United States Office of Air Quality EPA-450/5-79-002
Environmental Protection Planning and Standards February 1979
Agency Research Triangle Park NC 27711
Air
Cost and Economic
Impact Assessment for
Alternative Levels of the
National Ambient Air
Quality Standards
for Ozone
-------
EPA-450/5-79-002
Cost and Economic Impact
Assessment for Alternative
Levels of the National Ambient
Air Quality Standard for Ozone
Strategies and Air Standards Division
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
February 1979
-------
PREFACE
In accordance with the provisions of Sections 108 and 109 of the
Clean Air Act as amended, the Environmental Protection Agency has
conducted a review of the criteria upon which the existing primary and
secondary photochemical oxidant standards are based. The Act specifically
requires that National Ambient Air Quality Standards be based solely on
scientific criteria relating to the level that should be attained to
adequately protect public health and welfare. Based on the wording of
the Act and its legislative history, EPA interprets the Act as excluding
any consideration of the costs of achieving those stanndards or the existence
of technology to bring about the needed reductions in emissions. However,
in compliance with the requirements of Executive Orders 11821 and 11949
and OMB circular A-107 and with the provisions of the recently issued
Executive Order 12044 for rulemaking proceedings which are currently
pending, EPA has prepared an assessment of the potential cost and economic
impacts associated with efforts to attain the promulgated standard as well
as alternative levels of the standard. This document presents the results
of this assessment.
The purpose of the analysis contained herein is to estimate the
relative ranges of national control costs for alternative levels of the
ozone standard. In addition, in order to compare the relative implications
of alternative standards, the range in the number of Air Quality Control
Regions (AQCRs) which might be expected to attain the alternative standards
given various assumptions is also indicated. Because of the many uncer-
tainties in projecting emission levels and air quality levels and in
-------
determining effective control strategies throughout the nation, it is
important to fully recognize that the results of the analysis should be
viewed only as general guidance which provides an indication of relative
differences in the attainment picture and the associated costs between
alternative levels of the standard. The analysis cannot be used to
precisely determine how many or which specific AQCRs will attain a given
ozone standard through particular control strategies. Rather, attainment
status and control requirements for attainment will have to be determined
for each geographical area based on the unique conditions that are
inherent for that area. Likewise, this analysis cannot ascertain with a
great degree of precision the costs of control strategies that will be
required for all areas of the country to attain alternative standards.
Since the actual control costs will be extremely variable, this analysis
is useful in only presenting the relative implications for costs between
alternative levels of the standard.
-------
TABLE OF CONTENTS
Page
Chapter 1.0 Introduction and Summary 1-1
1.1 Approach and Methodology 1-1
1.2 Limitations of the Analysis 1-4
1.3 Major Findings 1-9
Chapter 2.0 Baseline Air Quality and Allowable Emission
Levels 2-1
2.1 Existing Ambient Air Quality Levels 2-1
2.2 Allowable Emission Levels 2-2
2.3 References 2-8
Chapter 3.0 Projection of Emissions in 1987 and Reasonable
Levels of Control Achievable 3-1
3.1 Motor Vehicle Emissions in 1987 3-2
3.2 Projection of Uncontrolled Emissions from
Stationary Sources 3-4
3.3 Emission Projections with Controls on New
Stationary Sources 3-7
3.4 Reasonably Available Control Measures for
Existing Sources 3-8
3.5 Total Emission Reductions and Additional
Reduction Required 3-13
3.6 Variation of Results 3-15
3.7 References 3-20
Chapter 4.0 National Costs of Control for Alternative
Levels of the Standard 4-1
4.1 Costs for the Federal Motor Vehicle Control
Program 4-1
4.2 Costs for New and Modified Stationary Sources 4-2
4.3 Costs of Applying RACT for Stationary Sources 4-4
111
-------
Page
4.4 Costs of Applying RACT for Mobile Sources 4-7
4.5 Total Cost of Applying RACT for Alternative
Standards 4-9
4.6 Summary of Costs for Identified Measures 4-16
4.7 Estimated Cost of Attainment 4-18
4.8 References 4-23
Chapter 5.0 Economic Impact of Reasonably Available
Control Measures on Selected Industries 5-1
5.1 Introduction 5-1
