EPA-450/2-75-001
January 1975
TECHNICAL SUPPORT DOCUMENT -
EPA REGULATIONS FOR PREVENTING THE
SIGNIFICANT DETERIORATION OF AIR QUALITY
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
-------
EPA-450/2-75-001
TECHNICAL SUPPORT DOCUMENT -
EPA REGULATIONS FOR PREVENTING THE
SIGNIFICANT DETERIORATION OF AIR QUALITY
U.S. Environmental Protection Agency
Office of Air Quality Planning & Standards
Research Triangle Park, N. C.
January 1975
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors
and grantees, and nonprofit organizations - as supplies permit - from
the Air Pollution Technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina 27711; or, for a
fee, from the National Technical Information Service, 5285 Port Royal
Road, Springfield, Virginia 22161.
Publication No. EPA-450/2-75-001
-------
CONTENTS
1. BACKGROUND 1
11. REASONS FOR SELECTING A THREE-CLASS AIR QUALITY INCREMENT
PLAN 5
111. REASONS FOR REJECTING ALTERNATIVE PLANS 10
1. Volume Averaging (Sierra Club Plan) 10
2. Emission Limitation Plan 11
3. Emission Charge Plan 12
4. Per Capita Emission Plan (NRDC Plan) 14
5. Percent Increase Above Existing Air Quality Level . . 15
6. Local Definition Plan 17
7. Establishment of Absolute Air Quality Levels Below
the National Standards 18
IV. SELECTION OF AIR QUALITY INCREMENTS 19
V. POLLUTANTS COVERED BY THE REGULATION 21
VI. SOURCES SUBJECT TO REVIEW 27
VI1. AIR QUALITY MONITORING REQUIREMENTS 29
Vll1. IMPACT OF THE REGULATIONS ON GENERAL GROWTH AND
DEVELOPMENT 30
IX. IMPACT ON POWER GENERATION AND NEW ENERGY SOURCES .... 30
X. INTERSTATE BOUNDARY CONFLICTS 32
XI. BEST AVAILABLE CONTROL TECHNOLOGY ... 34
Xll. INITIAL CLASSIFICATION AND CRITERIA FOR RECLASSIFICATION . 35
Xlll. EFFECTIVE DATE FOR SOURCE REVIEW AND DETERMINATION OF
SIGNIFICANT DETERIORATION 37
REFERENCE 6 41
REFERENCE 11 52
REFERENCE 12 71
REFERENCE 13 80
REFERENCED 81
REFERENCED 99
REFERENCE 16 107
REFERENCE 17 Ill
REFERENCE 20 115
-------
This document provides a detailed discussion of the
technical and policy considerations which form the basis
for EPA's regulations for prevention of significant air
quality deterioration, which were published on December 5,
1974 (39 FR 42510). Generally, the body of this document
contains a summary of each major issue and references
more detailed materials which have been prepared on that
issue.
I. Background
At the time the Administrator was required to approve
or disapprove State Implementation Plans (SIP's), he
believed that he was required under §110 of the Clean
Air Act (the "Act") to approve SIP's which attained and
maintained national ambient air quality standards promul-
gated under §109 of the Act. He did not believe that he
had authority to disapprove SIP's which attained and main-
tained national ambient standards but did not prevent
deterioration of existing air qualiry cleaner than the
standards.
On May 24, 1972, the Sierra Club and other groups
filed suit in the U. S. District Court for the District
of Columbia seeking a declaratory judgment and injunction
requiring the Administrator to disapprove all State imple-
mentation plans which did not contain procedures for pre-
venting significant deterioration in any portion of any
State where air quality is superior to national standards.
On May 30, 1972, the District Court for the District of
1
-------
_ 2 —
Columbia granted the plaintiffs" motion and issued a pre-
liminary injunction requiring the Administrator to review
all State plans and "disapprove any portion of a State
plan which fails to effectively prevent significant deteri-
oration of existing air quality." (344 F.Supp. 253) The
preliminary injunction also required the Administrator to
promulgate regulations "as to any State plan which he
finds, on the basis of his review, either permits the signi-
ficant deterioration of existing air quality in any portion
of any State or fails to take the measures necessary to
prevent such significant deterioration." On November 1,
1972, the decision of the District Court was affirmed by
the U. S. Court of Appeals for the District of Columbia
Circuit on the basis of the District Court opinion. (4
ERC 1815) On June 11, 1973, the Supreme Court, by an
equally divided court, affirmed the judgment of the Court
of Appeals; no opinion was issued. (412 U.S. 541)
Each State plan has been reviewed in accordance with
the preliminary injunction issued by the District Court.
Although many State plans included regulations which have
the potential for resulting in the attainment of air
quality better than that required by the national standards,
and although some State plans contained general policy
statements indicating an intent to prevent or minimize
deterioration of air quality, none was found to contain
explicit and enforceable regulations for implementing such
1
a policy.
-------
- 3 -
Accordingly, all State plans were disapproved by the
Administrator on November 9, 1972 (37 FR 23836), insofar
as they failed to provide for the prevention of significant
deterioration. This disapproval did not affect the status
of any previously or subsequently approved regulations
designed to provide for the attainment and maintenance of
national ambient air quality standards.
On July 16, 1973 (38 FR 18986), an initial notice of
proposed rulemaking was published which set forth four
alternative plans for preventing significant deterioration,
and which solicited widespread public involvement in all
aspects of the significant deterioration issue. A series
of public hearings were held and over 300 written comments
were submitted in response to this proposal. The hearing
records and the written comments are available for inspection
at the EPA Freedom of Information Office, 401 M Street,
S.W., Washington, D. C.
Since neither the Clean Air Act nor the Court order
define "significant deterioration," the initial proposals
focused on the conceptual basis for regulations. The com-
ments received on the proposed regulations therefore tended
primarily to discuss conceptual issues such as the roles
of federal and state/local governments, rather than de-
tailed comments regarding implementation of the regulations.
Accordingly, on August 26, 1974 (39 FR 31000), the Adminis-
trator reproposed regulations based on the conceptual
approach he had determined appropriate. The purpose of
-------
.. 4 -
the reproposal was to focus more clearly on procedural
and technical issues. Final regulations were published
on December 5, 1974 (39 FR 42510).
Briefly, the regulations promulgated on December 5,
call for the establishment of "classes" of different
allowable incremental increases in total suspended parti-
culates (TSP) and sulfur dioxide (SO^,) . Class I applies
to areas in which practically any change in air quality
would be considered significant; Class II applies to areas
in which deterioration normally accompanying moderate
well-controlled growth would be consideired insigificant;
and Class III applies to those areas in which deterioration
up to the national standards would be considered insigni-
ficant. All areas of the country are designated Class II
initially, but provisions are included to allow States. Federal
Land Managers and Indian governing bodie;s to request re-
designation of any area to accommodate social, economic,
and environmental needs.
The plan is implemented through a preconstruction
review of specified source categories to determine whether
these sources would cause a violation of the appropriate
increments. The regulations also include a provision
requiring the use of best available control technology
on sources covered by the regulation. Finally, the regu-
lation provides procedures for public comment on each
application for permission to construct and for delegating
-------
- 5 -
the responsibility for implementing the new source review
procedures to States or local governmental units.
II. Reasons for selecting a three-class air quality
increment plan
The decision to adopt a three-class increment plan
was based on the conclusion that consideration of varying
social, economic, and environmental factors in different
areas would result in varying definirions of what is
"significant."
Section 109 of the Clean Air Act requires the Admin-
istrator to establish national secondary ambient air quality
standards, "to protect the public welfare from any known
or anticipated adverse effects," including, as specified
by section 302(h), "effects on soils, water, crops, vegeta-
tion, man-made materials, animals, wildlife, weather,
visibility, and climate, damage to and deterioration of
property, and hazards to transportation, as well as effects
on economic values and on personal comfort and well-being."
Such national standards must be based on air quality
criteria which, under section 108, must "reflect the
latest scientific knowledge useful in indicating the kind
and extent of all identifiable effects on public health
and welfare which may be expected from the presence (of
air pollutants) in the ambient air, in varying quantities."
Thus, standard-setting under section 109 protects against
demonstrable or predictable adverse effects which can be
-------
- 6 -
quantitatively related to pollutant concentrations in
the ambient air.
Limitations on air quality that result in cleaner
air than the national ambient air quality standards cannot
therefore be based on any quantitative measure of harm to
either public health or welfare. This is not, however,
to say that there are no possible unquantifled adverse
effects on public health or welfare below the levels of the
national standards. Examples of such unquantified effects
involve the transformation of sulfur dioxide into suspended
sulfates and sulfuric acid aerosols, resulting in
possible effects on nealth, visibility, climatic changes,
2
acidity of rain, and deterioration of materials.
Since there is no way to relate "significance" of
deterioration of air quality to any adverse effects re-
sulting from air quality levels cleaner than the national
standards, EPA concluded that the determination of what
is "significant" deterioration must take: into account
factors other than air quality alone. For example, re-
latively minor deterioration of the aesthetic quality of
the air may be very significant in a recreational area
in which great pride (and economic development) is derived
from the "clean air". Conversely, in areas with severe
-------
_ '7 _
unemployment and little recreational value, the same
level of deterioration might very well be considered
"insignificant" in comparison to the favorable impact
-of new industrial growth with resultant employment and
other economic opportunities.
EPA's belief that any determination of what deteriora-
tion increment should be prohibited as "significant"
must take into account the practical effects of such
determination is supported by the decision directing EPA
to prevent such deterioration and the legislative history
of the Clean Air Act. In its decision directing EPA to
promulgate regulations preventing significant deterioration
of air quality, the District Court relied heavily on
language in the Senate Report on the bill that became the
Clean Air Act (S.Rep. 91-1196, at 91st Cong. 2d Sess. at
2, 1970).
"In areas where current air pollution
levels are already equal to or better than
the air quality goals, the Secretary shall
not approve any implementation plan which
does not provide, to 'the maximum extent
practicable, for the continued maintenance
of sucn amtoient air quality." (emphasis
added) *
The significant deterioration regulations have great
potential impact on the nature, extent, and location of
future industrial, commercial, and residential development
The requirement in these regulations that all new
sources in the 18 listed categories use best available
control technology is also part of the attempt to limit
air quality deterioration "to the maximum extent
practicable."
-------
throughout the United States, and specifically their
impact on the utilization of the Nation's mineral re-
sources, the availability of employment and housing in
many areas, and the costs of producing and transporting
electricity and manufactured goods. The Administrator
believes that it is most important to recognize and
consider these implications, since the consideration of
air quality factors alone provides no basis for selecting
one deterioration increment over another.
Although some have argued that economic and social
factors should have no bearing on the definition of signi-
ficant deterioration and that only air quality factors
should be considered in establishing a single nationwide
definition of significant deterioration, none of the
3,4,5
comments suggesting changes to the increments pro-
posed by the Administrator, or proposing alternate plans,
offered any justification for the numbers which were
selected.
It has been argued that the provision for Class III
areas is unresponsive to the court's order in that it
permits deterioration of air to the national standards
in these areas. Although this result could also occur
in Class I or Class II regions where the difference
between existing air quality and the national standard is
less than the prescribed air quality increment, all such
comments focused on the provision for Class III areas.
-------
_ Q _
Unless "significant deterioration" is defined as a per-
centage of the "unused" air resource, any air quality
increment plan, regardless of how small the increment is,
could allow deterioration up to the national standard in
some instances. As discussed below, a plan that is dependent
on a knowledge of existing air quality is virtually un-
workable. Therefore, the fact that air quality could,
in some, instances, increase to the national standard does
not, in the Administrator's opinion, make the regulations
inconsistent with the Court's ruling or the Clean Air Act.
Since the regulations are based on the premise that
the significance of deterioration must be determined
partially on the basis of the practical effect of the
determination and the present or intended use of the
land (e.g., pristine recreational area or commercial-indus-
trial area), the Administrator believes that the people
in the area for which the determination is being made
should make the determination of what level of
deterioration should be considered significant. EPA
lacks the resources or knowledge to determine what air
quality deterioration would be considered significant
in every area of the country. However, the Administrator
will not approve requested designations which are arbitrary
or capricious. *
Redesignations protested by other States or Indian
Governing Bodies are discussed in Section X, infra.
-------
- 10 -
III. Reasons for rejecting alternative plans
1. Volume averaging (Sierra Club Plan)
The details of this plan are presented in Reference 6.
Under this plan, air quality deterioration would be de-
finded in terms of the concentration of air pollutants
averaged over a one kilometer sphere, the center of which
would be the point emissions are released to the atmosphere.
Although basically an air quality increment plan, the
volume averaging plan is intended to limit the total amount
of pollutants emitted to the atmosphere, rather than
directly limit ground level concentrations. Although total
atmospheric burden is better related to effects such as
acid rain and visibility reduction than is ground-level
air quality, most other unquantified effects at levels
below the national standards are related to ground or
near-ground level concentrations (corrosion rates, plant
effects, suspected health effects, etc.). Since the use
of ground-level air quality at individual points (as
opposed to spatial averages) is consistent with the way
the national ambient air quality standards (NAAQS) were
established and the way attainment of the NAAQS is
determined, the NAAQS provide a meaningful range of
values from which to select an appropriate
-------
- 11 -
deterioration increment. No such "guidance" is avail-
able under the volume averaging plan and the "definition"
of significant deterioration under this plan tends to be
much more arbitrary.
Furthermore, although the average volumetric concen-
tration is easily calculated for a single source, the ad-
ditive effect of multiple sources is much more dificult
to compute. Since this plan contains no inherent advan-
tages, yet is technically more difficult to implement than
a ground-level air quality increment plan, a volume averag-
ing plan was not used.
In addition to the conceptual difficulties with a
volume averaging plan, the specific increments proposed
by the Sierra Club would impose severe growth restrictions,
especially on energy-related facilities. The Sierra Club
has presented no rationale for these specific increments.
2. Emission limitation plan
A typical emission limitation plan was described
in EPA's initial proposal of July 16, 1973 (38 FR 18986
at 18991). This plan would have limited total emissions
of TSP and SC>2 over a relatively large area and indirectly
prevented the significant deterioration of air quality.
-------
- 12 -
Basically, this plan focuses on total atmospheric load-
ing, and as such has many of the same advantages and dis-
advantages as discussed for the volume averaging plan. It
has a significant advantage in terms of technical simplicity
over an air quality increment plan, in that emission data
are easier to acquire and more reliable than estimates of
air quality computed using atmospheric simulation models.
However, its principal disadvantage is the lack of consistency
between emissions and air quality, particularly on a nation-
wide basis. As discussed above, a plan which focuses on
ground-level air quality, in accordance with traditional air
quality management practice, is considered preferable to an
average atmospheric burden plan. In addition, the averaging
of emissions over a large region could permit the creation of
small areas of substantial air quality deterioration. This
might be inconsistent with the court's order to prevent
"significant deterioration in any portion of any state."
The growth and development impacts of this plan were
7,8,9,10
analyzed in several studies and are summarized in a
11
separate report. These studies indicate that economic growth
would be restricted in a number of instances and that these
restrictions would be most severe for coal-fired power plants.
3. Emission charge plan
The general reasoning behind such a plan is that secondary
NAAQS comprise adequate upper limits on pollutant concentrations,
-------
- 13 -
but air quality superior to those limits is desirable. The
emission charge would provide a continuous incentive for
sources to seek and apply emission controls to minimize their
emission charges. The collective effect of these individual
cost minimizations would be to maintain air quality at levels
superior to NAAQS in most areas. The level of air quality
maintained would be a function of the emission charge rate,
the development potential of the area, and the state-of-the-art
of emission control.
The major advantages of this plan are that the cost of
emitting would be "internalized: i.e., it would be taken
into consideration in the normal economic appraisal of plant
design and location alternatives. Sources would have numerous
options as to control method, cost, and degree of control
from which to make the optimum choice. The state-of-the-art
of emission control would be continuously advanced. Finally,
the means of enforcement would be charge collection which is
relatively easily administered.
Unfortunately, several problems attend such a plan.
In view of the requirement that "significant
deterioration of air quality is to be prevented by the emis-
sion charge, some relationship between the charge rate and
the resultant air quality must be found. Such a relationship
is not presently available. Even if this relationship were
available, the emission charge rate would have to vary from
place to place to offset the variation in developmental
-------
- 14 -
potential offered by different land areas and the variable
capacity of the air to disperse waste under different meteoro-
logical and topographical conditions. Also, there is a ques-
tion as to whether the Clean Air Act provides adequate
legal authority for EPA to implement such a plan. But most
important, an emission charge would not guarantee that sig-
nificant deterioration could not take place in some portions
of some states. Consequently, the emission charge, while
possessing some desirable attributes, does not appear to
be a practical means of preventing significant deteriora-
tion of air quality, especially on a nationwide basis.
4. Per capita emission plan (NRDC plan)
The Natural Resources Defense Council (NRDC) proposed
a per capita emission plan. Under this plan the total
emissions in clean areas, plus a five percent increase,
would be divided by the total population in clean areas to
arrive at the allowed per capita emissions. The total
emissions allowed in any area would then be calculated as
(the population in the area) times (the per capita emission
rate). The primary advantages claimed for this proposal
are the emphasis on omissions rather than air quality, and
the relationship between the level of emissions and the pop-
ulation served. The latter advantage cited by NRDC would
-------
- 15 -
in many cases represent a major disadvantage. Because
part of the motivation to prevent significant deteriora-
tion is concern for currently unquantified but suspected
low level effects, it does not seem reasonable to force
new polluting development to locate in areas of high pop-
ulation. This result would add new pollution in areas
where air quality levels are already relatively high and
would result in many more people being exposed to the
added pollutants than if sources were located in less
populated areas.
This plan would tend to prevent development of cur-
rently needed natural resources such as low sulfur coal
and oil shale which are located in areas of very low
population. In addition, the location of many other
facilities such as smelters, paper mills, phosphate rock
processing, and oil shale retorting are determined by the
location of natural resources, not the population served.
Under the per capita emission plan it is unlikely that
facilities such as these could be built.
