposals. In many
cases, more specific technical  issues are
discussed in the Preamble to the Part 60
Regulations and  as such  the reader  is
referred to that Preamble. Specifically,
the Part 60 Preamble addresses the fol-
lowing topics:  data handling and report-
Inn techniques: requirements for report-
ing repairs and replacement parts  used;
location  of   monitoring   Instruments:
changes to  span  requirements, operating
                              HOIIAl MOUTH, VOL 40. NO. 1*4—MONDAY, OCTOBEI *, 1975
                                                              III-113

-------
                                              RULES AND REGULATIONS
 frequency requirements, sulfuric ncld and
 nitric  acid  plant conversion  factors:
 and. for opacity monitoring equipment.
 chances in the cycling time and in allcn-
 mcnt  procedures. The  reader  is cau-
 tioned, however, that specific reference
 to regulations In  the Part 60 Preamble
 IK strictly to federal ..ew Source Perform-
 ance Regulations rather than SUile and
 local  control  agency regulations  which
 affect existing sources and which are part
 of an applicable plan.
  In  addition to the  many  technical
 comments  received,  a number of legal
 issues  were raised. Several  commentors
 questioned EPA's  statutory  authority .to
 promulcate these regulations and pointed
out other alleged legal defects in the pro-
 posal. The  Administrator has considered
 these comments, and has found them un-
 persuasive.
  One commentor argued that new 40
 CFR 51.19(e)  will require "revisions" to
 existing state plans: that'"revisions" may
 be called for under Section 110(a) (2(H>
 of the Clean Air Act only where EPA has
 found  that there are  "improved or more
expeditious methods" for achieving am-
 bient standards or that a state plan is
 "substantially inadequate" to achieve the
standards:  that the  new regulation is
 based upon neither of these findings, and
 that therefore there is no statutory au-
 thority for the regulation. This  argu-
 ment fails to take cognizance of Section
 110 (ii) of the Act. which man-
dates that all state implementation plans
contain  self-monitoring requirements.
The fact that EPA  originally accepted
plans  without these requirements be-
cause of substantial uncertainty as to the
reliability of self-monitoring equipment
does not  negate  the mandate of the
statute.
  In essence, new 5 51.19(e) does not call
for "revisions" as contemplated by the
 Act. but for supplements to the original
plans to make them complete. At any
rate, it is the Administrator's Judgment
that the new self-monitoring require-
 ments  will result in a "more expeditious"
achievement of the  ambient standards.
Since  these requirements are valuable
enforcement tools and indicators of mal-
 functions, they should lead  to a net de-
crease in emissions.
  Other commentors argued that even if
EPA has statutory authority  to require
self-monitoring, it has no authority to
impose specific  minimum requirements
for state plans,  to require "continuous"
monitoring, or to  require monitoring of
oxygen, which is not a pollutant.  These
comments fail to  consider that a basic
precept of administrative law  is that an
agency may fill in  the broad directives of
legislation  with precise  regulatory re-
quirements. More specifically, the  Ad-
ministrator has authority under Section
 301 fa) of the Clean Air Act to promul-
gate "such regulations as are necessary
to carry out his functions under the Act".
Courts have long upheld  the authority of
agencies to promulgate more specific re-
quirements  than  are set forth in en-
abling legislation, so long as the require-
ments  are reasonably related to the pur-
 poses of the  legislation. Since the Act
requires self-monitoring without further
guidance. EPA surely has the authority
to set specific rrtiuirements in order to
carry out its function of assuring that the
Act is properly implemented.
  In EPA's  Judcnicnt, the  requirements
set forth  in ? 51.19'e)  are necessary to
assure that  each state's self-monitoring
program Is sufficient to comply with the
Act's mandate. The fact that oxypon and
carbon  dioxide  are  not  air pollutants
controlled under the Act is legally ir-
relevant, since in EPA's judgment,  they
must  be  monitored in order to convert
measured emission data to units of emis-
sion standards.
  Other  commentors have argued  that
the self-monitoring requirements violate
the protection against self-incrimination
provided in the Fifth Amendment to the
U.S. Constitution, and that the informa-
tion obtained  from the monitoring is so
unreliable as to be Invalid evidence for
use in court.