5.2 Petroleum Refining 5-2
5.3 Retail Gasoline Service Stations 5-5
5.4 Gasoline Bulk Plants 5-11
5.5 Automobile Assembly Plants 5-16
5.6 Metal Furniture Industry 5-20
5.7 References 5-25
Appendix A Ozone Design Values for 90 Air Quality Control
Regions A-l
Appendix B Mobile Source Emission Factors B-l
Appendic C Analysis of Costs for Hydrocarbon Control
Measures C-l
Appendix D Analysis of Costs for the Federal Motor Vehicle
Control Program D-l
Appendix E Cost/Effectiveness of Inspection and
Maintenance Programs E-l
IV
-------
1.0 INTRODUCTION AND SUMMARY
The Clean Air Act, as amended in 1977, requires the Administrator
of the Environmental Protection Agency to periodically review the basis
of the ambient air quality standards. The National Ambient Air Quality
Standard for photochemical oxidants has been reviewed and a new form of
the standard has been promulgated. As part of this review procedure,
this report presents the results of an analysis of the potential impact
of feasible changes in the standard on national costs of control and the
attainment status for various areas of the country. An analysis of the
potential socio-economic impact of the alternative standards has not
been conducted for every affected industry and region of the country,
although an assessment of the impact of control costs on selected industries
is presented to give an indication of the magnitude of the impacts.
1.1 APPROACH AND METHODOLOGY
This report includes an analysis of 90 Air Quality Control Regions
(AQCRs) which currently exhibit ambient ozone concentrations in excess of
the current photochemcial oxidant standard (.08 ppm hourly average not to
be exceeded more than once per year). For each AQCR this analysis estimates
potential emissions in 1987 and potential emission reductions achievable
with the Federal Motor Vehicle Control Program (FMVCP), new and modified
source control, application of reasonably available control technology
(RACT) on existing stationary sources and further motor vehicle controls
through inspection/maintenance and transportation control plans. Based
on the projected emission reductions, control costs are estimated for
applying technology in an attempt to attain alternative standard levels.
While the analysis considers each AQCR separately, the results are presented
in aggregate form for all 90 AQCRs instead of each individual AQCR.
1-1
-------
This analysis uses the 1975 emissions inventory from NEDS as the
baseline emissions for each of the AQCRs. From this base 1987 emissions
are projected based on statutory automotive emissions standards and
reasonable ranges of assumptions for growth in vehicle miles traveled
(VMT), growth in production for new sources, and retirement rates for
existing sources. In addition, assumptions relating to the control of
new and modified sources are made in order to determine emissions with
only new source and statutory motor vehicle controls.
In order to calculate the emission reduction required, if any, for
existing sources to ensure that alternative standard levels are met, the
following approach is employed. For each standard level, the maximum
allowable emission levels are calculated for each AQCR based on baseline
ambient concentrations, 1975 emissions of non-methane hydrocarbons (NMHC),
the level of the alternative standard, zero background ozone concentrations
in urbanized areas, and assumed relationships between hydrocarbon
emissions and ambient ozone concentrations according to alternaitve
modeling approaches. These two approaches are the linear rollback model
and the Empirical Kinetic Modeling Approach (EKMA).
Once the allowable emissions have been calculated for each AQCR,
they are compared with projected 1987 emissions levels taking motor vehicle
tailpipe controls and new stationary source controls as the baseline. If
additional reductions are required, available control measures are then
placed on existing stationary sources that have not been replaced as well
as in-use vehicle inspection/maintenance programs or VMT reduction plans for
control of mobile source emissions. RACT is assumed only for those specific
stationary source categories which have been identified and for which
1-2
-------
estimates of control technology and efficiencies have been made. While
these sources represent the primary sources of volatile organic compounds
(VOC), other sources exist, primarily small solvent evaporation sources,
which have not been identified by EPA. As an extension of the analysis,
reasonable levels of control for these miscellaneous solvent sources are
assumed in order to determine their effect on the results of the analysis.