5. Percent increase above existing air quality level
This approach would limit deterioration to fixed
percentages above existing air quality levels. The pro-
blem with the plan is in the difficulty of defining
"existing" air quality. Even with no change in man-made
-------
- 16 -
emissions, air quality levels in "clean" areas can vary
substantially from day to clay, year to year due to changes
in natural emissions and variations in meteorological con-
12
ditions.* A further problem with this plan is that air
quality monitoring is presently concentrated in heavily
polluted urban areas, which constitute only a small portion
of the total area of the country. In the relatively
clean rural areas which are principally affected by the
significant deterioration regulations, there is very
little monitoring being done. Although atmospheric diffu-
sion modeling could be used to estimate existing air
quality levels, the difficulty in quantifying natural
sources of particulate matter would make the results quite
questionable. Also, models; for estimating existing
concentrations of reactive pollutants (nitrogen dioxide
and photochemical oxidants/hydrocarbons) are not available
as discussed below. In summary, accurate definition of
"For example, Cape Hatteras, North Carolina, showed a
yearly average of (of TSP) 58 ug/m3 in 1968 and 110 ,ag/m
in 1972. The station in the Black Hills National Forest,
South Dakota, had an annual average of 10 /ug/m in 1968
and 37 jag/m in 1964. The range of 24-hour averages for
a single station is often from a few micrograms per cubic
meter to over 150 jug/m . For example, in Rio Arriba
County, New Mexico, the lowest reading was 9 jug/m and
the highest 257 jug/irT during 1964.. . . . Although
attempts to measure SC> at twenty-eight non-urban
stations have been conducted over the past six years, the
results of this effort forced the conclusion that it will
be essentially impossible to establish any sort of base-
line, nationally or area-wide, using the standard reference
method. This is largely due to the insensitivity of the
technique." (Reference 12 at pp. 5-6). These data also
influenced EPA's dicision not to have a single inviolate
deterioration increment for the whole country.
-------
— T_ 7 —
existing air quality in clean areas of the country is not
available nor can such data reasonably be obtained.*
6. Local definition plan
This plan would have prevented significant deteriora-
tion by requiring local determination, on a case-by-case
basis, of the significance of the air quality impact of
major new sources. This plan recognized the variability
between areas and called for a subjective decision-making
procedure to be implemented at the local level. A major
disadvantage is that the local definition plan uses what
is essentially a "sliding baseline," and deterioration is
always measured relative to the current air quality. Hence,
there is no control over the ultimate level of deterioration,
which could progress in finite increments up to the level
of the secondary standards. Also, the long range impact
of deterioration is not restricted to the local area. The
regulations proposed on July 16, 1973, associated with
this plan required public comment from within "the area
significantly affected by the potential emissions".
The plan selected does not present these difficulties
although it refers to increases above a baseline level.
The increases are stated in absolute terms, not as a per-
centage of the baseline. Thus, the baseline may be taken
as zero and subsequent air quality changes determined by
modeling the air quality effects of increased or decreased
emissions which were not included within the baseline.
-------
- 18 -
However, it is entirely possible that the cumulative ef-
fects of sources in "growth-oriented" regions could have
a significant impact on the air quality of neighboring
"clean-air oriented" regions, and these neighboring re-
gions would thereby lose control over their own environ-
ment. Although the State, rather than the local popu-
lation, has final -authority for the definition of signi-
ficant deterioration (which tends to mitigate this con-
cern) , the impact of sources in one region on the air
quality in neighboring regions remains a problem under
this plan which could be especially serious in interstate
situations.
7. Establishment of absolute air quality levels
below the national standards.
In effect, this plan would result in more restrictive
air quality "standards" which would plaice an absolute
ceiling on air quality concentrations. However, the
amount of deterioration which would be permitted under
this plan would vary significantly throughout the countr/
depending on existing air quality le;vels. In very clean
areas, a substantial amount of deterioration would be
permitted, while in areas that presently exceed the selected
"standard", absolutely no deterioration would be permitted.
Therefore, this plan could be very inequitable, especially
when applied on a nationwide basis.
-------
- 19 -
IV. Selection of air quality increments
The Class I increments represent extremely stringent
deterioration criteria, and application of these increments
would prohibit the introduction of even one small fossil
14
fuel fired power plant or municipal incinerator unless
existing emissions are reduced. Strong incentives there-
fore exist in Class I areas for improved emission control
technology to reduce existing emissions and for introduc-
tion only of low-pollution development. Although Class I
could be applied to a semi-urban or urban area in which
it was desired to inhibit further development, it is
anticipated that Class I would normally be applied to
those ultra-clean areas such as national and state forests
and parks, and other recreational areas in which it is
desired to prevent essentially any deterioration of
air quality.
The Class II increments were established to permit
what, in the Administrator's judgment, constitutes moderate,
well controlled growth. Large point sources, aggregation
of several moderately-sized sources, and large-scale
general growth would be permitted only in Class III areas.
-------
- 20 -
For example, the power plant size that normally would be
permitted at a single location within a Class II area
is approximately 1000 MW. Assuming that such a source
used up 90% of the allowable increment in a Class II area,
a similar source could not be located within 25 miles of
the first plant. Similarly, typical coal gasification
plants, oil shale processing facilities, and petroleum
refineries would not be expected individually to exceed
the Class II increments in most areas. However, the
Class II increments would prevent the aggregation of
such sources within close proximity of each other. It
should be noted that the foregoing statements are generali-
zations, since the impact of the Class II increments with
respect to the source size permitted and the distance
which must be maintained between major sources is highly
dependent on the meteorology, topography, and stack heights
applicable in each specific situation.
Comments were specifically requested in the August 27,
1974, proposal as to whether the Class II increment should
be doubled. Power companies generally supported such a
change, "Awhile other comments from the industrial sector
indicated that the increments were adequate for well-
controlled growth. Power companies indicated that many
-------
- 21 -
new plants would be much larger than those which would
be allowed in a Class II area (approximately 1000 mega-
watts) , and that the Class II increment ought to accomo-
date such development. None of the comments presented
any reasons for permitting such development in a Class II
rather than a Class III area, except that the initial
designation of all areas will be Class II. The Admin-
istrator continues to feel that a Class II increment should
be compatible with moderate, well-controlled development
in a nationwide context, and that large-scale develop-
ment should be permitted only in con5unction with a
consicious decision to redesignate tne area as Class III.
V. Pollutants covered by the regulation
Many comments have criticized the omission of carbon
monoxide (CO), nitrogen dioxide (NO ', Hydrocarbons (HC),
and photochemical oxidants (Ox) from the regulations. As
indicated on July 16, 1973, and August 27, 1974, and
in previous actions involving indirect, source review
(38 FR 29893 at 29894, 39 FR 7270 at 7272, and 39 FR
25292 at 25295), existing mathematical modeling proce-
dures are not adequate to determine rhe impact, prior to
construction, of individual sources on air quality
-------
- 22 -
concentrations of reactive pollutants (NO,, and Ox) .
These pollutants are formed in a complex atmospheric
reaction which is dependent on a number of variables such
as temperature, humidity, solar intensity, concentration
and chemical reactivity of hydrocarbon precursors, and
concentration of nitrogen oxides. Diffusion/photochemical
models are still in the research stage and generally have
not focused on the impact of individual sources. There-
fore, the only presently available technique for relating
emissions to air quality ::or these pollutants is the area-
wide proportional model used for demonstrating the adequacy
12
of control strategies.
The proportional model assumes that air quality is
proportional to emissions in the area. To determine how
much emissions can be increased without increasing ambient
air quality levels by more than a given increment, it is
necessary to know at some point in time both the air
quality and the emission levels in the area involved. The
permissible emission increase is then equal to the allowable
percentage increase in air quality levels.
As indicated above, the air quality data which must
be available in order to use the proportional model are
-------
- 23 -
very limited in presently clean areas (even more so than
for TSP and S0_). Furthermore, the lower the levels are
of identified emissions or air quality, the less accurate
the results that can be obtained from using the proportional
model. In very clean areas with virtually no baseline
emissions, the addition of a major source may increase
emissions in the area by several orders of magnitude. In-
creasing available air quality data (assuming that such
data are available) by several orders of magnitude would
yeild highly questionable results, for several reasons.
First, the accuracy of air quality data at low ambient
levels is very poor. Multiplying very small air quality
levels by several orders of magnitude multiplies the
substantial initial range of error by a similar amount.
Secondly, if air quality is presently below the detection
limit of the measurement method, no prediction can be
made (i.e., the emission increase would be multiplied
by zero). Finally, there are natural sources of hydro-
carbons which contribute to the photochemical reaction;
however, the magnitude of these emissions cannot be
adequately quantified at this time. Thus, identified
emission levels in "clean" areas may be far less than
-------
- 24 -
actual emission levels contributing to the air quality in
the area. In contrast, the proportional model is adequate
for control strategy development in urban areas where
measured air quality data are available and the aggregate
impact of many identified sources is being anlayzed. How-
ever, it is inappropriate for analyzing the incremental
impact of individual new sources, particularly in "clean"
areas.
Although CO is stable; and emissions can be ade-
quately related to air quality, CO is almost exclusively
emitted by mobile sources (85% in 1970). Therefore, in
order to treat CO in the same manner as TSP or S0_, signi-
ficant deterioration considerations would need to be in-
corporated into EPA's indirect source review procedure
(39 FR 25292). However, Congress has recently expressed
reservations about the indirect source review program
by prohibiting EPA from using fiscal year 1975 funds for
any program which could restrict parking facilities.
Even if the question of indirect source review is favorably
resolved, only CO can be adequately dealt with through
this program.
Other factors which mitigate against the need to
specifically include the automotive pollutants under these
-------
- 25 •-
regulations are as follows:
1. The Federal Motor Vehicle Control Program will
result in sizeable reductions in emissions on a nation-
wide basis for CO and EC for many years into the future.
This will be true even if the auto emission standards
are frozen for 5 years at the level presently required in
California, as the Administrator has proposed in his re-
commended changes to the Clean Air Act. That is, the
nationwide emissions of CO and HC will be continually
reduced through the mid-1980's, and the effect of any
suspension of the present statutory standards would only
1 8
reduce the rate at which emissions are decreased.
2. A basic requirement for sources subject to the
regulations is the application of Best Available Control
Technology (BACT). The Federal Motor Vehicle Control
Program accomplishes this for individual motor vehicles.
New source performance standards (NS?S) have already been
established under Part 60 of this chapter for many of the
source categories subject to the regulation. Where practi-
cable, emission limitations for CO, NO , and HC have been
X.
promulgated for those sources presently subject to Part 60.
Although some of the source categories are not yet included
-------
- 26 -
in Part 60, either (1) those that are not covered are
not significant emitters of CO, NO , or HC; or (2)
X
control technology for these pollutants is unavailable
or an emission limitation is impractical (e.g. since HC
emissions from coke ovens cannot be adequately measured,
a numerical HC emission limitation is not feasible for this
type of source).
3. At ambient air quality levels for CO approaching
the national standards, the only known or suspected effects
are those relating to human health. At such levels, there
are no known effects on plants, animals, materials, visi-
bility, or other non-health related effects. With respect
to the health effects of CO, the National Academy of Sciences
has recently concluded that "(o)n the basis of data avail-
able at this time, we cannot identify any population group
that would be adversely affected by carbon monoxide ex-
19
posure from ambient air if statutory standards are met".
An additional step which could be taken to minimize
emissions of CO, NO , and HC appears to be in the area of
X
minimizing vehicle miles of travel (VMT). Plans for re-
ducing VMT and minimizing future VMT growth have been
developed as part of the Transportation Control Plans (TCP)
submitted by the States or promulgated by EPA. Since the
TCPs focus on major metropolitan areas, the flexibility avail-
able in designing these plans would be more limited when applied
-------
- 27 -
to rural and outlying areas. It is clear, however, that
comprehensive transporation planning offers an appropriate
mechanism for minimizing VMT growth in such areas. It
is not clear, however, how EPA might become involved in
comprehensive transportation planning throughout the
country under these regulations, although states may
wish to consider such an approach when developing their
own plans to prevent significant deterioration. States,
of course, are not precluded from including other more
comprehensive measures for dealing with HC, CO, and NO
X
in their own plans.
Some difficult additional questions arise as to how
this concept of VMT minimization could be incorporated
into these significant deterioration regulations. Would
the addition of a VMT increment, similar to the air
quality increment approach used in these regulations, be
appropriate? How would it be enforced? The Administrator
has solicited additional comments on this issue and may
modify the regulation at a later date if workable pro-
cedures in this area can be developed.
VI. Sources subject to review
The 18 sources categories which are covered by the
regulation, except for fuel conversion plants, are the
-------
- 28 -
largest present emitters of SO- and TSP on a nationwide
20
basis. Fuel conversion plants (coal gasification and
liquefication, oil shale processing, etc.) were included
due to their significant growth potential, particularly
in presently clean areas. The air quality impact of
average-size sources subject to review is analyzed in re-
ferences 10 and 16. The air quality impact of sources
not included in the 18 categories is taken into account,
since the total air quality deterioration above the base-
line is taken into account when an application to con-
struct a new source in one of the 18 categories is
reviewed.
Although the preconstruction review of smaller
sources is not so critical during the initial phase of
this program and such review would contribute greatly
to the difficulty of initial implementation of the
program, inclusion will become more important as the air
quality approaches the
-------
- 29 -
allowable increment. Therefore, the Agency is considering
the addition of'sources to the list, perhaps using the addi-
tion of sources subject to the new source performance standards
under 40 CFR Part 60 as a way of "phasing in" additional sources
to be covered by these regulations.
VII. Air Quality Monitoring Requirements
The July 16, 1973, proposals included a provision re-
quiring that source owners must install air quality monitoring
instruments to assess the impact of their facility. This provi-
sion was not included in the final regulations. The accuracy
of the measurement methods in relation to the air quality
increments and the year to year variability of air quality
12
data severely limits the usefulness of the data collected.
In addition, the use of air quality monitoring to determine
whether the applicable increments are being maintained assumes
that the baseline air quality from which deterioration is
measured (i.e., 1974 air quality) is accurately known. As
indicated in Section III.5 above, such baseline data are not
generally available. Baseline air quality data are not needed
in order to implement the regulations, since significant
deterioration is defined in terms of air quality increments
rather than absolute air quality levels.
In actual practice, it is anticipated that assessment of
the available increment will normally be accomplished through
21,22
an accounting procedure whereby atmospheric modeling of
individual sources will be used to keep track of the availeble
-------
- 30 -
(or "unused") increment as sources and emission are increased
or decreased. After a source has been given permission to
construct based upon a diffusion model, it would be very
inequitable to require the source to shut down because the
EPA or State approved model was inaccurate. Therefore, air
quality data are not needed either to define an air quality
baseline or to assess the amount of deterioration that has
occurred.
VIII. Impact of the regulations on general growth and develop-
ment
In general, the impact analyses that have been performed
focus on the Class II areas, since the Class I increments are
intended to prohibit all significant growth and no new re-
strictions are created by the Class III "increments." Reference
9 is a summary of the impacts of the various options proposed
on July 16, 1973, and is based on the date contained in References
7, 8, 9, and 10. Alternative I discussed in these references
(the single air quality increment plan) is virtually identical
to the final Class II increments. Additional information on
this issue, including a discussion of whether the regulations
might inhibit development of new communities and increase
urban sprawl and congestion, is contained in Reference 23,
pp. 8-13.
IX. Impact on power generation and new energy sources
Reference 15 provides the basis for concluding that a
1000 MW power plant will generally not exceed the Class II
-------
- 31 -
increments. This reference also provides the basis tor in-
creasing the Class II increment for S02 r'rom 300 ug/m^, maxi-
mum 3-hour average, to 700 ug/m3.
A study was done to determine whether the size of a
power plant that could be constructed in a Class II area
would be affected by the use of flue gas scrubbers to meet
the new source performance standards (NSPS). It was concluded
that the ground level concentration resulting from a cold
scrubber plume would be increased by 20-80%, depending on
°the meteorological conditions assumed, as compared to an
unscrubbed, and therefore warmer, plume. However, for a hypo-
thetical 1,000 MW power plant meeting NSPS with a scrubber
and assuming no plume reheat, the predicted concentrations
for specific meteorological conditions in the upper Ohio
valley were still well within the Class II increments.
Reference 16 discusses the long-range transport of pollu-
tants, particularly in relation to the violation of the Class
I increments by a power plant in an adjacent Class II area.
It is apparent that a 1000 MW power plant can endanger the
Class I increment for S02 at distances of 50-60 miles down-
wind, depending on the persistence of poor meteorological
conditions.
Reference 23, pp. 21-23 summarizes the impact that the
Class II increments (equivalent to the original Option #1)
would have on the development of new sources of energy.
-------
- 32 -
X. Interstate boundary conflicts
Calculations have shown that because of the small air
quality increments specified for Class I areas, these levels
can be violated by a source located many miles inside an
adjacent Class II or III area. For example, a power plant
which just meets the Class II increment for SO,, could under
some conditions violate the Class I increment for SO,, 60 or
£
miles away. Under the significant deterioration regulations,
a source would not be allowed to construct if it would violate
an air quality increment either in the area where the source
is to be located or in any neighboring area in the State.
Therefore, wherever a Class I area adjoins a Class II or III
area, the potential growth restrictions, especially for power
plant development, extends well beyond the Class I boundaries
into the adjacent areas. A similar situation exists, to a
greater or lesser degree, wherever areas of different classi-
fication adjoin each other. Therefore, the area with the
less restrictive classification should include an additional
area at the periphery where it is clearly recognized that
development will be somewhat restricted due to the adjacent
"cleaner" area. As a result, a Class I redesignation could
be fairly limited in size, yet the adjoining Class II or Class
III areas would need to cover a substantial area in order to
fully utilize the Class II or III increment. Again, it
should be clear that the Class II or III increment could
only be fully utilized toward the center of the area and that
-------
- 33 -
at the periphery, permissible siting of new sources will be
dictated by the adjoining Class I area rather than the Class
II or III increment.
The distance a large source would need to be located
away from a Class I boundary is more dependent on the
meteorological conditions in the area than the size of the
source. Where very long pollutant travel times from the
source to the receptor are involved, the assumptions concern-
ing the persistence of wind direction ana atmospheric stability
are critical. At some point, it can be assumed that a receptor
will be virtually unaffected by a source, regardless of the
source strength, since the critical meteorological conditions
would not be expected to persist long enough to move the pollu-
tants from source to receptor for any significant period of
time. This distance is, of course, dependent on local
meterological conditions, but for most areas the maximum
distance would be 60 to 100 miles.