  There  are two reasons why the self-
incrimination  argument is invalid. First,
the self-incrimination privilege does not
apply to corporations, and it is probable
that a great majority of the sources  cov-
ered by these requirements will be owned
by  corporations.  Secondly, courts have
continually  recognized an  exception to
the privilege  for  "records required by
law",  such  as  the self-monitoring  and
reporting procedures which are required
by the Clean Air Act. As to the validity
of evidence  issue, in EPA's opinion, the
required  performance specifications will
assure that self-monitoring equipment
will be sufficiently reliable  to withstand
attacks in court.
  Finally, some  comments reflected a
misunderstanding of EPA's  suggestion
that states explore with counsel ways to
draft their regulations so as to automati-
cally  incorporate by reference  future
additions  to Appendix P and avoid  the
time-consuming  plan revision  process.
(EPA pointed  out that public participa-
tion would still be assured, since EPA's
proposed revisions to Appendix P would
always be subject to public comment on
a nation-wide  basis.)
  EPA's purpose  was merely to suggest
an  approach that a state  may  wish to
follow i!  the  approach would  be legal
under that  state's law. EPA  offers no
opinion as  to whether any state   law
would allow this. Such a determination
is up to the  individual states.
  Summary of Revisions and Clori/ica-
tions   to  the  Proposed  Regulations.
Briefly, the revisions  and clarifications to
the proposed regulations Include:
  (1)  A clarification to indicate that con-
tinuous emission monitors are  not re-
quired for sources unless  such sources
arc subject to an applicable  emission
limitation of an approved SIP.
  (2)   A  revision to require  emission
monitors for oxides  of nitrogen in  only
those AQCR's where the Administrator
has  specifically  called for  a  control
strategy for nitrogen dioxide.
  (3) A revision to include a general  pro-
vision to exempt any source that clearly
demonstrates that It will cease operation
within five years of the inclusion of moni-
torinc requirements for the source in
Appendix P.
  <4> Revisions to exempt smaller-sized
sources  t»nd  infrequently used  sources
within the specified source cateRories
  <5> A  revision  to the data reporting
requirements to-requlre the submillal by
tlir source of the State, emission data in
excess of the applicable  emission limita-
tion for  both opacity and  gaspous  pol-
lutants, rather than all measured data, as
proposed A provision has been added to
require information on  the cause of all
noted violations of applicable regulations.
  (6' A clarification to indicate that the
continuous monitoring of oxygen is not
required  unless the continuous monitor-
ing  of sulfur dioxide  and/or nitrogen
oxides emissions is required by the appli-
cable SIP.
  (1) A revision to allow the placement
of continuous emission  monitors at "a
representative location" on the exhaust
gas  system  rather than at  "the most
representative location" as required by
the proposed regulations.
  '8i A  revision  to delete the require-
ments of new performance tests each
time  the continuous monitorlne  equip-
ment is repaired or modified. However, a
new provision is included to require that
a report  of all repairs and  maintenance
performed during the quarter shall be re-
ported by the source to the  State.
  (9> A modification to provide sources
18 months rather than one year after
approval or promulgation of the revised
SIP to comply with the continuous moni-
toring regulations adopted by the States.
  (10) A modification to provide States
one  year, rather than  the six  months
after the promulgation  of these regula-
tions in the FEDERAL REGISTER to submit
plan revisions to satisfy the requirements
promulgated herein.
  Requirements of States. States shall be
required  to  revise their SIP's by Octo-
ber 6.1976 to include legally enforceable
procedures to require emission monitor-
ing,  recording and reporting, as  a mini-
mum for those sources specified in the
regulations  promulgated herein. While
minimum requirements have been estab-
lished. States may. as they deem appro-
priate, expand these requirements.
  The regulations  promulgated  herein
have been revised In light of the various
comments to generally  provide  a more
limited introduction into this new meth-
odology.  Cooperation   among  affected
parties, i.e.. State and local control agen-
cies, sources, instrument manufacturers
and  suppliers, and this Agency is neces-
sary  to  move  successfully  forward in
these areas  of emission monitoring and
reporting prescribed in the Clean Air
Act. Assistance can be obtained from the
EPA Regional Offices in relation to the
technical and procedural aspects of these
regulations.
  Copies of documents referenced in this
Preamble are available for public inspec-
tion at the EPA Freedom of Information
Center. 401 M Street. S.W., Washington.
D.C.  20460.  The  Agency has not pre-
pared an environmental impact state-
ment for these regulations  since they
                              PSBSIAl OieiiTSB, VOl. «0, WO 194—MONDAY, OCTOBER 6,  1*75
                                                   III-114