In many instances, control of identified sources and reasonable transpor-
tation controls will not result in the required emission reductions,
meaning that additional control measures will have to be applied in order
to meet the alternative standards. This study does not consider
additional control measures, rather only the additional emission reductions
needed are indicated.
Once the control measures have been identified, the costs of the measures
for the specific sources are then analyzed and estimated. The average cost
per ton of emissions controlled is established for general categories of
sources and combined with estimates of emission reductions achievable for
the 90 AQCRs in order to obtain total annual costs of control in 1987.
Costs are included for the FMVCP, new source control, and application of
RACT for identifiable sources to the extent required for each of the alter-
native levels of the standard. For areas that need additional control
beyond the identified control measures, a rough estimate of the additional
cost to attain the alternative standard is made which will reflect the
magnitude of differences in total costs of the alternative levels of the
standard. The costs presented in this analysis represent direct annualized
costs of control which will be incurred in 1987, when it is assumed that
1-3
-------
the control measures will be applied fully to the applicable sources.
No estimate of total capital costs is made. In addition, there is no
consideration of the secondary costs (i.e., costs beyond those encountered
by the source being controlled) or benefits associated with the control
measures.
1.2 LIMITATIONS OF THE ANALYSIS
In any analysis of this type many simplifying assumptions have to
be made because of the uncertainties that surround the problem. The
choice of assumptions, analytical tools, data bases, and approach will
profoundly affect the conclusions of the analysis. This analysis attempts
to place reasonable ranges on the assumptions in order to estimate the
range of national costs for controls necessary to attain a given ozone level.
Since the assumptions are so crucial, this section outlines the basic
assumptions employed in the analysis and the limitations the choice of
these assumptions place upon the results of the analysis. Because of the
uncertainties, results using differing assumptions are presented in this
analysis.
First of all, this analysis considers 90 AQCRs for which validated
data on ambient concentrations exist which indicate violations of the
current standard or which contain urbanized areas with populations
greater than 200,000 which have been designated non-attainment by EPA.
Undoubtedly, there are many other smaller urban areas as well as rural areas
which currently also experience violations of the standard. However, these
latter areas are not considered in the analysis since validated data have
not been compiled for these areas. While this exclusion will understate
the costs somewhat, it is not expected to be significant since these areas
1-4
-------
probably do not significantly exceed the standard and since the AQCRs
which are considered represent well over 60 percent of the nation's population.
In order to calculate the emission reductions required in each of
the AQCRs in 1987, assumptions must be made concerning the emissions
inventory, baseline air quality values, growth rates, and the modeling
technique to be used.
This analysis uses the preliminary 1975 emissions inventory from
the NEDS Emission Summary Report, which is the only source of aggregate
information on emission data for all 90 AQCRs. No attempt was made to
verify the emissions inventory with data from state and local agencies,
whose inventories are in some cases at variance with NEDS. Although
there are some inconsistencies between NEDS and local inventories, NEDS
nonetheless provides a consistent procedure for conducting the inventories
throughout each AQCR and is believed to be the best available source of
emissions data for national assessments.
Several reasons exist for the differences between NEDS and local
inventories. However, the most significant difference is that NEDS
includes significant emission contributions from miscellaneous
sources of organic solvent applications. Local inventories do not
usually include the small miscellaneous sources whose emissions are
relatively easy to calculate nationally based on national production
statistics but are extremely difficult to calculate on a local basis.
Much uncertainty exists as to the exact composition of sources in this
ill-defined miscellaneous category. EPA currently has underway an extensive
study to better define, classify and locate the sources of solvent evaporation.