The regulations include provisions for dealing with
problems created when a State or Indian Governing Body wishes
to designate one or more of its areas in such a way that it
will have a negative impact on other States or Indian Reserva-
tions. These regulations provide that a State or Indian
Governing Body must take into account the effect of proposed
redesignations on other States, Indian Reservations, and
regional and national interests. Where no State or Indian
Governing Body protests the redesignation of another State or
-------
- 34 -
Indian Reservation, the Administrator will only review the
redesignation to determine whether it is arbitrary and
capricious. However, where a State or Indian Governing
Body protests a redesignation to the State proposing the
redesignation and to the Administrator, the Administrator
will take an expanded role of review in which he will balance
the competing interests involved.
XI. Best available control technology
In the original proposal, two alternative definitions
of Best Available Control Technology (BACT) were discussed.
Under both alternatives, a case-by-case review to determine
BACT was required of each source for which new source performance
standards were not applicable,. Under the first alternative,
the attainment of new source performance standards (NSPS)
promulgated pursuant to section 111 of the Act would be equiva-
lent to application of BACT for all sources except for sulfur
dioxide emission from fossil fuel-fired steam electric power
plants: for these plants a C5ise-by-case review was required
to determine if emissions could be reduced to below NSPS,.
Under the second alternative, fossil fuel-fired steam electric
power plants were treated like all other sources for which
NSPS are applicable.
In the final regulations, the second alternative is
incorporated: power plants would not be subjected to the
special BACT review because requiring such a review would be
inconsistent with the NSPS. Further, the requirement for
application of BACT for control of hydrocarbons, oxides of
-------
- 35 -
nitrogen, and carbon monoxide has also been deleted because
this requirement was inconsistent with the restriction of
these regulations to particulate matter and sulfur dioxide.
As discussed in Section V, BACT for these pollutants is already
required, to the extent presently feasible, for the 18 source
types subject to review.
XII. Initial classification and criteria for reclassification
There was considerable divergence of opinion over the
initial classification of all areas. Industrial groups
generally supported an initial designation of Class III so
as to minimize disruption of projects scheduled to commerce
construction in the near future. Environmental groups supported
an initial designation of Class I, fearing that a Class II
or III designation would permit air quality deterioration of
some clean areas before States could act to redesignate areas
to a more restrictive classification. The Administrator con-
tinues to feel that an initial Class II designation repre-
sents the most reasonable compromise between these widely
differing positions. Also, since the regulations apply only
to sources which commence construction after June 1, 1975,
the Administrator feels that this deferral should reduce
disruption to the industrial sector while permitting States
sufficient time to consider reclassifying any area either to
Class I or III before requests for approval must be acted upon.
In order that the Administrator will have an adequate
basis for determining whether an application to redesignate
-------
- 36 -
an area should be approved or disapproved, the regulations
require that the necessary information be a part of the hear-
ing record on the proposed designation. Specifically, the
hearing record must show that the social, environmental, and
economic effects of the proposed redesignation have been
evaluated for the area being reclassified as well as for
adjacent areas and that regional and national interests have
been considered. EPA will disapprove a proposed redesignation
if the State has not properly examined the effects of the
redesignation or has arbitrarily and capriciously disregarded
such effects.
It is not anticipated that there will be large scale
redesignations to Class III, for several reasons. First,
the initial Class II increment will accommodate most moderate,
well controlled development, as discussed above. Secondly
since a State must fully discuss the reasons for a proposed
redesignation at a public hearing, a State that proposes a
redesignation to Class III without showing a definite need
for such redesignation would likely meet with considerable
resistence at the public hearing. Finally, should a State
submit such a proposed redesignation without adequate justi-
fication, EPA would disapprove the proposal as having arbitrarily
and capriciously disregarded the environmental effects of
the redesignation.
-------
- 37 -
XIII. Effective date for source review dnd determination of
significant deterioration
The December 5, regulations establish the air quality
concentration during 1974 as the baseline for determining
whether significant deterioration has occurred. The Sierra
Club has contended that significant deterioration should be
measured from May 30, 1972, the date that the implementation
plans were to be approved or disapproved by the Administrator.
Because of the far-reaching implications of these regulations
on growth and development, it is important that the allowable
deterioration increments be wisely used and that the allocation
of limited air resources be planned and managed with full con-
sideration of the impact on future growth and development. To
make the baseline for measuring deterioration retroactive to
May 1972 would require that the effect of sources commencing
construction since that date be counted against the allowable
increment. Yet the States, in granting such sources permis-
sion to construct, had no opportunity to consider the full
implications of their decision (such implications being only
now available). Furthermore, to subject sources constructed,
or which started construction, during the period from May 1972
to June 1975 to the significant deterioration requirements
(which could result in clofeing those sources violating the
allowable increments) would, in the Administrator's judgment,
be grossly unfair.
The Sierra Club has suggested that as a minimum, such
sources be required to install BACT on a retrofit basis.
-------
- 38 -
There would be relatively little benefit to be gained by
such a requirement, since the overwhelming majority of emis-
sions from sources subject to significant deterioration review
have been subject to new source performance standards since
12
August 1971. New source performance standards have been
in effect for the following sources commencing construction
after August 17, 1971: fossil fuel fired steam electric
generating plants (by far the most important source category
both in terms of the number of new plants constructed and in
terms of emissions of both pcirticulate matter and SOo) , in-
cinerators, portland cement plants, and sulfuric acid plants.
NSPS have been applicable since June 11, 1973, to petroleum
refinery catalytic regenerators and process gas combustion
and steel mill basic oxygen furnaces.
For those sources commencing construction since May 1972
and which are not subject to NSPS, emission control equipment
has, in most cases, been incorporated as part of the original
design. Although such equipment may be quite efficient and
may represent a significant investment, it may not be equivalent
to BACT. The retroactive application of BACT could then re-
sult in scrapping virtually brand new control equipment in
order to obtain only modest additional emission reduction.
In the Administrator's judgment, this result is both unwise
and inequitable.
-------
- 39 -
REFERENCES
1. "Sierra Club et al Litigation - Significant Deterioration,"
B. J. Steigerwald, September 27, 1972.
2. "Draft Summary Report on Suspended Sulfates and Sulfuric Acid
Aerosols," EPA Office of Research and Development, March 1974.
3. "Summary of Responses Received Regarding the Prevention of Significant
Deterioration."
4. "Summary of Responses Received Regarding the August 27, 1974, Proposal
to prevent Significant Deterioration of Air Quality."
5. "Summary of State Responses on 'Significant Deterioration' Proposal."
* 6. "Sierra Club Proposal," November 8, 1973.
7. "The Impact of Proposed Non-Degradation Regulations on Economic Growth,"
Volumes 1 and 2, Harbridge House, Inc., November 1973.
8. "Implications of Non-Degradation Policies on Clean Air Regions: A Case
Study of the Dallas-Ft. Worth AQCR (215)," U.S. Dept. of Commerce,
May 1974.
9. "Analysis of the U.S. EPA's Proposals to Prevent Significant Deterioration
Relative to the Development Outlook for New York State," New York State
Department of Environmental Conservations October 1973.
10. "Impact of the Proposed Non-Degradation Alternatives on New Power
Plants," TRW, Inc., September 28, 1973.
* 11. "Economic Growth and Development Impacts of Proposals to Prevent
Significant Deterioration of Air Quality."
* 12. "Scientific Factors Bearing on Regulatory Policies to Assure Non-Degradation
of Air Quality."
* 13. "Availability of Air Quality Data in Areas Generally Below the NAAQS."
* 14. "Technical Data in Support of Significant Deterioration Issue."
* 15. "Non-Degradation and Power Plant Size," J. A. Tikvart, August 12, 1974.
* 16. "Significant Deterioration in Zone I Areas and the Relative Location of
Power Plants," J. A. Tikvart, October 15, 1974.
* 17. "Discussion Paper on the magnitude of the Class II Increment in the
Significant Deterioration Regulations."
18. "Air Quality Impact of Alternative Emission Standards for Light Duty
Vehicles."
-------
- 40 -
19. "Air Quality and Automobile Emission Control," National Academy of
Sciences, September 1974, Volume 1, p. 32.
* 20. "Emissions of Sources Subject to Significant Deterioration Issue."
21. "Guidelines for Air Quality Maintenance Planning and Analysis,
Volume 10: Reviewing New Stationary Sources," EPA, September 1974.
22. "Guidelines for Air Quality Maintenance Planning and Analysis,
Volume 12: Applying Atmospheric Simulation Models to Air Quality
Maintenance Areas," EPA, September 1974.
23. "Findings of Task Force on Significant Deterioration," R. G. Rhoads,
December 20, 1973.
24. "The Largest Annual Average,, Maximum 24-hour and Minimum 3-hour
Concentrations of Sulfur Dioxide Produced Per Year by a Modern
1000-MW Electric Power Plant Meeting the New Source Performance
Standards for Sulfur Dioxide Emissions," Enviroplan, Inc., 1974.
References with an asterisk are reprinted as attachments to this
document. The other references, which are too lengthy to reprint
as part of this document, are available for public inspection at
the EPA Freedom of Information Center, 401 M. Street, S.W.,
Washington, D.C., 20460.
-------
Reference 6
SIERRA CLUB PROPOSAL 8 Nov 1973
The plan proposed by the Sierra Club to prevent significant deterio-
ration of air quality is based on a volume averaging or mixing zone type
of approach. (Attachment 1 is the testimony given by Mr. L. Moss at Washington)
Under this plan the pollutant concentrations would be limited to certain
specified incremental levels in a spheroid surrounding the point source.
Figure 1 depicts the situation. Table 1 shows the allowed increases in con-
centration allowed. In addition to the incremental increases in the volume
around the sphere, they also recommend that the total emissions over an
2
AQCR be limited to increases of 1 Ton/yr/mile for all criteria pollutants
2
except CO where 20 Ton/yr/mile would be allowed.
Al 1 cwed_ Emi;;sions Under the Si erra Club Proposal
The actual allowed emissions under this type of plan will vary from
location to location but if some simplifying assumptions, are made, it is
possible to compute the emissions allowed under the increment. (Attachment 2
shows info nation supplied by Dr.. M. Williams on behalf of the Sierra Club.)_
For purposes of their discussion of potential impact the Sierra Club assumed:
1. an annual wine! speed of 6 m/sec (13.2 ra/hr) and
?, a 250 m stack height.
They felt chat both of those assumptions were reasonable for power plants
in the southwest. While this ts true, it appears that the annual wind speed
may be too high for many areas and the "allowed development' less.
Following the Sierra Club assumptions the maximum allowed emission rate
is 5.35 T/day or 1950 tons/year. This is limited by the annual increment
* , -3
of 4 ug/m .
41
-------
42
-C
CM
+
CO
t=
ro
CM
O)
CD
p.
V
o
c
>• CL
O)
t- O
O C
o
C CJ
o
•p- en
+* c
03 -r-
-i-> cn
C na
O) S-
(/) O)
O) >
s- <
Q.
cu
o:
-------
43
TABLE 1
INCREASES IN POLLUTANT
ALLClvJ') UNDER SILRKA
cor;ccNTRAV
CLU3 PRul'OS
ALLOWED INCREASE (ug/M
POLLUTANT
TSP
so2
HC
N0x
CO
1 HR
100
TOO
100
100
2000
24 HR
50
50
50
50
1000
; ON*
AL~
3) FOR
ANNUAL
4
4
4
4
80
* All concentrations are averaged over a one kilometer sphere
around the source
-------
44
Req u ired LcveIs of _C o n_tro 1_ Joohno!ex iy
In their presentations at. public hearings and in written comments, the
Sierra Club indicated that a large amount of industrial development would be
allowed. (Attachment 3 shows development the Sierra Club claims is allowed.
Attachment 4 is material submitted by Dr. M. Williams to support these assump-
tions,) Their assumptions on control technology were reviewed by Control
Technology, OAWP which stated, "For the most part the Sierra Club has assumed
the host of all possible situations—an abundance of natural gas, raw mate-
rials wiuh low pollution potential, and an ability to develop any level of
control technology. Many of these assumptions are unrealistic particularly
in view of the fuel shortage. Costs of control have received no considera-
tion at oil." Table 2 shows some of the plant types, analyzed under this
procedure, the Sierra Club assumptions on control technology and comments
of EPA engineers.
The impact on coal fired power plants and petroleum refineries is
e^p^c-'^lly sever*"-;. For example, if NSPS were applied to a cool fired power
plant the maximum plant size would be 52.8 MW. If state- of the art stack
gas cleaning (80-05% sufficient) were combined with O.b4% Western coal, the
maximum allowaole plant size would ba 200 MW. The Sierra Club claims that.
97% SCL removal on 0.5% western coal is available and as such the sulfur
dioxide limitation would allow a 2200 MVJ power plant to be built.
A similar situation exists 'if the oxides of nitrogen are considered.
A plant which meets NSPS (0.7 Ibs NO/106 BTU) would be limited in size to
A
about 94 MW. If NO emissions are limited to 0.45 lbs/106 BTU (a level
possibly attainable) the plant would be limited to about 150 MW. If NOV
A
-------
_J
«X
to
O
Q-
O
o;
Q_
G")
4->
C
QJ
O
C.J
OJ "O 4-
r— i— O
JD O
fO 4- QJ
r— 4~>
•t- CO ro
•"C i — 4-J
.> G*>
ro ra
3:
>> OO O
r— 4-> i —
i— C. QJ
rO CU JO
C VI
O QJ GO
•r- S_ C
4~> D- O
rO QJ T-
C S- GO
GO
+J r— -r-
0 0 E
C i- QJ
4-J
i — c c:
rC1 O •! —
O (_>
O QJ
to V)
to ro
Vs- CU
£'J. co i- •
LO CT. 0 4->
• • QJ i-
C CM -D ro
O
s_
40
c~
O
o
4->
c
QJ
l/l
QJ
S-
O.
4—
O
4-^
•i —
E
•i —
i —
QJ
4->
^3
i —
O
GO
_c
*i
«
cr
ro
i — i-
r— QJ
ro 3 >
0 , .
4-> i — cu
3 4-J r—
i— O JD
i — CJ ro
04-r-
Q. i. -r-
Qj re
r— D- >
r— ra
ro CU
4-J I/)
4- re •!-
0 i-
QJ r—
>, O- ro
Cr, o 0
0 u
1 — GO
0. E S.-
C QJ 3
.1= 4-J 4-
CJ VI r-
O >, 3
4-> GO V>
'" 45
V) QJ
4-J r~ ||
C .O
It?
i—
C~
$~
QJ
3:
o
a.
*X^J •
QJ 3:
^_ y~
•,—
4- O
CM
i — 1£J
rci
O QJ
0 JO
5 O
QJ 4->
C
"O
4— QJ
O 4-*
O
QJ QJ
N Q.
•r- X
IO QJ
(D 4-1
i — C
•r- rO
ra r—
> Q-
ra
O
to C
>>•-
ra 4->
S to
i — •»—
ro X
OJ
4-J
O 4-
c o
r— C
•r"" Q
O -r-
V)
S- C
H3 03
4- 0.
r— X
3 UJ
GO
•
5 O'l
o >
—i «t
4->
O
*o
G1 i-
ra 4J 4->
CT. O C
c: c
t— 0
ro GO
S- i— <4-
3 O O
4-> i-
ro 4-^ i —
z: c QJ
O >
(J QJ
— 1
CO X
O
S£^ •
" un
>>
ro •
"O *3~ **
-\ ti
r— QJ
ja "4-J
JO CD ro
r— i-
O JD 4-J
C". rO V)
O i— C
"•r- O
O ro E
LO > CJ
i — ro 13
Q)
GO
C
QJ
Q-
X
QJ
QJ
tr.
!_
ra
r—
ro
4-'
3
jQ
QJ
r—
^£}
•t —
GO
ro
QJ
4-
>,
r—
.Q
•i —
GO
GO
O
Ci.
•
QJ
r—
JD
ro
O
GO
ro
a>
a:
V)
c
o
4J
E
v>
to
, —
o
i-
4_>
{—
O
V)
ra
"a.
s_
o
4-
QJ
N
GO
y—
cn
to
QJ
X3
•^
QJ
O
QJ
X
f_
ro
5^
O
&o:
•r- K"'
GO
Q.I
T3 •
CO
QJ »
4-' GO
U i-
QJ a,'
O-i —
X -r-
LU O
• QJ
CM S-
">T
ro O
C3 4-*
•^^^
LL. i —
O ro
s; o
c
Lf) V)
CM QJ
GO
II 3
c:
,
r— JD
4-J QJ
0 > T3
C QJ QJ
VI T-
ro £T ra
JC O i —
•r- GO
"O tO
CJ GO QJ
ro E 0
4-> QJ JC
10 4->
CM
GO O C
ro to ro
_c
r— 40
O •
J- «3 S-
4-J CJ
c « S
O "0 O
(-) QJ i —
CM
CM
UJ
_J
ca
r3^
h-
cc
'- '.
_J
CJ>
-
cc
CC5
UJ
t—
<^
Q-
•;?.
1 — 1
C^
00
c.
•
CM
Vi
CI
C
• r—
•*->
O
r:
t~,
to
t-
.
CO
»1
r—
ro
O
(J
I/)
r.o
>
o
*u
Q)
>
O
rr
QJ
O'
C. "
to
t-'
LO
CO
1
c
CX
.
CM
f
t —
ro
O
0
oo
^r
i.:',
C'
GO
J~~)
i —
ro
•
c;
(•o
»»
TD n.
CU I—
> DC
q
cu o
4- t—
CXI X
o c
GO ;_;
i
^3
4->
(C
.
CM
^
QJ
"0
H5
s_
u
C/)
^
.
C
GO
i-
CJ
r—
•( —
O
JD
4-J
ro
_t—
O
4->
GO
rU
cr
,
ro
i-
CT.
O
, —
O
4->
C
O
U
X(/l
C !-
zr GJ
j—
• c
CO JL
, —
ro
O
E
QJ
CM
O
to
fe<
o
cr,
o
c
»l—
s_
., —
4-
to
ro
CJ
4-> GO
O J~
3 QJ
O ^
S- 0
D_ JD
u~.
O-
_J
Cl-
a -T!
N lj
13 T
CJ i-
"
O OJ
I — • *r -
1 — (/)
O
CVI
CM
CM
r\j
i
O re
O XJ
C\J
C-j
CO
CO
ro
u_
CM
O
to
C3
n.
G-) TJ
h- QJ
>> QJ
4_J "Cj" c"" -t—
C QJ O V,
r3 S- c
,._. .,- 4- O
•'I ' • - (. ;
o o
rx —-
CM
O 13
GO QJ
t-
" Q)
Cu X3
f-
CJ
(.'
^
>—
t" ]
CI
•I—
4-
QJ
CX
QJ
4->
f "~
QJ
E
^j
CO-
CX.
o
CJ
S-
QJ
GO
ro
O
o^
.
ra
-£Z
I/)
rd
f—
I .'-.
b
JD
C
O
(O
0
£
•t—
VI
'O
t-'J
r —
ro
O
0
CO
-------
46
if further reduced to 0.3 lbs/10 BTU (a level possibly attainable by one
manufacturer) the maximum power plant size would be 220 MW.
Thus the maximum coal fired power plant which could possibly be con-
structed under the Sierra Club proposal using loA'-sulfur coal, stack gas
cleaning and ideal furnace design for NO control is 200-220 MW. In areas
A
where very low sulfur coal is not available, the maximum plant size would
be less than 100 MW. Current planning of coal fired power plants estimates
that the average plant size is to be about 600 MW.
Economic Impact of Sierra Club Proposal
It is extremely difficult to estimate the economic impact of the Sierra
Club proposal, because the actua" allowed emission rate will vary from loca-
tion to location. However, it appears th:»t many industries and some light
development would bo able to bo undertaken withc-.it a substantial incrocir-c
in cost over what now is required.
The Sierra Club proposal cioes_ appear to have a \ery substantial imptct.
on the power and oil industries. If, as described previoi'sly., the maximum
coal fired power plant size is limited to 220 "'.K1., constraining coal fired
plants to these levels would result in either a rapidly decreasing rel'ioi'.ce
on coal and a switch to nuclear po'ver or very large numbers of very small
power plants at a much higher cost. In dddltion to the-.1 higher capital cost,
the resulting increases in distribution and coal hauling facilities would
also result in considerably higher costs.
The restrictions applied to the size of new or expanded oil refineries
would also impose a substantial economic burden, It would result in greater
numbers of small refineries with resulting higher transportation and
-------
47
distribution costs. In addition, to build eve;" these small refineries
would require the consumption of scarce natural gas and thus divert this
from other uses.
Severe limitations are also placed on conventional copper smelters.
Maximum allowed size under the Sierra Club proposal is 88 T/day or 32,000
T/yr; average design size for smelters is 100,000 T/yr. The hydrometalur-
gical processes are not available for all types of copper ores.
Technic_a]_ Piffjcul ties i n Implementfng Sierra Club Proposal
One point emphasized in the Sierra Club proposal was that areas where
existing sources could be controlled there would be much greater room for
the expansion of new industrial processes. While this is probably true, to
some degree, there are two major technical problems associated with carrying
out the type of proposal suggested by the Sierra Club. The first of these
involves the monitoring of pollutants at the low levels and in the volume
suggested by the Sierra Club. Current instrumentation for sulfur dioxide,
niv.roqen oxides •, and hydrocarbons operating at stationsrv ground level moni-
toring stations is not sufficiently accurate to distinguish the proposed
Siena dub increments from random raise associated v/ith the day-to-day
operation of the instrument, To adequately handle th^ type of monitoring
required by the Sierra Club would require that monitoring be done from
instrumented helicopters or aircraft. It is not presently possible to
perform this type of analysis.
Monitoring could possibly be carried out by remote sensing devices such
as the lider but presently these also do not have sufficient accuracy.
-------
48
The second major difficulty with the Sierra Club proposal lies with
the ability to do the type of modeling required. While as the Sierra Club
states the modeling required under their proposal for a single isolated
source is relatively simple, the extension of modeling to multiple sources
areas becomes extremely difficult. The current gaussion diffusion models
were developed to predict, concentrations at a point on the ground or in
space. They are presently accurate to within a factor of two half the time.
The extension of these models to integrate the concentration in a spheroid
some distance removed from the source does not presently appear reasonable.
From an enforcement viewpoint the capability for using reductions from
existing sources probably only exists In the very near vicinity of large
isolated sources which can reduce emissions considerably.
Admi nl s trail ve D;'t f f i cuHi c-s i n Imp! ement i nq SJ_erra_ Cl_i'_b_ Proposal
The volume average concept espoused by the Sierra Cli.o i:-, an entirely
different process of air quality if'anegem^iit than tbM. currently ongoing for
the attainment ana mp.intfc-iicirice of NAAQS. To handle both the attainment and
mainte.'.rjncu of NAAOS by means of ground levels concentrations and to prevent
significant deterioration via the volume average techniques would appear to
drastically ccpiOuund the administrative burden.
In addition, the source size selected by the Sierra Club as requiring
review would impose a greater burden on regulatory agencies. The size
selected was 1 Ton/yr for all pollutants except. CO end 20 Tons/yr for CO.
This size would include all commercial and industrial facilities, develop-
ments of 50-100 homos heated with oil or 75-100 ho.r,cs heated with gas plus
-------
49
all highway, aiv.l many streets. It is impossible co estimate the number ot
sources whit.li would require review but. it would be a very large increase as
most state regulation cell for review only if the source is greater than
100 T/yr.
Summary
The pi on proposed by the Sierra Club allows for small increases in
concentrations over a 1 KM sphere around the. source. Their calculations
show that this plan would allow sources in a clean area to emit 5.35 T/day
(1950 T/yr) of any pollutant. Assumptions made by the Sierra Club in
estimating the impact were beyond the state of the art for oil refineriess
power plr.ins and coke ovens. Monitoring one1 mode ling techniques are not
available i/hiui can be used to keep track of changes in baseline air quality
to alloy/ fj»- inductions in emissions from existing sources. Plan would
impose adminif Lrc.live buroej'S in using an entirely different procedure for
attainment o.-id maintenance of NAAQS and prevention cf significant deterio-
ration.
-------
50
COMMENTS ON SIERRA CLUB PROPOSAL
1. Restrictions on Coal Fired Power Plants
In area where low sulfur fuel (0.5%) is available, new coal fired
power plants would be restricted to 200-220MW assuming that NO can be held
A
to 0.3 Ibs/ 10 BTU(may be poosible). Where low sulfur fuel is not
available plant size would be less than 100 MW. Both of these sizes are
generally not economical facilities. Reasonably sized additions to
existing plants may be possible. Basic thrust of the proposal seems to
be that all new coal fired power plants will have to be built only near
existing facilities. No consideration has been given to the cost of
control.
2. Restrictions on Oil Refineries
New refineries would be limited to 65-100,000 bbl/day if all possible
technology is used and no consideration is given to cost. Since current
construction and mosification averages 150,000 bbl/day, the most feasible
location for new refineries jeems to be near sites occupied by existing
refineries which can be very well controlled.
3. Monitoring Ability
Capability decs not currently exist to routinly perform the monitoring
needed to carry out the Sierra Club proposal. To require that this
monitoring be carried out by aircraft or remote sensing devices is not
presently feasible and if it be becomes feasible would be extremely
expensive.
4. Modeling Ability
Capability does not exist to perform the type of volume average
modeling in mul tip]? 'scu-'ce areas seemingly required by the Sierra Club
-------
51
to the degree of accuracy the proposal would require,
5. Administrative Problems
Current state strategics for attainment and maintenance of ambient
standards requires conrol technology and plans based on ground level
concentrations. The Sierra Club proposal would require a separate procedure
and the resultant monitor ing, modeling etc. to be carried out for new
sources. There does not seem to be any direct comparison between the
two approaches and a large additional workload would likely result.
-------
Reference 11
ECONOMIC GROWTH AND DEVELOPMENT IMPACTS OF PROPOSALS TO PREVENT
SIGNIFICANT DETERIORATION OF AIR QUALITY
This paper attempts to answer some critical questions on the extent
and nature of any economic growth limitations, altered development
patterns, and dislocations that may result from final promulgation of
three of the four proposed regulations to prevent significant deterior-
ation of air quality. The conclusions are based upon the findings of
four studies undertaken by EPA, the Department of Commerce, the Depart-
ment of the Interior, and the State of New York. The extent to which
the studies of selected industries and regions reflect the impact upon
the nation as a whole must be viewed with caution. Projected growth,
site peculiarities, changing economics, energy use, technology and
modeling techniques for assessing air quality impacts could all alter
the findings. We have identified these factors where they are relevant.
I. Summary of Findings
A. Alternative I: Ambient Air Quality Increment Plan
1. Regional Growth Impacts
a. The proposed increments appear adequate to allow projected
economic growth in most urbanizing areas to 1980, while significant restric-
tions and/or altered development patterns will be experienced by 1990.
b. Dispersion of development in urban areas in order to reduce
pollutant concentration could increase transportation-related emissions.
2. Isolated Industrial and Energy-Related Sources
a. Careful siting and the use of tall stacks would allow most
isolated industrial development.
S2
-------
1 > . ; . h I i I i ' )i i i I. e'. ;s t s ale i nij >' : e( I I;1 >ei) a n> ",' . - . ) > ;
if. in e • dar !:o comply w>tii the eir guaj.iay inci.va • it it must:
t* bui.lo a tailor st..ck
o apply stack gas scrubbing for SOp control
o incur transportation and land costs where associated
iridustr j ,;1 activities :nusL; be separated geographically
or where it uiuo t move larther from its market,
f> scale; cio.v'i its plnnt.
c .. Jncrei^,- at , ; 1 c-eats imposed by the air quality JncreramLS
: ,c -; . cas^: j-ej, : ! /• l\ sruil.l . Incremental costs Tor ,~>a-\cr
plant:-, x.liich I'l.-ot X'SPS or v/nsre power plant size must be limited
fiub: b/ i.j.ajly. 'iatl stc.c",s v/ould be a mjch lower cost alternative
v/c.i. % I-'-'-';.1 ciii.(.A\'' in such situations,
o, J"i.nirjg o^era i ic>ns afiSL'clr.rted \vitii new cement / copper,
and i re n aad. otcraJ plai-ts violate the proposed TSP increment in
B. A! iarr. i v- El: }yni ss aois tJ.-nsi/ty P-I":
a. Grov.aaa restrictions can be severe
o Dj>3 las-Fort .Vortli violates the TSP increment
in 6 years .
o Dev"l op> "»ont is restricted in most of New York.
State by 1980.
b. The impact on regional growth varies widely wi th the
sixo o/ the overaglng area and the existing emission baseline.
Region;;! inequities arise with averaging areas significantly
smaller or larger than the proposed 10,000 square miles.
c. Po>.A-r plants absorb a major portion of the allowable
increments in giowth areas, liov/ever, the restriction of access-
ible etcotiic i. o\.'cr in turn restricts industrial and commercial
de-- :'loy , -na . Growth restrictions can be severe in such
in.J a r 'a .:. .
-------
54
d. Norlhpj i. ("irv.'i S'l.jin- power complex • Id bo
res 1 r it ! c -d.
2, ']';o~Jat_f d ."* > <.';'••. M ' L! 1 .JILML 1''"1- r9Y~Iic'J'^i °' <.'(J.uZ.c-L'\i
a. An emission ] i mj 1 ation p.l an impact,; !.,;>st heavily
upojj powcr : 3"I ;j nl. 13.
b, la the abseno • of baseline aberrations industrial
grov/th ];^/;')ected i.o 19:-jO would not hu affected by the increment.
c- v/ith the exception of power plants, no new industrial
,sources violate the increment
«p most growth is in priority I and II regions and
thus i.u riot covered by this regulation as proposed.
C. A1 L_e r n i t_i \-_o _ IV : _ _ A r e a Classi _t i_c: Qt.i_on_ S_c \ i enie
Most oi the1 comnie.ij Is pertinent to Alternative I apply
to IV. However, it is difficult to evaluate Alternative IV,
because ve can not anticipate how states will classify legions.
-L • Hen ion61 Orowi ij_ _[1L'LX4£^.S
a. Maintaining some areas at Zone 1 levels while
allowing reasonable and planned growth in adjacent areas
appears to be feasible.
b. A significant number of exceptions to zone II increments
would be required by 1990 in high growth areas.
2 • ,Indujs^tji;\i: • 1 and D j.er_qyj- Rel at ed Sour c e s
a. Zone 1 restricts most industrial activities in or
near the zone with the exception of large (1000,000 TPY)
Kraft pu]p and paper mills and small, well controlled power
planes (150 MW meeting NSPS).
-------
55
II. C' >}}f I !•• - on;;
A. A XC.MI u KJ j) I r ij'i h !'«_'! oil , i n" o i •' H t'1' L! iCj ! r.u.M i! ; \,o ' .1 d
bo luos I oj it'ctive in both ] >j o: • i -'j'v.i ' >ii cloc.n a "J r ai.d ace v.'iiodaL—
inn plaiiiji'u oco':>Oin,ic i.fro~,.'trj. Po I ~! ov/i.ng ere tic proposed "..one
d e f i n i t .1 u • • _- :
1. A VUT->- retrictj N/O zouy to mninLain r i r ^odlity in pristine
areas.
2. An -i j/(j'-'inodj u.'.e zone, .s.iiailar to A",. Leru,. Li ve 1, to
p^rro.n' -1: r!;..° ,. - .-;v^:~; iry/vr. rri ixl c/?velo; •;;.;• r\+- for .-1 ".iL.; c:r economic
sizes.
3. A tliivd zone, a] lov/ing grov/l:h up to the r.ocond :ry
standa': '-j.-J i. o :
a. uJl'/.v areas v;ith e;ic(.)pl;ions to be planned rather than
allo\7 the(i: Lo be1 developed p.i ece-inea3 on Llie } "i-.is OL economic
pressure.';
b. avo:d the unnianagable and uiinocesr arv burden of. grantinq
e*ACtipL j-OLi.. ' ^ iu.j-j u^-.v ^.^^wu-^i... j.^-v»<.i.'i - - -j ''-"- a pi .-T.r ~ c1. ''"zi'C'.*! !"• r."1 '
B. rj'lie ,i n torinedj ate zone: should have an exception r;:ehe;rc
which \/oj I c- jiorin.it the consideration OL othc.-r socio~cc or.o;ric
tradeoffs in (a) industrial rlant siting or si:--ing (b) natural
resource eevelopraerit (e.g., open pit mining) cad (c) colloca-
tion of industries. Such exceptions should b.' poss'_b ie
v/ithoxil TI -zoning an entire ai ea for degrodatioa ap to the
socoiidaj'y standards. In laany cases the full i no/• vce I up to
seeoiidarv nnctht not ]v- rec[U.ired, or only a s;n;;il pr,rtion of the
xone m.iqnL lie affected by the source.
C. Ta.l .1 stacks arc- the least cost option for reducing ground
level concvntrati oi:s ..
1. Doubling stae/.. lieight reduces groand concentration by
about 70 jjc'i'cent.
2. I1 lore strjngeni increments v/onld be po. slblc for indiisl
soure^s :i n Zone IT ir ta] 1 stacks were allovrec;.
-------
55
3. The cost of doubling stack height (from 400 to 800 ft.) is
$4MM while the cost of scrubber may be $50MM, both for a 1000 MW power
plant.
4. In cases where scrubbing is not possible, the incremental cost
of scaling down from 1000 MW to 300 MW plants is $123 MM (per 1000 MW
of total capacity).
D. The air quality increments proposed for Zone II will generally
provide an adequate margin for growth through 1980. The proposed 24-hr.
particulate increment may require further analysis.
-------
57
III. Methodoloc
The analyses that have been used to assess the economic
growth and development impacts of the throe alternative signi-
ficant deterioration regulations cover a v/ide range cf new
development situations. The studies arc characterized by
two approaches: analysis of impact in specific prototype regions,
and analysis of impact on isolated new industrial and energy-
related sources. Each approach is diecursed below:
t
A. Prototype Regional Dcvclopmont
Prototypical regional developments were identified for
study by EPA to reflect (1) trend development in a growing
metropolitan areas, (2) resource-based development in a clean
air area and (3) development in a rural state. The regions
which were selected for study are the Greater Boston Metropolitan
Area, the Four Corners Region and the State of Iowa respectively.
Additional information is available on two additional regions:
(a) the Dallas-Fort Worth area in a study undertaken by the
Department of Commerce, and (b) New York State, in a study done
by the State Department of Environmental Conservation.
1. EPA Study
The EPA Study of Boston, Four Corners, and Iowa used
regional economic growth plans and projections to estimate
emissions. An air quality dispersion model was employed to
estimate the ambient air quality effects of projected emissions.
Both air quality and emission estimates were then compared with
proposed non-degradation increments to determine the extent of
any potential restrictions on regional economic growth.
This study, performed under contract by Harbridge
House and Environmental Research and Technology, used the
following methodology:
-------
58
a. Planned and projected economic growth to 1980 for
each area was obtained from several sources including the
Office of Business Economics, Department of Commerce, regional
and state planning councils, the Battelle economic base study
for the Four Corners Area, and updated industrial projections
from each of the major source industries. Growth projections
were derived on a county by county basis.
b. 1972 emissions inventories were obtained from EPA.
c. Emissions from new sources were estimated assuming
attain'rrnt of New Source Performance Standards (NSPS) . Published
EPA emission factors were used. Where emission factors were
unavailable, best available control technology was assumed to
lead to a 99 percent reduction in particulate emissions and a
90 percent reduction in sulfur dioxide emissions. No control
technology was assumed for open pit operations. Area source
emissions for Bocton were based on emission factors developed
for the Northeastern U.S. These factors take residential and
i_Oiiunc-j-1—Lc.i_i_ ii^ciuj.iL
-------
59
3. New York State
The New York State analysis differed from the other
two in the manner in which ambient air quality projections
were made from emission estimates. No diffusion models were
used. A linear relationship (or rollback model) was assumed
between emissions and ambient conditions within each county.
Ground concentrations from planned fossil-fuel power plants
were obtained from available projections.
B. Isolated Industrial and Energy-Related Sources*
1. EPA Analysis
Sixteen major new source categories were identified
in the proposed regulations as the major sources of S02 and
TSP. However, in reality each major source category tends to be
associated with geographically collocated industrial activity.
An analysis was therefore performed of major source "building
blocks" composed of the base industry and collocated industrial
activity. In addition, new towns and industrial parks were
analyzed as concentrated new components of urban growth. The
methodology followed by Karbridge House in developing building
blocks was as follows:
a. Principal links between the industries, their
markets, and suppliers were identified using the National Input/
Output Table (U.S. Department of Commerce, 1969) with appropriate
updating. An investigation of each industry yielded information
on location decisions and key determinants of the geographic
location of related industrial activities.
b. Locations of the building blocks, with the except-
ion of power plants, were determined nationwide by Air Quality
Control Region using industrial growth projections from trade
journals and CBE information. The emission contribution was
then calculated for each AQCR, and the emission density was
estimated both for the existing AQCR and for areas the size cf
the suggested 10,000 square mile area.