Conclusive results are not yet available.
1-5
-------
The design ozone values used in this analysis represent only an
approximation of the ambient ozone concentration that on the average
will be exceeded once per year in each AQCR. These values were obtained
from an analysis of air quality data for the years 1975 to 1977 using
methods described in Guideline for Interpretation of Ozone Air Quality
Standards (OAQPS No. 1.2-108). Analysis of air quality data indicates
that this value should fall between the third and fourth highest daily
averages. These approximate values were derived only for analytical purposes
for use in this assessment. In the SIP revisions submitted to EPA, the states
will calculate the actual design values which will be used for attainment
determinations and for planning purposes.
Growth rates are crucial parameters in determining the overall
emission reduction which will be required in 1987. This analysis uses
ranges of national average rates of growth in industrial production for
various classes of industrial sources as well as ranges of national average
rates of growth in vehicle miles traveled. The range brackets the individual
growth rates for specific areas. This approach of using national growth
rates simplifies the growth projections but introduces some distortions
for some local areas. First of all, there is the implicit assumption that
every AQCR has the same mix of sources as the nationwide mix on which the
average growth rates are based. Secondly, all areas of the country will not
experience the same rate of growth. For example, many highly industrialized
areas with ozone problems may experience lower rates of growth than less
industrialized areas which are attracting new industries. Ideally, for
each AQCR, the analysis should use regional growth rates which reflect
local economic conditions and the mix of sources in an area. However, results
from this analysis show that for standards of .10 ppm or higher, at most
four AQCRs change attainment status by shifting from high to low growth rates.
This result is not considered significant.
1-6
-------
Another variable in the analysis is the modeling techniques employed
to estimate the allowable emissions for areas to attain alternative
ozone standards. Depending on the technique used, wide variations can
result in allowable emission levels and the concomitant reduction in
projected emissions that is required. This analysis includes results from
application of the linear rollback model as well as the Empirical Kinetic
Modeling Approach (EKMA). A judgement as to which modeling approach is
more valid or accurate is not made.
The emission reductions that can be achieved with application of
various controls measures is still another source of variability in the
analysis. The emissions inventory is segmented according to 10 broad
categories of industrial sources and three categories of mobile sources.
Each of the stationary source categories is composed of many diverse
industrial sources which are subject to varying levels of control. Thus,
a weighted average control level has to be estimated for each broad category
based on the assumed control levels and overall emissions for each
individual source category. When this weighted average emission potential
is applied to each broad category in each AQCR, it is again assumed that
the mix of sources in each AQCR is the same as the nationnwide mix.
Thus, this assumption could well affect the results of the analysis for
any AQCR.
Furthermore, RACT has not been identified or estimated with any
certainty for all sources of hydrocarbons. These sources are primarily
in the solvent evaporation category, where the specific sources have not
been identified, control technology and efficiency have not been assessed,
and whether or not these sources are amenable to control in the first place
1-7
-------
has not been determined. Without this information currently available
on the remaining sources, further controls are subject to speculation.
As an extension of the analysis, this analysis assumes control of these
sources in order to determine their impact on control costs and attainment
projections.
The approach used in defining control costs is also likely to
introduce another area of variability in the analysis. For each of
the many sources of VOC, the cost per ton of pollutant removed was
determined based on estimated capital and operating costs for "typical"
model facilities. Costs for actual facilities are likely to vary
considerably, both on low and high sides, due to individual circumstances
of the plants. However, the typical plant costs are considered to be
representative of the industry as a whole. Once again, to aggregate the
control costs for the broad categories of sources, the costs were
weighted based on the nationwide mix of sources, which is not representative
for all AQCRs.