*See Table I
-------
60
c. Ambient air quality diffusion modelling was
performed for 35 facilities of differing sizes in 10 major
source categories. General:.zed meteorological assumptions were
used for three source categories: iron, and steel, zinc, and
aluminum (i.e., flat terrain and good dispersion character-
istics were assumed). Seven source categories were modelled
within the three regional analyses. Therefore we were able
to take into account both the actual meteorological conditions
i
reflecting the worst case at a specific site (e.g., low wind
and unstable atmosphere) and any interaction with projected
area sources.
2. Department of Interior Study
A study conducted for the Department of Interior by
TRW examined projected emissions from new energy facilities*
to determine if they would be allowed in an emission density
scheme. TRW also applied generalir-'od factors indicating
the relationships between emissions and air quality for each
geographic region of the country to estimate ambient air
quality impacts.
*See Table 1
-------
61
TABLE I
/
The EPA Analysis examined the 16 industries are listed below:
c Power Plants
© Coal Cleaning Plants (Thermal Dryers)
9 Kraft Pulp Mill Recovery Furnaces
9 Portland Cernen I Plants
9 Primary Zinc Smelters
9 Iron and Steel Metallurgical Furnaces
-------
62
IV. Impact Analys i_s of Alt : ornat ijye_ Plans i
This section is divided into three parts: a discussion
of the findings from analyses of three of the four proposed
alternatives. The State Determination alternative is not
susceptible to analysis. The findings are divided between
regional growth implications and industrial or energy-
related impacts.
A« Alternatj^ve^I: Air Quality Increment Plan
1 . Regi ona 1 Gr owt h Impac t s
a., The proposed annual EC>2 and TSP increments appear
to be adequate to allow planned and projected economic growth
in most urbanizing areas to 1980. Some areas would be affected,
however, and the number of ereas which experience limitations
on growth would increase where greater use of high sulfur coal
by existing plants becorres essential in place of low sulfur
fuels .
ronr-entrations of SOo reached
3 3
5 ug/m in Boston by 1980; 8 ug/m in Dallas/Fort Worth by 1930—
both well below the proposed increments.
• Any constraints on growth in the Boston or Dallas/
Fort Worth areas will be related to restrictions imposed by
secondary TSP standards and not to proposed significant deteriora-
tion levels.
• Two areas could be affected by the increments in
New York State by 1980, however, the State does not find the
impact on economic development to be significant: (a) the
levels are only slightly above the limits, and (b) resiting
of a few sources would reduce area concentrations to allowable
limits. Moreover, because the New York State analysis did not
utilize site-specific diffusion modelling, siting and stack heights
which minimize ground concentrations were not taiken into
account and the numbers reflect higher concentrations than
those achievable. Specifically:
-------
63
— The lower Hudson Valley, would appear to exceed
the proposed increments by 1980.
i. This area exceeds the proposed SC>2 increment
by approximately 3 ug/m3 because of the contribution of new
fossil-fueled power facilities brought on line by 1980. The
increase could be even greater if another projected plant were
located in the area.
ii. General population growth and economic develop-
ment (SO percent increase in the number of households) could
cause the particulate increment to be exceeded by 6 ug/m3.
— One additional area, the Greater Rochester Urban
area, would approach the allowable SO2 increment by 1980.
(Projections range from 4 to 16 ug/m3).
b. Many areas would exceed the proposed increments by 1990
and would experience limitations on future growth after 1990.
• Neither Dallas/Fort Worth nor Boston would appear to
violate the increments by 1'jyu.
® Five areas in New York State may exceed the TSP
increment by 1990. Increments of up to 20-30 ug/m3 are possible
based upon projected growth.
• Approximately nine areas in New York State violate
the S02 increment by 1990.
• Power plant siting and control technology is especially
critical if all reasonable growth is to be accommodated in high
growth areas. Restriction of accessible electric power, which
absorbs a major portion of the allowable increment, in turn
restricts industrial and comirerical development.
2. Indu s t r i a 1 and Em; r cr y - gel a to d, J3our c_eg
a. Careful siting and the use of taller than traditional
stacks v/ould allow roost new industrial energy development.
• The impact of the air quality increment on allowable
plant size is highly dependent upon stack height. As stack
height is doubled, ground concentration of pollutant is reduced
by 70 percent (see Exhibit I).
-------
64
• With traditional stack heights, the proposed incre-
ments for Alternative I would bo violated by: 1000 MW power
plants, large copper smelters, zinc smeJters, cement plants,
and iron and steel mills. (3 hour SC\_ and annual TSP standards)
-?x
« If taller stacks are used (2 times traditional heights,
see Exhibit I), copper, cement, and iron and steel building
blocks still violate the TS.? increments because of associated
mining operations. In each case the plant itself causes no
violation but would have to locat-.e some small distance from
the mine-. Incremental transportation costs have not been
assessed.
o It may be possible to reduce mine emissions below the
increment in humid areas by wotting the site. However, in
arid areas there is no technology for reducing mine emissions,
and a violation of the TSP increment would result.
• The analysis is also dependent upon dispersion model
-!C.r-;-i-lr«^x4_-J/-^v-lc, ^* f^^r-*^. .3 1 ^^r *^ ,^4- ^ s*y. ^1 ^j^, , -,v^^q 4- ^%~ ^^•^-,.^'U.,.- 7\ ^"| _, ~.l.
C-~ U. ^ -I •' il J ' i- - _. JL---J- -- A —
located under poor dispersion conditions could cause ground
concentrations two to four times greater than an identical
plant located under the favorable dispersion conditions
assumed in the modelling effort. Thus, a fixed deterioration
increment could be considerably mere constraining on plant
size in areas with poor dispersion characteristics.
b. The assumptions on fuel utilization are important in
drawing conclusions on the constraints imposed by the proposed
ambient levels. The TRW analysis indicated that with the use
of tall stacks (i.e., 800 feet for power plants):
11 - Power plants burning 3 percent sulfur coal with
wet scrubbing technology would be limited in size
to about 1000 MW in Appalachia and perhaps 1400
MW in the Rockies; 1500 MW plants could be built
in all other areas."
-------
65
"- Power plants burning 0.7 percent sulfur coal
without wet scrubbing technology would be
limited in size to about 800 MW in Appalachia
and about 1200 MW in Rockies; 1500 MW plants
could be built in all other areas."
"- Power plants burning 1.5 percent sulfur coal with
wet scrubbers and 0.3 percent coal without wet
scruVbers could be built in .1500 MW units anywhere."
"Thus, the air quality increment plan does not appear
to be overly restrictive, particularly considering that
in most areas similar facilities could be sited 25-50
miles of one another."
c. Additional costs are imposed upon a new plant if in
order to comply with the air quality increment it must (a)
build a taller stack than usual; (b) apply stack gas scrubbing
for SG2 removal (low sulfur fuel burning power plants only);
(c) increase transportation and land costs where associated
industrial activities must be sepairated geographically or it
must move further from the market; or; (d) scale down its plant.
Incremental costs imposed by the air quality increment appear
to be relatively small, with the exception of power plants
forced to add a scrubber while burning fuel which meets NSPS
or power plants limited substantially in size.
• The cost of a stack approximately quadruples as its
height doubles. A recent EPA study shows that for a large
power plant the cost of a 400 foot stack (concrete 40 foot
diameter) is about $1.25 MM while an 800 foot stack costs $5
MM--thus an incremental cost of almost $4 MM.
® The cost of adding a stack gas scrubber for SO2
control to a low sulfur coal burning power plant is about
$50 MM. For plants which could meet NSPS without such a
scrubber, this would be an incremental cost of meeting tho
nondegradation increment.
-------
66
« A power plant using coal from an open mine might
also Lave to locate 10-20 miles froru the mine to avoid
violating the T3P increment. No estimate has yet been developed
for the additional transportation cost.
e An alternative to building a stack twice as high would
be building a plant only 3C percent as large. For example, the
construction cost of a 300 KW power plant would be approximately
30 percent more per unit of capacity than for a 1000 MW plant.
The investment cost for a 1000 KW power plant is about $410/kw.
t
The cost of a 300 MW plant would then be $533/kw or au incre-
mental cost of $123/kw. Thus the cost penalty is $123 million
for building 1000 KW of total capacity in units of 300 MW rather
than one 1000 MW plant. In view of the potential magnitude
of such additional cost it is unlikely that a plant would scale
down as long as it has the viable option of building a taller
stack.
• A further cost of disaggregating a single large power
plant into two smaller facilities is the additional land
required for power transmission rights-of-way and other
facilities that would have to be duplicated,, It is estimated
that 15 percent additional land would be required as a result
of a second site. A single 3000 MW coal fired power plant
requires 12,000 acreas. Thus an additional 1000 acres would
be required.
B• Alterna tiye jTIj Emissions Increment Plan
1. Regional Grov/th Impacts
a. The impact on regional growth varies widely with
the size of the averaging area and the existing emission base-
line. Regional inequities arise with averaging areas signifi-
cantly smaller or larger than the propose;d 10,000 square miles.
» Projected economic growth in the Greater Boston
Metropolitan Area falls within a Priority I area and would not
be subject to this regulation. However, projected emission
density growth for both GO2 and particulates exceeds the
-------
67
allowable increments by a factor of five. The entire State
of Massachusetts could not support new power plant growth
projected for the Boston area alone.
a No growbh restrictions are found for the Four Corners
area c?ven if the large AQCR is disaggregated into 10,000 square
mile averaging areas. Large sources are dispersed evenly
throughout the entire region.
9 Dallas/Fort Worth vio3at.es the TSP emission limitation
in six years under the 120 percent rule. Extensive controls
applied to area sources as well as complete retrofit of all
industries v/ith BACT would allow projected growth to be
accomcdated for up to eleven or twelve years.
b. Power plants absorb a major portion of the allowable
increments in growth areas, however the restriction of accessible
electric power in turn restricts industrial and commercial
9 The Flan would have critical consequences for develop-
ment in New York State. The plan would restrict development after
1980 in much of the State (excluding AQCR's or counties with
significant portions at or above Federal secondary standards
with implementation plans in effect) . The major impact would
be a severe restriction in additional fossil-fuel power plants
which in turn could curtail growth in industrial and ccmmercial
activity.
e Planned mine mouth power plant development in the
Northern Great Plains coal formation could not be accommodated
unless the averaging area were expanded to encompass six
adjacent states.
c. There are .no identifiable impacts on patterns of
development within a region, as siting is not a consideration
under this plan.
-------
68
2 . In du s b. r ia 1 Gr owth and Ene rgy- Re 1 at e d S our c es
a. An emission limitation plan impacts most heavily
upon power plants and related fuel development.
o Power plants account for 84 percent of TSP
emissions and 98 percent of BQ2 emissions from all new
industrial sources.
o Coal fired power plants in current clean air areas
could be restricted in size to as little as 300 KW or as much
as 1900.MW for every 10,000 square mile area depending upon
the baseline.
b. In the absence of baseline aberrations, industrial
growth projected to 1980 would not be affected by the increments,
with the exception of any indirect impacts due to restrictions
in power plant development.
« Hew industrial development to I960, in its projected
locations, v/ill require only a fraction of the allowable incre-'
rnerrL.
e Most new industrial sources are projected for
location in the rural and semi-rural areas of Priority I and
II AQCR's and would therefore not be subject to Alternative II
as proposed.
c• Alternative IV: Area Classification Plan
1. Regional Growth Impacts
a. Maintaining some areas of Zone 1 levels while
allowing reasonable and planned growth in adjacent areas appears
to be feasible.
9 The State of New York has indicated that in 29 areas
of study they were able to devise a crude area classification
scheme which would maximize allowable growth to 1980 in the
state. This plan gave 25 areas Zone 1 status for TSP,
13 areas Zone 1 status for SO2 and exemption status to only
two areas. Exemption status would be required for many more
areas (nine) by 1990. This zoning is somewhat misleading
in that all areas which currently exceed secondary standards
were"classified as Zone 1.
-------
« The Grand Canyon could be designated Zone 1
without .restricting resource development in the Four Corners
Region.
b. A significant number of exceptions would be required
by 1990 in high growth areas. (See Alternative I discussion.)
2• Industrial and Energy-Related Sources
Zone 1 restricts most industrial activities in the
area with the exception of large (100,000 ton per year) Kraft
Pulp and Paper Mills and small, well controlled power plants
(150 MW meeting NSPS).
-------
70
EH
H
to
H
M
CO
O
CO
I
oo
s;
CO"
w i ^.
H!> 3
?H| &
0| O
E tf
o o
U| U
W H
S^
fc tl
CO
g
H
a
&
u
£3
a
CO
EH
CO
to
n
co
ft
/
• co
in
H
CN
H
O
CN
in
vo
(N
4J
d
ra
H
ft
r4
rl (U
o s
m o
ft
tf
QJ g
M a
•H
3 O
tr o
CU O
« rH
O
O
H
•tf
•
H
CN
00
CNO
oo
O
CO
00
in
•
^
kD
oo
\ .
CNf>
kD O
00
rl
0)
4->
rH CU
QJ d
E -H
to 2
M M
OJ OJ
ft ft
ft ft
0 0
u u
><
ft
EH
O
O
O
O
r-
r-l
I
m
•
00
\H
CNp^
in n
• • rH
o
in
H
\-.
CNf^.
CN 00
rH
4-1
d
(0
H
ft
-P >i
d r4
GJ 'M
E (0
v QJ 3
O 5>
>H
ft
EH
0
O
0
O
LO
CN
I
00
•
o
00
•
o
H
CO
•
CN
4J
d
rd
H
ft
ft
H
3
ft
4J
4H
(0
M
2
>H
ft
EH
O
O
O
O
O
rH
m co , mo
in oo 1 CN o
ro Q)
(-H
r1-! •
~ -P 13
4-) 0
C O co £
^OCNCN CNin r04J-HS-l
• • H HO
O O ft rH CO 4-1
fO -H 5-1
-P 3 i QJ
« D1 H ft
QJ
H rj;
fO co • ,fi
M -H d
«> CJ o -P
CN > >t -H CO
r>>, CO-
0 D' C -H
. , _ QJ co
r^ Ti fj >,
u a c H
rO ra 0 rj
-P U d
W -P d
o o in o d ^c!
oocnl CNO 4-iradrH
rH CN 00 CO rH rj QJ
Q) ft 0 'O
H SH O
•y H E S»g
d ra o
H 10 4J •+-« M (J U
QJ rH d -— 44 jj O
CN o in CN in 4J -p ft its -ri
. r-l4JrHdH4->CO(l)tO
o dcuioftft-PCdu^
fO E H H d f8 0 ^| 4-1
H CO ft 3 E fO H -rl 3 4_|
ft ft 3 rH ft CO O -••!
, J-l 4-> d ft (0 U) Tf
^ Q) C 4-> -H rH -H
0 QJftaj4HECJOJ£r;r<
^ ^ftErs^dcJQjo
5 O QJ rl rH -H ^J fQ
m f-s. a)uorv;fONW4H QJ .
f^ o o in o o c;
HrHcNCN rH5-lM!-l!Hr4!HM 4-1 C
OOOOOOOd 0-rH
4H 4-1 4H 4H 4-1 4-1 4--I O U
•H d ra
4J 4J jj 4J 4J 4J 4-> 4-' O ^
4J (U(Ua>QJ(U(UQJS-i-,H -H
d Ct)0a)QJQ)QJCUO4JX
(t) J-i 4n 4H 4H MH MH 4H 4~l JJU 1 O
rH QJ 4-1 O ft J-|
ft 4-1 d O O O O O O O J-i -H p.
rH rt! OOOinOOOftua
§QJH •• CNOOHCNCNCNCN 4Jrd
g ft 'C QJ d
d co >i QJ cQComcocococorc;oaJ
-H H SH co 3 3 -3 3 r5 3 3 4J O rO
§U QJJ-) O 10 CD 10 W W) CO CO -I
d CUfO ti rlMrlr)rlrHrl4JHrl
H -H 4J? o ajajajQJOJQjajrarau
< N coco; M >>>>>>>,dd
ft -P O ra
>H >H >H 4-1 4-> 4J 4J 4-> -P 4J -rl
P4ftft > OJCUOJCUQJQJQJCO .|j ±>
EH H EH H H 0) Q) O O st <£> CN m "si1 si1 ^ fi; (i, |TJ
O O O OJ IS! IS)
CN 00 CO SH \ \
H H oo Pi H| • CN|
-------
Reference 12
SCIENTIFIC FACTORS BEARING ON REGULATORY
POLICIES TO ASSURE NON-DEGRADATION
OF AIR QUALITY
I. Overview
Most scenarios under which regulatory policies to assure non-
degradation are conceived imply the existence of information and/or
techniques which do not always exist. Specifically, assumptions are
made as to abilities to measure, within limits of accuracy and precision
which would be the result of the regulatory policy, the concentrations
of the pollutants under consideration. In addition, it is implied
usually that, once measured, these concentrations can be compared with
"baseline" concentrations which are assumed to be known. Finally, it
is often assumed that mathematical simulation techniques can be used to
predict the air quality impact of land use policies sufficiently well
to help define regulatory policy. It should be obvious that the know-
ledge and techniques mentioned above cannot be perfect and thus that
the limits which exist on our ability to measure, model the transport,
and determine "baseline" levels of the criteria air pollutants must,
in turn, constrain regulatory policy options to some extent. It is the
purpose of this paper to set forth the limitations which exist on our
abilities in these areas so as to document the necessity for discarding
many regulatory options which might otherwise seem obvious.
-------
72
II. Measurement Capabilities
Total Suspended Participate (TSP): The federal reference method
for this pollutant is both sensitive (to 1 yg/m3) and has reasonably
low relative variation *j as shown:
Ambient Level Measured (yg/m3) Relative Variation (yg/m3)
1 2
150 4.5
260 7.8
The technique appears adequate for most regulatory options related to
non-degradation.
Sulfur Dioxide (S02): The federal reference method (prosaniline)
for this pollutant is of limited sensitivity and rather poor relative
variation. The lower limit of sensitivity of the method is on the close
order of 30 yg/m3 but has not been rigorously defined. The relative
variations tabulated below are quite adequate for many regulatory
purposes, but appear deficient for application to non-degradation of
areas relatively low (below secondary standard) in S02-
Ambient Level Measured (yg/m3) Relative Variation (yg/m3)
45 40
350 65
1300 120
Carbon Monoxide(CO): The federal reference method (non-dispersive
IR) for this pollutant is of fair sensitivity (300 yg/m3) but rather
poor relative variation as shown:
V Two measurements differing by an amount less than the relative
variation cannot be interpreted as being significantly different to
a 95% level of confidence.
-------
73
Ambient Level Measured (ug/m3) Relative Variation dig/in3)
300 300
10,000 2500
40,000 2500
Since ambient concentration in clean environments may range as low as
150 Mg/m3, it seems clear that the technique would not be applicable
to many non-degradation situations.
Photochemical Oxidants as Ozone: The federal reference method for this
pollutant has sensitivity to approximately 50 yg/m3. The relative
variation of the method is reasonably good, as tabulated:
Ambie_nt_ Level Measured (yg/m3) Relative Variation (ug/m3)
50 10
160 12
At the present time, not enough is known about prevailing ambient levels
to determine whether these performance characteristics are adequate,
but there is some indication that levels below trie sensitivity of the
method are not uncommon.
Hydrocarbons: Due to the fact that the present federal reference
method for total hydrocarbons does not distinguish between hydrocarbons
from natural sources (trees, etc.), the levels of which may range widely,
and man-generated hydrocarbons, it would be impossible to substantiate
environmental degradation in the sense of this pollutant using the
federal reference method.
Nitrogen Dioxide (N02): A federal reference method for this
pollutant has not been promulgated at this time. Although there should
be at least one adequate method among the existing interim methods, it
-------
74
seems essential to first promulgate a federal reference method for use
in this area before attempting to develop regulations.
Addendum: While the above performance discussions do define
constraints placed on regulatory policy decisions by the capacity of the
analytic methods, there is a further overlay of parameters which impact
on the performance of field samplers using these methods. In order to
analyse air quality it is necessary to pass a known amount of ambient
air through the measurement device in a known period of time. Ideally,
the air drawn through the device is all of consistent quality and is
also representative of the area under consideration. This latter
factor may, however, be a function of terrain, climatic conditions and
proximity to sources of interference and thus all of these will bear
on the number of measurements required to determine air quality to a
desired degree of confidence. This additional influence on performance
further mitigates against the use of air quality measurements in any
active regulatory role vis-a-vis non-degradation.
III. Baseline Air Quality Information
TSP: The available data base on non-urban TSP levels is more
extensive than that for any of the other criteria pollutants. Measure-
ments have been taken over the period 1964-1972 at some forty non-urban
National Air Sampling Network (NASN) stations.
The overall average TSP concentrations at all NASN non-urban
stations for the period 1964-1972 was 35 yg/m3 but the individual 9 year
station averages ranged from 10 yg/m3 (Yellowstone., Wyoming) to 68ijg/m3
-------
75
(Cape Hatteras, N.C.). Individual yearly averages, of course, show
even more variation and ranged from a low of 8 ;ig/m! (Yellowstone, 1967
and 1968) to a high of 110 ng/m3 (Cape Hatteras, 1972.) Individual
24-hour values show a very wide variation with a low of lyg/m3 in a
number of stations to a high of 312yg/m3 (White Pine County, Nevada,
1965).
It is obvious from this information that there is extreme variation
in TSP concentration between individual station annual averages and
between the lowest and highest individual 24-hour value. Therefore,
any decisions with respect to significant deterioration of air quality
using these data must be made with full cognizance of the variability
discussed above. Further, there is no significant difference between
TSP data collected at non-urban NASN sites in 1972 and data compiled
from either 1969-1972 or from 1964-1972; hence, 1972 data would be as
good as starting point as any other single year from which to determine
significant deterioration provided it is desirous to establish a single
nationally applied baseline year. However, a nationally applied
single-year baseline does not appear to be feasible because of the
considerable year to year variability in data obtained from specific
non-urban sites. Data from several years at a given sight are evidently
necessary to establish a baseline level with any degree of confidence.
SO^: Although attempts to measure SC^ at twenty-eight non-urban
stations have been conducted over the past six years, the results of
this effort force the conclusion that it will be essentially impossible
to establish any sort of baseline, nationally or area-wide, using the
standard reference method. This is largely due to the insensitivity of
the technique.
-------
76
Other Criteria Pollutants: Although scattered data exist on
concentrations of the other criteria pollutants in rural and wilderness
areas, these data are in no way sufficient, either geographically or
in terms of span of time covered, to suggest baseline concentrations for
use in non-degradation enforcement activities.
Assessment: Extensive experience with measurement of suspended
particulates (TSP) shows the danger of attempting to arrive at a
determination of baseline levels from insufficient information. Further,
the wide variability over time of concentrations measured at any point
make it clear that extensive evidence would be required to document.
degradation of air quality from any established baseline. Although the
standard reference method for TSP appears adequate to such a task, should
it be determined necessary, this i; not the case with most other criteria
pollutants. In order to develop (as necessary), collaboratively test,
and then promulgate appropriate analytic methods; and then apply them to
determination of baseline air quality, nationally, would require the
expenditure of resources on the order of hundreds of millions of dollars
over a period of five to ten years. Finally, the use of this information
for comparative purposes to show the degradation of air quality at a
given site or set of sites would require extensive further measurement
over periods of at least a year and perhaps more.
IV. Modelling Capabilities
In discussing this subject it is first essential to point out the
necessity for excluding all of the photochemical pollutants (hydrocarbons,
nitrogen oxides and oxidants) from any but very theoretical consideration.
-------
At. this time meteorological dispersion/simulation modelling of these
pollutants is in the research stage only and validated models are not
available. Previously proposed State Implementation Plans were based
on an empirical relationship between hydrocarbons and ozone. Further
research and development in the area of simulation of these pollutants
is confined to atttempts to treat such heavily polluted areas as the
Los Angeles Basin and would need to be drastically redirected to address
relatively clean air regions (to account for hydrocarbons from
vegetation, for example). In terms of the conservative pollutants
(sulfur dioxide, carbon monoxide, and particulates) the use of mathemat-
ical modelling is often considered in two contexts: (1) use to
calculate a_ctu_a]_ concentrations of pollutants under a given set of
circumstances and (2) use to calculate the changes in concentrations of
pollutants implied by changed circumstances (such as additional sources).
The former use is the less developed of the two. For example, there is
virtually no information or applicable past experience from which one
could develop models to predict baseline air quality. However, the use
of semi-empirical "Gaussian plume" models to relate pollutant concentration
increases to additional sources of emission is reasonably well developed.
These models have been assembled and validated for point, multiple point,
and certain types of area sources. It should be pointed out, however,
that the accuracy which can be expected of the models decreases as
distance from the source increases and that many non-degradation applications
require extending the application of the models beyond distance limits
(perhaps ten miles) for which they have previously been validated.
Further, the uniformity of meteorology and accuracy with which meteoro-
logical parameters are known over the region being simulated affects
-------
78
the accuracy which can be expected. In general, we would expect to be
able to predict the hourly average air quality increment due to
additional point sources to within a factor of two (i.e., a predicted
increment of 10yg/m3, would have a "range of truth" of from 5 to 20yg/m3)
at least half tne time. Predicted annual average increments will range
within a factor of 1.5 of the observed average increment at least half
the time. Thus even in the best case the possible errors in the modelling
process due to for example, uncertainties in meteorological parameters
may overwhelm the air quality increment under consideration.
Mathematical simulation does, however, provide a consistant,
reproducible way to estimate "hypothetical" air quality increments for
conservative pollutants. The development of enforcement scenarios
utilizing mathematical models and controlling against such hypothetical
(but uniformly reproducible) air quality increments, appear to have
some merit.
V. Conclusions
Measurement: Although the federal reference method for suspended
particulate is evidently adequate, the federal reference methods for
other criteria pollutants at low (clean environment) concentrations
suffer varying degrees of inadequacy. Extensive modification of existing
methods or the development of new methods would be required in order
to develop a capacity for determining significant deterioration using a
"ground truth" approach.
Extant Air Quality Data: Only data on suspended particulate
concentrations are extensive enough to be meaningful. The major
conclusion which can be drawn is that vast numbers of measurements are
-------
79
required to first determine a baseline level and then further extensive
measurements are needed to establish degradation from that level.
Model 1ing/Simulation: The use of simulation techniques to predict
concentrations of the non-conservative pollutants (oxidants, hydrocarbons,
and nitrogen oxides) is scientifically indefensible at the present time.
Simulation techniques may have some utility in non-degradation enforce-
ment in that they allow the establishment of a logically consistent
enforcement posture predicated on limiting the incremental increase in
"hypothetical" concentrations of conservative pollutants.
-------
Reference 1 3
AVAILABILITY OF AIR QUALITY DATA IN AREAS
GENERALLY BELOW THE NAAQS
In order to estimate the number of monitoring stations which
are located in areas which are generally below the NAAQS, the 1973
Monitoring and Air Quality Trends 'Report (EPA -45C/1-74-007) was
examined to determine the number of urban versus rural sites. Although
it is probably improper to assume that urban areas are above standards
while all rural areas are below standards, the following table does
illustrate how little monitoring is being done In the rural, generally
clean, areas.
Pollutant
TSP
SOo
CO
Oxidant(03)
NO.,
Total number
of sites
3591
2132
289
252
Number of sites
reporting at least
one valid quarter
of data
3390
1615
213
188
588
Number of valid
ruraj__ s i tes
136
55
0
0
40
80
-------
Reference 14
TECHNICAL DATA IN SUPPORT
OF
SIGNIFICANT DETERIORATION ISSUE
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Water Programs
Office of Air Quality Planning and Standards
Research Triangle Park, N.C. 27711
-------
TECHNICAL SUPPORTING DATA TO ACCOMPANY
PROPOSED RUtE MAKING ON PREVENTION
OF SIGNIFICANT AIR QUALITY DETERIORATION
FEDERAL REGISTER DATED JULY 16, 1973
VOLUME 38, NO. 135, PART IV, PAGE 18985
All data presented herein are estimated, based upon typical con-
ditions. Many of the specific values represent averages from which
tne range of values exceeds several orders of magnitude. Although the
data are adequate as a guide for impact assessments of alternative plans,
tney are not intended for use in assessing the impact, on individual
sources, facilities, or AQCRs.
TABLE 1. SOURCE CATEGORIES
All of the options set forth in the proposed rule making for pre-
venting s iqrr, • u.artt air quality deterioration require review of certa>n
types of sSdtiorMry sources. Table 1 identities the sixteen sources
subject to review along with the related information on annual construction
estimates and -,-ource size statistics. This list represents the Administrator'
best judgmert ,?^ to which sources individually have the potential for causing
"significant deterioration." The sixteen source categories account fcr
approximately 30 percent of the particulate matter and 75 percent of the
sulfur dioxide emitted into the atmosphere eacn year nationwide, and a:count
for essentially all of these pollutants emitted in clean areas.
TABLE 2. COMPUTATION OF EMISSIONS FROM SELEcTZD CATEGORIES
Table 2 shows emissions estimates for the sixteen source categon >s
identified in F, ble 1. Estimates assume the application of "best available
control technoh gy" .jnd are for an average size facility.
-------
TABLE 3. CALCULATED MAXIMUM POLLUTANT CONCENTRATIONS FOR SELECTED
POWER PLANTS USING OBSERVED STABILITY WIND ROSES - GOOD
DISPERSION AND TERRAIN ASSUMED
Table 3 shows the calculated total suspended particulate (TSP)
and sulfur dioxide ($02) concentrations associated with fourteen different
power plants. The list represents a variety of coal burning power plants
with different capacities, volumes, stack gas temperatures and stack
heights.
TABLE 4. CALCULATED IMPACT OF SELECTED POWER PLANTS ON SHORT-TERM
AIR QUALITY
This table shows the estimated maximum short-term pollutant concen-
trations in the vicinity of five large power plants with various stack
configurations. The modified values represent those concentrations
corrected to reflect application of best available control technology.
TABLE 5. CALCULATED IMPACT OF SELECTED AREA SOURCES ON SHORT-TERM
AIR QUALITY
Emissions of S02 and TSP from three area sources plus the predicted
impact of these sources on air quality are listed in Table 5. All sources
are assumed to burn .3% sulfur oil.
TABLE 6. CALCULATED POLLUTANT CONCENTRATIONS RESULTING FROM POINT
SOURCES IN HYPOTHETICAL AQCRs
Table 6 illustrates the impact of five point sources on air quality
in hypothetical AQCRs. Resulting concentrations assume application of
"best available control technology" and are given for both open and
valley terrain.
TABLE 7. CURRENT AMBIENT AIR QUALITY STANDARDS
Table 7 identifies current primary and secondary ambient air quality
standards for TSP and
-------
TABU 8. ESTIMATED COSTS OF EMISSION CONTROLS FOR SELECTED SOURCE
CATEGORIES
Table 8 provides an estimate of control costs for selected source
categories. Cost figures assume application of best available control
technology.
FIGURE 1. DISTRIBUTION OF PRIORITY III AQCRs BY CAPACITY TO SUPPORT
INCREASES IN TOTAL S02 EMISSIONS
Figure 1 is a cumulative distribution of AQCRs based on their
capacity to support additional S02 emissions. The distribution assumes
a maximum allowable increase in emissions of 20% over those levels that
existed in a sample baseline year. Interpretation of the figure is best
illustrated through an example: approximately 40 percent of all
Priority III AQC^s cannot support a total increase in S02 emissions of
30,000 tons per year. Alternatively, 60 percent of the Priority Hi
«
AQCRs can support a single source, or combination of sources, resulting
in 30,000 tons of annual S0£ emissions.
FIGURE 2. DISTRIBUTION OF PRIORITY III AQCRs BY CAPACITY TO SUPPORT
INCREASES IN TSP EMISSIONS.
This is a distribution similar to Figure 1 but based on TSP emi MODS.
For example, approximately 83 percent of all Priority III AQCRs cannot
support a total annual TSP emission increase of 40,000 tons. Figure 2 also
assumes a 20 percent allowable increase in TSP emissions over the baseline
values.
NOTF: Unless specified otherwise, data for power plants are based 0' the
alternate definition of best available control technology.
-------
85
TABLE 1
SOURCE CATEGORIES REVIEWED
SOURCE CATEGORY
Fossil Fuel Fired Steam Electric
Power Plants
Coal Cleaning Plants
Kraft Pulp Mills (Recovery Furnaces)
Portland Cement Plants
Primary Zinc Smelters
Iron and Steel Mill
Metallurgical Furnaces
Primary Aluminum Ore Reduction
Primary Copper Smelters
Municipal Incinerators
Sulfuric Acid Plants
Petroleum Refineries (Cat. Crackers)
Lime Plants
Phosphate Rock Processing Plants
(Process Unit)
By Product Coke Oven Batteries
Sulfur Recovery Plants
Carbon Black Plants
EXPECTED
NO. PER YEAR
40
AFFECTED
SIZE
SIZE DISTRIBUTION
AVG. RANGE
100 MW 600 N!W
50-1200 MW
9
15
6
2
ALL
ALL
ALL
ALL
465 TPH 100-1000 TP^
800 TPD 500-1500 TP;
100 TPH 75-300 TPH
280 TPD 200-900 TPD
(Metal)
8 Elec Arc
6 BOP
1
2
25
2
3
7
4
3.5
8
1.4
ALL
ALL
ALL
*
250 TPD
ALL
ALL 10x1
ALL
ALL
ALL
ALL
ALL
200T/heat
480 TPD
260 TPD
(Metal)
300 TPD
750 TPD
04bbl/day
450 TPD
1200 TPD
180 TPH
140 TPD
120 TPD
100-325 T/hea1
95-740 TPD
125-1000 TPD
50-1000 TPD
500-1500 TPD
40-200 x 103
bbl/day
240-600 TPD
550-1650 TPD
70-360 TPH
27-270 TPD
109-270 TPD
-------
86
ts>
LU
o:
o
UJ
I—
~
_J O
-
LU OO ^
1 — H-
C9 OO
z a:
(— ' IE
h~~ *— *
c£
LU s:
Q- 1—1
O H-
CM
O
CO
o
1- -
CJ
i—
CO
VO
O
««^^
^•^
jQ
J-™1
CM
_
:z>
I —
VO
o
C^
in
r—
0
„ .
:c
h-
CO
vo
3 O
21 r—
O X
o
VO O
o
o
VO
-fc— »
E en
0) C
4-> ro
CO r—
D_
"O
cu s-
i- cu
•I— S
U- 0
o_
1 —
QJ U
3 •!-
U_ 1_
•M
r— O
•r- CO
in r—
in LU
0
u.
1
1
LO
1 — ^
f—
o
0
vo
LO
1
1—
"i/r
f~)
r-~
OO
r—
d
C£
•^s^
1—
LO
vo
t/>
4_>
c
nj
r"
0.
en
c
•r—
C
rO
cu
r. *
ro
O
CJ
1
1
O
o
Ol
O
o
00
1
1
H-
"en"
o
i—
LO
LO
r—
a:
:r
%^^i
^.
oo
oo
<^~s
in
cu
CJ
rO
in c
r— 3
•r- U_
?y
^
Q. J-
i— CU
3 >
0- O
o
+-> cu
i+- a:
c
^
1
1
LO
CM
! .
O
VO
CO
1
1
H-
\
in
J3
oo
O
C£
31
h-
O
O
in
*i_>
c
ro
oZ
-M
C
CU
E
CO
CJ
•a
c
rO
r*—
•*->
0
o.
0
o
o
oo
CD
CD
i —
O
VO
CO
1—
in
JO
r—
r^
^
oo
1 —
in
r*i
0
r~
^^
(O
0
**^
1—
O
VO
LO
I/)
i.
cu
-M
^— •
CU
E
CO
0
c
•r*
M
^
L.
rO
.S
^
a.
i
i
LO
CM
0
vo
r^
00
I
1
1—
^*
M
.a
,__
oo
o
pv
~T~
^^^
t—
0
o
r— -
*
*
*
in
CU
O
10
c
r— i.
r— 3
•r- U.
s:
r—
i— fQ
CU CJ
cu •»-
4J O>
CO i-
3
C 1^
ra ro
•P
§£
S-
1— 1
i
CD
LO
LO
O
VO
J**»^
00
t
1 —
{/}
f°l
LO
VO
^^
ra
o
""•*^
t—
o
00
cu
i-
o
E
3
c
•r-o
E
3 C.
r— O
4-}
>>o
&. 3
E W
•r- or
i.
D_
O
o
o
0
LO
1 — -
o
vo
t-N.
CO
H-
•»^
in
c\
r—
CM
O
CM
t—
in
-C)
f—
CO
VO
d
;>^
ro *~~**
O CU
->^ J_
1— 0
0 •
«* 0
O C
r— O
CJ
in
i_
CU
-U
r—
cu
E
CO
J^
cu
0-
Q.
O
CJ
>j
C—
ra
te
f-
CL.
i
LO
p^
O
o
LO
^
i
1
1—
in
f~*
r—
LO
VO
r—
>j
ro
Q
""^Sv
t—
O
O
oo
IV)
i-
0
4_>
(13
SH
(U
C
»^-
CJ
c
(—4
1—
ro
a.
•r—
0
c
3
1-L
O
O
1
1
o
o
•r—
SB. ^"~N
cu in
c i.
•r- CU
4- J*
ai u
a; ro
s-
EO
3
cu •
i — -4->
O ra
J_ CJ
co
O-
-------
37
ce
oo
C3
un
2: oo
z i— i -— -
< 5: a:
UJ O_ >-
co ~-^
o
LT>
O
un
.
•-> IE
Q- >-"
O H-
o
10
i-»
co
o
to
CO
o o
O W3
«* r^
CO CO
Q
UJ
HI
z
1— «
1 —
o
0
CM
UJ
CQ
~^.
00 (—
o
O LO
UJ
Q-
>-
t—
t—
eC to
| i \
ex. c
ro
O.
O)
1
1
I—
-•^
•!->
ro
CQ
X3
CM
rO
Q
t—
O
to
c
rO
Q_
S_
CU
o
U
CU
0£-
3
•^
OO
1
1
H-
^
to
CM
fO
Q
1—
O
CM
to
C
(O
r—
Q_
«3
CQ
O
XI
i-
il
-t-J
00
ro
U
CU
^
CU
S-
o
c:
o
-o
c:
ro
CU
CU
+->
E
X
0
a.
Q.
rO
OJ
S-
ro
CU
CU
I —
•X
•*
CU
_a
ro
-i->
OO
_c
-i-j
•r-
ro
CU
O
-C
5
c
+3
ro
'cu
s-
r—
ra
o
•r-
!=
ro
-o
CU
O)
J^
| —
*
*
rnaces .
-*
l*-
d
0)
Cn
X
o
l/l
-Q
T3
C
-------
88
3:
VOID
(—to
i/i
E
--»
en
3:00
10
rOl/)
oro«j-
...
Or^in
iocvjr^
r— CM
oo
f^.
ro
.
o
oo
O LiJ
a- 1—
O (±1
i z:
E
cr>
28
LU
_ I
o a:
t-Q
i— . oo
><
cf Q
LU LU
i—oo
«X CQ
_JO
fT3
+j
00 (
O
OJ OJ
$
O
.1
-a
(O
VI
+J
C
f
3
5 I
I 5 5
5 I.
s
01
c
en
c
-------
89
TABLE 4
CALCULATED IMPACT OF SELECTED POWER PLANTS ON SHORT-TERM AIR QUALITY
FUEL MEASURED MODIFIED* DISTANCE TO MAX.
SOURCE DESCRIPTION PRESENT ASSUMED 1-HR. S02 3-HR. S02 CONCENTRATION (KM)
1000 MW Plant 1.5%S 0.7%S 1048 yg/m3 273 yg/m3 5-6
Flat Terrain Coal Coal
2-450 Ft. Stacks
1500 MW Plant 3%S 0.7%S 2358 wg/ra3 319 yg/m3 4
Flat Terrain Coal Coal
10-250 Ft. Stacks
120 MW Plant 1.3XS 0.7%S 733 yg/m3 275 yg/m3 3
2-265 Ft. Stacks Coal Coal
455 MW Plant 2.2%S 0.7%5. 825 yg/m3 150.8 yg/m3 3
Rolling Terrain Coal Coal
3-400 Ft. Stacks
1700 MW Plant 3.8%S 0.7%S 2600 yg/m3 304 yg/m3 18
Flat Terrain Coal Coal
2-600 Ft. Stack
1-800 Ft. Stack
*Modified impact based on measured data X (i '7) to adjust emissions
to reflect NSPS.
-------
90
TABLE 5
CALCULATED IMPACT OF SELECTED AREA SOURCES ON SHORT-TERM AIR QUALITY
SOURCE DESCRIPTION FUEL
Polo Grounds - NYC 0.3%S
1,614 Units Residual
Area = 0.9 Sq. Km. Oil
Parkchester Manor - NYC 0.3%S
12,271 Units Residual
Area = 0.8 Sq. Km Oil
Co-op City - NYC 0.3%S
15,375 Units Residual
Area = 1.1 Sq. Km Oil
EMISSION
(TONS/YR.)
S02 TSP
26 14.1
151 73
376 184
3-HR S02
21.5
35.4
70.18
1-HR. TSP
(pg/m3)
18
30
59
-------
91
ro
CO
LU
I
QQ
et
r~"
2:
o
cc.
LJ_
^3 t,
2: £•£
t— t i- •»
<_?
^ ^
~> •
LJ_ ^-
I— m
z:
UJ 2;
O •— i
z:
O 00
O LU
o
I— o:
z: — >
< 0
h- 00
ZD
—1 1—
_J Z
O l-H
Q- 0
Q.
Q
UU _l
h- —, CO
-
UJ
_J
_l
ct • — •
> °E
EZ
cn
-a
** X
CM
O
oo
CO
E
CD
3.
Q.
CO
1—
•ZL
t—i
o
cu
^.
o
(_)
cu
Q. -!->
>> 0
1— 3
-o
CU O
U i-
t- Q-
3 1
O ••-
CO OQ
•"^
o
LO
•
O O O O
i— *3" CM
1 i
UD CO
CM i—
.v
^
• * .
CM LO
CM ^t
i— CO
r---
i i i
o
o
LO oo vo o
CO CM
1 1
CO O
CM i—
*
1 * 1
O LO
i — t~^
S-
s-^ O
3 •>->
>> s: ro
C s-
cu o cu
C 0 C
•i— r^ *r-
>4- — o
4-> CU C
C CC 4J l— i
ro c
•— E ro i—
Q- 3 r— ro
ai o_ a.
4-> r— T-
c o s- o
O) S- CU •!-
E -M 2 c
CU CU O 3
o a. o_ 5:
O
O
>, O)
•O N
3 T-
CO
-o
ro
•o
c
ro
+J
co
o
CO
co
ro
c
o
o
S-
c:
CU
ro
c:
ro
Q.
T3
E
ro
CO
C£L
t_)
O"
*x.
*
cu
Ol
s-
ro
•o
C
ro
o
ro
i-
o
CO
o
cu
3
"O
CO
E
O
ro
^
c
cu
o
c
o
o
cu
03
a.
o
u
o
03
-a
cu
CO
O)
CO
C
o
,
CO CT
CU O
S- i—
0. O
O) C
S- .C
o
CU OJ
I— I—
-Q
fa i —
4-> O
S-
co 4->
•r- C
-c o
+J O
c cu
vi ro[
cu i— i
IS ro
CD >
CL> CO
-C OJ
t— 03
-------
3-Hour
24-Hour
Annual
92
TABLE 7
CURRENT AMBIENT AIR QUALITY STANDARDS
TSP
(ug/m3)
—
150**
75*
SO?
(yg/m3)
1300**
365*
80*
* Primary Standards
** Secondary Standards
-------
93
TABLE 8
ESTIMATED COSTS* OF EMISSION CONTROLS
FOR SELECTED SOURCE CATEGORIES
SOURCE
Fossil Fuel Fired Steam Electric
Power Plants **
Coal Cleaning Plants
Kraft Pulp Mills (Recovery Furnaces
Portland Cement Plants
Primary Zinc Smelters
Iron and Steel Mill
Metallurgical Furnaces
Primary Aluminum Ore Reduction
Primary Copper Smelters
Municipal Incinerators
Sulfuric Acid Plants
Petroleum Refineries (Cat. Crackers)
Lime Plants
Phosphate Rock Processing Plants
(Process Unit)
By Product Coke Cven Batteries
Sulfur Recovery Plants
Carbon Black Plants
COST $(000)
CAPITAL ANNUAL
25,000
7,150
304
1,467
1,030
14,502
7,400
10,000
39,500
300
551
700
300
800
8,000
800
270
137
450
355
3,990
2,751
1,600
1 2 ,400
106
310
150
70.6
190
N/A **'
224
56
*Cost estimates are for an average size facility as
identified in Table 1 and assume application of
best available control technology.
**Assumes the primary definition of BACT for plants
requiring stack gas cleaning. Costs for the alterna-
tive definition of BACT would vary from 0 to this
figure, depending primarily upon availability of
low sulfur coal.
***0perating costs are not available for these systems.
-------
90.
80..
70-
0)
60..
o
-------
9b
90 I
80 I
701
60 J.
o>
c
501
CD
O
X
40 4
a:
c_>
cy
2 emissions. (Allowance is 20% increase over 1970* emissions.)
* ]?7Pu1S^ sample year and was selected because of data availability
at the time this figure was prepared.
-------
96
REFERENCES
"Air Quality Criteria for Carbon Monoxide,"
National Air Pollution Control Administration
Publication No. AP-62, Washington, D.C.
March 1970.
"Air Quality Criteria for Hydrocarbons," National
Air Pollution Control Administration Publication
No. AP-64, Washington, D.C. , March 1970.
"Air Quality Criteria for Nitrogen Oxides," Air
Pollution Control Office Publication No. AP-84,
Washington, D.C., January 1971.
"Air Quality Criteria for Particulate Matter,"
National Air Pollution Control Administration
Publication No. AP-49, Washington, D.C.,
January 1969
"Air Quality Criteria for Photochemical Oxidants,"
National Air Pollution Administration Publication
No. AP-63, Washington, D.C., March 1970.
"Air Quality Criteria for Sulfur Oxides, "National
Air Pollution Control Administration Publication
No. AP-50, Washington, D.C., January 1969
"Atmospheric Emissions from Petroleum Refineries -
A Guide for Measurement and Control," Public
Health Service Publication No. 763, United
States Government Printing Office, Washington,
D.C., 1960.
"Atmospheric Emissions from Sulfuric Acid
Manufacturing Processes," National Air Pollution
Control Administration Publication No. 999-AP-13,
Durham, North Carolina, September 1970.
"Compilation of Air Pollutant Emission Factors,"
Office of Air Programs Publication No. AP-42,
Research Triangle Park, North Carolina,
February 1972.
-------
97
"Control Techniques for Carbon Monoxide Emissions
from Stationary Sources," National Air Pollution
Control Administration Publication No. AP-65,
Washington, D.C., March 1970.
"Control Techniques for Carbon Monoxide, Nitrogen
Oxide, and Hydrocarbon Emissions from Mobile Sources,"
National Air Pollution Control Administration
Publication No. AP-66, Washington, D.C., March 1970.
"Control Techniques for Hydrocarbon and Organic
Solvent Emissions from Stationary Sources.,"
National Air Pollution Control Administration
Publication No. AP-68, Washington, D.C.
March 1970.
"Control Techniques for Nitrcgen Oxide Emissions
from Stationary Sources," National Air Pollution
Control Administration Publication No. AP-67
Washington, D.C., March 1970.
"Control Techniques for Particulate Air Pollutants,"
National Air Pollution Control Administration
Publication No. AP-51, Washington, D.C .January 1969
"Control Techniques for Sulfur Oxide Air Pollutants,"
National Air Pollution Control Administration
Publication No. AP-52, Washington, D.C., January 1969
Cuffe, S. T. and Gerstle, R. k., "Emissions from
Coal-Fired Power Plants: A Comprehensive Summary,"
Public Health Service Publication No. 999-AP-35,
Cincinnati, Ohio, 1967.
Kenline, P. A. and Hales, J. K. ,"Air Pollution and
the Kraft Pulping Industry - An Annotated Bibliography,"
Public Health Service Publication No. 999-AP-4,
November 1963.
Kreichelt, T. E., Kemnitz, D. A., and Cuffe, S. T.
"Atmospheric Emissions from the Manufacture of
Portland Cement, "Public Health Service Publication
No. 999-AP-17, Cincinnati, Ohio, 1967.
"National Emission Standards Study," A Report to
the Congress of the United States by the Secretary
of Health, Education, and Welfare in Compliance
with Public Law 90-148 Clean Air Act, as'Amended,
National Air Pollution Control Administration,
March 1970.
Pooler5F. Jr., "Potential Dispersion of Plumes
from Large Po^er Plants," Public Health Service
Publication No. 999-AP-16, Cincinnati, Ohio, 1965.
-------
98
Schueneman, J. J., High, ML D., and Bye, W. E
"Air Pollution Aspects of the Iron and Steel
Industry," Public Health Service Publication
No. 999-AP-l, Cincinnati, Ohio, June 1963.
Smith, W. S., "Atmospheric Emissions from
Fuel Oil Combustion," Public Health Service
Publication No. 999-AP-2, Cincinnati, Ohio,
November 1962.
Smith, W. S. and Gruber, C. W., "Atmospheric
Emissions from Coal Combustion - An Inventory
Guide," Public Health Service Publication
No. 999-AP-24, Cincinnati, Ohio, April 1966.
State Implementation Plans as required uv
Section 110 of the Clean Air Act, as amended.
Turner, D. B., "Workbook of Atmospheric
Dispersion Estimates," Office of Air Programs
Publication No. AP-26, Research Triangle Park,
North Carolina, July 1971.
-------
FROM:
TO:
Reference 15
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
.iiJiijKCT: Non-degradation and Power Plant Size
DA'Ilv.
AUG i 2 1974
Joseph A. Tikvart, Chief
Model Application Section, SRAB
Richard G. Rhoads
Special Assistant, OAQPS
THRU: Herschel H. Slater, Chief
Source Receptor Analysis Branch^, MDAD
At your request, I have reviewed the allowable concentration
increases applicable to a Class II area designation for prevention
of significant deterioration. The size (megawatts) power plant which
conforms with specified concentration increments was examined. Simi-
lar studies by TRW for the Department of the Interior and by Harbridge
H6use for EPA were also reviewed.
As a result of this study, I find it likely that an 800 MW power
plant which meets new source performance standards will not exceed the
allowable 24-hour concentration increment (100 yg/m3). Generally, this
same plant will not exceed a 3-hour concentration increment of 600 ug/m3.
(A stack height of about 400 feet has been assumed.) Also, in most
cases a 1,000 MW plant will not exceed the 24-hour increment. However,
for a plant of this size, an allowable 3-hour increment of 750 ng/m3 is
more appropriate. The'plant sizes discussed above will increase, if
greater stack heights or reduced pollutant emissions are considered.
It should be noted that a power plant in the size ranges discussed here
may endanger a Zone I increment for a distance of 80 miles downwind under
poor dispersion conditions.
As you are aware, the impact of individual power plants on air
quality is highly variable. To determine impacts for specific power
plants, a plant-by-plant analysis is required. The above plant size
estimates are general "rules of thumb" only.
The basis for these findings is documented in the enclosure.
Enclosure
EPA Form 1320-6 (Rev. 6-72)
-------
100
REVIEW OF NON-DEGRADATION LIMITATIONS ON POWER PLANT SIZE
TRW Study
The study by TRW attempts to relate an allowable power plant size
which meets non-degradation requirements to various areas of the United
States. It concludes that a power plant with a 400 foot high stack and
emissions which meet new source performance standards (NSPS) should be
less than 800 MW in size if located in the Southeastern United States.
If located in other parts of the United States, a size less than 1500 MW
is required.
Unfortunately, the report does not document the basis for relation-.
ships used between concentration averaging times or between regions.
Nor is the basic dispersion model which was used identified adequately.
Thus the conclusions of the study can be considered to be tenuous, at
best.
Harbridge House Study
The study by Harbridge House considers the impact of power plants
in three regions of the United States, e.g. Boston, Four Corners Area, and
Iowa. The impact of individual power plants in Boston were not con-
sidered in this report. The power plants considered in the Four Corners
Area and Iowa were assumed to meet NSPS and to have a stack height, of
250 feet. From the concentration estimates for these power plants, it
was concluded that a plant size ir the range of 500 to 800 MW allows the
acceptable 3-hour increment to be met. Concentration estimates for an
averaging time of 24 hours were not made.
-------
101
It should be noted that the model used in the Harbridge House
Study is basically applicable to estimating long-term average pollutant
concentrations. As such, the algorithm which determines the horizontal
plume dispersion is 3-4 times greater than a similar algorithm used in
estimating short-term (1 hour) concentrations. Data collected by TVA
around the Paradise Steam Plant indicates that the ratio of 3-hour
average concentrations to the 1-hour average was 0.8 as much as 20% of
the time. The peak, median, and minimum ratios were 0.94, 0.66, and
0.44, respectively. Thus it can be concluded that the 3-hour average
concentrations estimated by Harbridge House may be low by a factor of
2 to 4.
Source Receptor Analysis Branch (SRAB) Analysis
To obtain a crude estimate of the power plant size which will not
exceed the allowable concentration increments for a Zone II area, SRAB
employed the point source dispersion model described in "Requirements for
Preparation, Adoption and Submittal of Implementation Plans—Appendix A"
(FR 36 15494). This model of limited mixing is the critical plume dis-
persion model for sources with tall stacks. TVA has indicated that such
a model may be applicable at distances of 5-10 kilometers downwind from
the source. Meteorological conditions of a 500 meter mixing height and
a wind speed of 5 m/s were assumed in the model. While these conditions
severely limit dispersion, they are not uncommon in many parts of the
United States. More severe conditions also occur in many areas. The
concentration estimates from this model indicate that a power plant of
-------
102
600-800 MW should not endanger the applicable increment for 24-hour
averages; however, the size must be limited to 400--600 MW to meet the
allowable 3-hour increment. So that a 600-800 MW plant could meet both
the 3-hour and 24-hour increments, the allowable 3-hour increment would
have to be increased to 400-450 yg/m3.
A related question is concerned with the location of such a power
plant near a Zone I boundary. Zone I allowable concentration increments
are only 5-13% of Zone II allowable increments. By using the limited
mixing model indicated above, it was determined that a plant which just
meets the Zone II increment, may exceed the Zone I increment for a dis-
tance as great as 80 miles downwind. Thus it appears that such a power
plant should not be located any closer than 80 miles from the nearest
Zone I boundary.
To supplement the"basis for assuming that a power plant larger than
600-800 MW could exceed both the 3-hour and 24-hour Zone II allowable
increments, data from the EPA funded Wai den Power Plant Study were
employed. First, all power plants with actual stack heights in the
range of 400 to 600 feet were selected. The reason for this selection
is an arbitrary definition of good engineering practice (GEP) stack
height. In its simplest form, in flat terrain a GEP stack is one which
is 2-1/2 times the height of nearby structures. Since nearby structures
routinely include a boiler house the stack should be 2-1/2 times this
structure. From the data available it appears that typical boiler houses
on new plants are about 150 feet high; this would require a GEP stack of
-------
103
400 feet. Next, the relationship of stack height to surrounding terrain
was examined. Only those 15 plants with stacks 300 feet or higher than
surrounding terrain were analyzed further. These stacks were reviewed
for their impact on 1-hour, 24-hour, and annual average concentrations
with the assumption that they met NSPS. The impact with NSPS was then
compared to the allowable Zone II concentration increment.
The attached figures indicate plant size (MW) versus maximum concen-
trations for 3-hours and for 24-hours for plants assumed to meet NSPS.
Lines indicate the relationship between plant size and concentration that
is necessary for 2000, 1000, and 800 MW plants to meet required increments.
Plants above the line will exceed the allowable increment; plants below
will be within allowable concentrations. For the 24-hour average concen-
trations it is quite apparent that if the plants investigated were 2000 MW
in size, more than half would cause the 24-hour increment to be exceeded.
A 1000 MW plant would allow most of the plants to meet the allowable incre-
ment, while a 500 MW plant would insure that all plants would meet the
allowable increment. This is consistent with the earlier finding that a
600-800 MW plant is required to meet the allowable increment.
Estimates for 3-hour concentrations were not available for most of
the plants considered, only 1-hour averages. For 3 plants on which data
were available, the 3 hour/1 hour ratio varied from 0.44 to 0.69; this
is within the range of observed ratios discussed above. Thus to be con-
servative, the 3-hour ratio of 0.8 also discussed above is employed here.
The estimated maximum 1-hour concentrations were reduced by 20% to estimate
-------
104
the 3-hour concentrations. This is done with the understanding that this
ratio has wide variation.
For 3-hour average concentrations it is clear that any plant larger
than 400 MW can cause the 3-hour increment to be endangered. Though there
are plants that exceed the allowable increment, none of these plants have
stacks greater than 400 feet. It appears that if the allowable 3-hour
increment is doubled to 600 ug/m3, most 800 MW plants should not exceed
the standard. This estimate is somewhat greater than that obtained from
the limited mixing model analysis.
-------
CU
M
i-
aj
o
o_
s_
0)
O)
u
c
o
X
rt3
OCfFI
001
-------
01
M
i-
OJ
O
a.
c
O
c
01
c
O
X
ro
•a
0)
09L 01? L
0£L OZL DLL
(Em/6rt)
OOL
oe 02 OL
-------
Reference 16
Office of Air Quality Pianrrino and Standards
f'rsrarch Triannle Park, North Carolina 27711
Significant potcrioration in 7onc I /Yeas and tho Oct. IB, IC'7'-
Relative Location of Po'./or Plants
Joseph A. Tikvart, Cliiof
f'odel Application Section
The Files
Tho 7ono I allowable ?A hour isvornn^ concentration increments 1o
prevent significant deterioration for cno and particulate r>attr.r aro
5 iia/n3 and 10 ii'Ym3» respectively. It Is nossible that a nov;er plant
located in a Zone II or 7.rnc III aron could root the allov/ablo increment
for these latter areas hut still exceed the allowable increment for
adjacent Zone I nrca. This brief analysis considers the distance a plant
must to located frcr - 7rne I nrra to avoid exccedinq the^lTcvMDl a
concentration incrcmontr>.
The concentration that a source causes at a niven dov.-nv/ind distance
depends on the pollutant emissions, physical plant characteristics rnd
meteorological conditions. To simplify this analysis a few nenpraV'zpd
assumptions arc made. "The source is assured to bo a lOOO'-'l! pover r.rnt
which roots nc>v source nerfornnnco standards (iiSPS); thus its SO.^ and
particulate emissions are estirotftJ to be ^pproxinatoly 1000 o/s^and
100 p/s, respectively. Tho plant has a 400 foot hich r.tack v/ith a
pluino rise equal to the stack hninht, o.q., cfrective sback heioht is
approximately 250 roter^. It is further assumed that vertical nixino
of the plunc is linitfd to 500 raters and that the avorpno v/ind spend in
5 T'/S (IT. Hlo-^torT. nor Imur): ;--lio conui Liens ny cct:ur
with a steady vind (lir«^ct.i(Mi for periods up lo '~ hours; i-ov/over, 11;1 '.n:,:
direction itself may persist for considerably lonqor periods.
Such peteoroloqicnl condition'; rcsultina in poor dispersion of thn
pollutant arc not uncornon in ranv parts of the United Slates. F.'vor: ;n
areas rith ccncrally rood dispcrsun r;or,dit.ion.~t such adverse condition;;
may occur on a limited number of occasions.
I'ith the sbc.v*' irdicatrc! plrsit rivd rTtr~or''lor*i>Ttl '"'iji'iitfor'i. ^!'if>
paxiiur.\ ^-l-l/'-ir C°^ cr-^c^-'.ri^ic':; rrtr.rij uy Lr:. nlai;t vould 'r ;i|;pr'.)y"-
pately 100 yn/m3 at r •!i-tr:r,cn of 3 '.iloivctorr, •!f'7nvirvj frc~> tho nl:ni:.
Thus the cor.ccntrTjIcr .-Its1: r'::M,<; th^ "'TC IT Allcvnblo incrr "f-'t, i'i-c
channc 1n "0 cc:i';cnf'.ral!on \-it!i '.,'1st once dcvn-./ini frcn i(,n p!>ni. -\:.
shown in FifL-rr 1, It can bo s'-on tii^t if mo oVcny nf SH0 is .-.ssi'^r-''
the concentration roachos a level of r ya/r,i3 at a distance" of bO-100
kilometers. If an ro,, half-life of 1? or 4 hours is Assumed, th.o ''eve!
of 5 yQ/m3 is reached"'closer to the stack. It. should be noted that
beyond 80 kilorG'tcrs i;i.r concontrr'tion dccroasc-s vorr slov/ly with incrra^r
in distance. Siiice F1wrc 1 has a scale of 0-100 for concentration, these
107
-------
108
?.
''
values can F~o used as a prrccntriee. Thr-rcfor^ if n sipilar ?OOP"-'.'
plant in a 7onc III area caused n iraxInuTn concentration of 200 yn/;n3,
then a concentration of F>&'. of this value or 10 ua/ru3 v:ould result at
a distance of CO-1CO kilometers, 1f there was no 500 decay. Due to
tlio relative- stringency of the 7one I ?4 -hour ?nn Increment, it is the
rrost difficult to reot. In rest cases, if the ?4-hour increment is mot,
then the 3-hour and arnual incr 'rents are a"; so lil'ftly to bo net.
A distance of lnO Mlrretrvs at a wind srood of 'J rotors nor second
requires a travel tin-1 of nearly r> hours. Thu:;, if the phenomena which
causes the significant concentrations at 100 kilometers also "Insts for
C hours, a persistent wind ^>-r ]? hours is irplicit in the ;»nn'Nsis.
Since tli^.re is Vittln information on dispers'Icn beyond IPO k1lo!"eters
and any cssunption of persistence for rericds lonoer than 12 hours are.
extrenely tenuous, Fiourc 1 is not extended beyond 100 kiloincters.
Such an extension should be ottenpt^d only for specific tonooraphic and
cl1niatolof|ical conditions and cnly then vn'th orrat care.
Finure II indicates 24-hour participate concontrationr, caused by A
100C-!'.VI plant r.eetinn f'SPS. It can be seen that such a v-ell controlled
plant will meet the 7.one I allowable concentration increment of 10 yn/ri3 .
Even 1f S00 viith P. bnlf-life "o'- 4 hours Is converted entirolv to
sulfate and treated as particu'ate patter, ?. concentration of 10
is reached v.'ithin 10 Mlor^otors downwind of the source. For laroer
sources a proportional increase in raxinun concentrntion can b>e assumed,
and the percent decrease with distance irpliec! by Fini.'re II used.
Thus, from the above analysis it can bo concluded that: It is
reasonable to expect a 10QOM1! plant nretinq r'SPS to endanncr the Zone I
SCL Increment to a distance of 50-CO riles dov/nv/1nc'. For larner plants
located for example in a Zone III area this Increment vill be endannereci
for nreater distances dowr.vind, A ^ens,rate analysis for such silualions
on an individual basis appears ''eslrable, "bo Irpact of such sources on
the pcrticulate increment does not appear to ':-e a problem.
It should be noted that the above anr.lvsis is for o oenrral i?.ed sourc.'",
Thus, conclusions tfrwr about downv:in'J dist.apces at v.'in'ch the plant hrs
a sinnificant effect can Lc considered ''rules t;f thurb", at best. If
any changes in the; niiysic-il ciiarncterintics of the pl?nt or ^>o rcteoro-
logical conditions are rade, 5;inr,ificantly different concentrations rw
b"c estimated at dov/nwind distances. This is especially irportnnt since
relatively snail channes in concentration arc- critical in the consideration
of air quality deterioration.
cc: K. Perry ,^-~-~
-------
109
Figure 1
SO- Concentration versus Distance
Limited Mixing:
Wind Speed :
'C' Stability :
Plant size
500 m
5 m/s
1000 MW
No decay
20 si ww si TO"sir
Downwind Distance (Kilometers)
12 hour half-life
4 hour half-life
90 TOO
-------
no
Figure 2
Particulate Concentration versus Distance
Limited Mixing: 500m
Wind speed : 5 tn/s
"C" Stability :
Plant Size : lOOOMlN
Addition of S02
converted with a
4-hour half-mile
to (NH4)2 S04
Addition of S02
converted with a
12-hour half-life
to (NH4)2 S04
No conversion
0 10
20 30 4C 50 60 70 80 90 100
Downwind Distance (Kilometers)
-------
Reference 17
DISCUSSION PAPER ON THE
MAGNITUDE OF THE CLASS II INCREMENT
IN THE SIGNIFICANT DETERIORATION REGULATIONS
THE ISSUE
The draft regulations incorporate the same general definition of the
Class II increment as was incorporated in the proposal of July 16, 1973,
and incorporate the same provision that this increment serve as the initial
area classification nationwide. It has been suggested that the increment
be doubled.
EXPLANATION
The primary reason for doubling the increment is to permit a larger
growth potential - with primary concern for coal-fired power plants -
during the interim period until States redesignate their territory. The
provision for States to redesignate their territory according to local
desires is not an issue. The provision for EPA to perform an initial
designation as Class II is also not an issue. The issue is solely
whether the Class II increment should remain as proposed, or whether it
should be doubled.
DISCUSSION
Point 1. The Class II increment is adequate to permit normal,
well-controlled, commercial and industrial construction in most areas
of the country. The current increment was developed to permit the amount
and type of polluting growth which is, in the subjective judgment of EPA,
generally "acceptable" to affected populations in most areas, but to pre-
vent the type of growth (such as very large coal-fired power plants) which
generally creates a strong adverse reaction by many local populations.
The increment will permit construction of typical new coal-fired
power plants (which must meet New Source Performance Standards and must
use stack heights corresponding to good engineering practice) in the range
of from 800 megawatts to 1000 megawatts. The variation is due primarily
to differing terrain features and meteorological conditions in particular
sections of the country, and to specific siting conditions. These data
were developed by EPA, and are generally consistent with the studies
conducted by other agencies and contractors. The data are somewhat
more precise than earlier studies because most earlier studies could
not assess the impact of "good engineering practice" stack heights.
Ill
-------
2
The increment will therefore support new power plant units which are
slightly larger than the average individual units planned for construc-
tion. Testimony provided by a representative of El Paso Natural Gas
confirmed that the Class II increment would £.lso not restrict their
planned coal gassification facility. Similar analyses indicate that the
Class II increment will support larger than Average si/o fatilities oi
other1 major commercial and in.lustr in I .utivUies although large f,u ilities
(including oil shale extract!'in), would likely require redesignatiun to
Class III.
Point 2. The_ -J^s_s_J_J__jj)cr_ement wo u 1 _d_ (and_ shou 1 d) i nhi bit
r!_q^
_r.g.eiPo'l lution-prone facil i ties, and large cju_sters~p7
smaller facilities. Data on new fossil-fuel fired power plants currently
planned for construction through 1982 shows that future plants will be
substantially larger than those to which the public is currently accustomed.
The average new unit size is !>70 MW which is nearly double the average
size of existing plants. There are 103 new plants planned, with an
average size of 970 MW. The "argest plant will be 5200 MW. Six plants
will be larger than 2500 MW, 22 will be larger than 1500 MW, and 37
will be larger than 1000 MW. Most of these plants will be prohibited
from construction in Class II areas.
The large new plants wil" not be uniformly distributed. The 11
SERC (Southeast) plants will average 1480 MW each, the 28 ECAR (East-
Central) plants will average 1160 MW, and the 19 WSSC (Western) plants
will average 1050 MW. This trend toward substantially larger new plants,
and substantial expansion of existing plants, is expected to continue.
One of the major goals oi the "significant deterioration movement"
has been to insure that these very large plants are properly controlled
and located, but not necessarily prohibited. The plans for "massive
coal-fired power plants" were repeatedly referred to in the public debate
as specific examples of the t>pe of growth on which public attention
should be focused. It is therefore appropriate that the Class II incre-
ment should be sized to place these large plants in a category different
from that of the smaller plants.
Point 3. JjT§_J nitial designation as Clas s 11 will not, di srupt
consjtnj ctj o n of most large povv er plant's due to~~nT the six-month grace
period and~f27 the provision for rapid redes i gnat ion to CjAS_s_LLL_ The
initial designation will directly affect only those power plants which
(1) have not received construction approval prior to six months after
promulgation (prior to that time sources are not subject to review for
significant deterioration) and (2) have planning schedules which require
construction approval before the State can consider redesignation of the
local area (after the State has considered redesignation, it is the State
decision, not the initial designation, which affects the power' plants).
-------
113
3
The regulations do not require a State to redes i
-------
114
facilities. In so doing, it is likely that some forms of desirable
growth would also be prohibited as a byproduct.
Point 6.' Uojubljixj the Class II increment wo_ukl _ro<|uire develo])-
nien_t_ of entijej^ new rajjonale ajid_ j_u_s_t i f_u a (.Km _fo_r_ selection of_ the
nl'^1'n^:r^i11§H^- ^ne current increment permit^ introduction of well-
controlled, moderate sized facilities, and excludes the very large
pollution-prone facilities which often encounter massive public
resistance. The Class II increment is intended to apply to areas of
"moderate," "normal," or "average" growth, with populations which have
"typical" attitudes toward balancing of environnertal and other con-
siderations.
This rationale could not be applied to an increment which would
permit introduction of very large power plants. These massive facilities
do encounter strong opposition in many areas of the country. They are
not "moderate," "normal," or "average," and are not generally acceptable
to "typical" populations.
Point 7. To double the Class II increment^t this late date
wjthout stronrj_ju_stjfjc_at_i_on would likely be viewed with suspjcion by
the pub 1 i c and ,d j_sf a vo_r_ by the Cput;t. The current Class II increment
has received extensive debate and general acceptance in the public arena.
Although many comments questioned the basic concept of preventing signi-
ficant deterioration, there were no major chal'enges to the philosophy
that the Class II increment should permit a reasonable amount, of growth
in most areas of the country, except for challenges that the increment
is too large. In fact, there is nothing in the public record that
would support doubling of the Class II increment in these regulations.
The Acting Administrator stated in the July 16, 1973, proposal
that, in his judgment, "Deterioration above the Zone II levels would
constitute ... a significant deterioration in most areas of the country."
It is unlikely that the Court, and the public, could be provided any
new data or rationale which would justify changing that, judgment.
Possibly the most compelling argument against doubh'n
-------
Reference1 ?0
Emissions of Sources Subject to Significant Deterioration Regulations
The 1970 emissions for the 17 largest source categories are shown
in Table 1. The total 1970 emissions, excluding sources such as forest
fires and agricultural emissions, were: 22.3 million tons for TSP and
33.3 million tons for SO,,. Therefore, existing sources of the type
subject to the significant deterioration regulations emitted about 30%
of the TSP and 75% of the S02 emitted nationwide in 1970.
Table 1 - 1970 Emissions for the 17 Largest Source Categories, tons/yr.
Source Category TSP
p
Fossil Fuel Fired Steam 3,000,000 13,465,000
Electric Power Plants
Coal Cleaning Plants 102,000
Kraft Pulp Mills 796,000
Portland Cement Plants 280,000
Primary Zinc Smelters 22,000 140,000
Iron and Steel Mill 817,000
Metallurgical Furnaces
Primary Aluminum Ore 74,000
Reduction
Primary Copper Smelters 71,000 4,021,000
Municipal Incinerators 450,000
Sulfuric Acid Plants 1,757,000
Petroleum Refineries 211,000 4,393,000
Lime Plants 523,000
Phosphate Rock Processing 334,000
By Product Coke Oven 153,000 437,000
Batteries
Sulfur Recovery Plants 750,000
Carbon Black Plants 215,000
Primary Lead Smelters 1 ,200 494,000
TOTALS 7,249,000 25,457,000
115
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
3. RECIPIENT'S XCCESSION'NO.
4. TITLE AND SUBTITLE
Technical Support Document
EPA Regulations for Preventing the
Significant Deterioration of Air Quality
5. REPORT DATE
January 1975
6. PERFORMING ORGANIZATION CODE
N/A
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
EPA, OAQPS,OAWM
N/A
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Same
10. PROGRAM ELEMENT NO.
547853A991
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Same
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document provides a detailed explanation of the technical and policy
condiserations which form the basis for EPA's regulations for preventing the
significant deterioration of air-quality, which were published in the
Federal Register on December 5, 1974. The document discusses the reasons
for rejecting various alternative plans for preventing significant deterioration
of air quality; the basis for the air quality increments selected and the
pollutants and sources covered; anc the impact of the regulations on general
growth and development as well as on new energy sources. A number of the
references are included as appendices; however, several of the studies and
reports referenced are too voluminous to include in this document. These
reports are available for public Inspection at EPA's Freedom of Informantion
Center in Washington, D.C.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Prevention of significant deterioration
18. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (ThisReport)
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Form 2220-1 (9-73)
116
------- |