The range of assumptions made in this analysis in regard to these
points appear to be reasonable in terms of estimating costs of control
measures for hydrocarbon control. While the intended purpose of this
analysis is an estimation of the range of national costs, the approach
allows the estimation of the number of AQCRs which will attain alternative
levels of the standard. This is only a secondary result which
was not considered in the original design of the study. Since the approach
of this study can possibly introduce more errors for an individual AQCR,
1-8
-------
the estimation of the number of AQCRs which will attain alternative
standard levels is particularly sensitive to the choice of assumptions.
The estimates of attainment status included in this analysis are only
approximate and are valid only for the assumptions and approach used;
thus, the estimates should be viewed with full recognition of the
limitations of this analysis. Any choice of different assumptions or
refinement of the assumptions to a fine-tuned basis was beyond the scope
of this study and the attainment numbers are presented only to indicate
the relative differences between alternative levels of the ozone standard.
Finally, this analysis does not include a rigorous economic impact
assessment on the affected industries or the impact of growth restrictions
on the economies of affected urbanized areas. An analysis of the economic
impact of the revised standard on the numerous industries affected is not
included as part of this analysis. However, EPA has conducted economic
impact studies for the major emission sources, though these constitute
only a portion of the total number of sources affected. Economic analyses
of all affected industries are underway and significant results are
expected within two to three years. Furthermore, the Clean Air Act does
not permit the consideration of costs and economic impact in the setting
of national ambient air quality standards. The incomplete character of
the economic analysis for all affected sources and areas does not mitigate
EPA's legal burden to promulgate the standard.
1.3 MAJOR FINDINGS
While the results of this analysis are sensitive to the assumptions
used, several conclusions are evident which indicate the relative
1-9
-------
differences between alternative levels of the standard and the implications
for control. This section summarizes the major conclusions of the analysis.
Table 1-1 summarizes the results.
1.3.1 Attainment Status for Major AQCRs
Many areas of the country will not be able to attain a .08 ppm
level based on the statistical form of the standard by 1987. Based on
the results from the linear rollback model, anywhere from 15 to 50 of the
major AQCRs in the country will require stationary control measures beyond
NSPS and RACT as well as additional transportation control measures in order
to approach the standard. The lower estimate reflects a low growth
scenario with extensive RACT measures placed on all significant stationary
source categories, including the solvent evaporation source categories
which have yet to be assessed in detail to determine their amenability
to control. On the other hand, the higher estimate is based on a higher
growth scenario without control of those solvent evaporation categories
which have not been assessed. For a .10 ppm standard level, seven to 20 AQCRs
may not achieve the standard by 1987 with the control measures identified
in this analysis. Finally, at a .12 ppm standard level, all but five AQCRs
could attain the standard under the low growth/high control scenario,
while all but 13 could attain under the less optimistic case.
The application of EKMA in the analysis does not lead to as optimistic
results since EKMA, with the assumptions used in this analysis, tends to
require significantly more control than rollback. The results from EKMA
could vary appreciably for any urbanized area depending primarily upon the
1-10
-------
CO
co
LU
<:
Q
CO
Z O
< rvi
o
CO
1 LU
co :r
o h-
o
u_
_J O
o
o: co
I i
s: LU
o >
O LU
LU CO
r- :r>
LU
CtL
u_ o
o u_
>- r^-
cc. co
co
cu
_a
rO
«%
rO
T3 4->
CU E --^
4J CU 00
ro E E
r^ *r *f"
4-> rO r
00 4-> i
LU 4-> <-
< CQ
O 4- *f>
o -
CU
Ol 00
E 4->
ra oo
C£. O
U
ft
.a
A
rO
S- 00
O CU
M- S-
00 00 00
4-> rO E
00 CU O
O ^ <-
O r
T3 i
l|_ 0) T-
O >- CQ
M-
CU !- fcO
CO4-J '
C E
ra CU
a: T3
^D r^
i. >
rO ra
oo TJ "O 00
CC E O) CU'--
O ra -I- S- O
cr4-> M- 3 cn
< CO T- 00
i*- E CU O
O en CU S
E -0 4->
S. T- *-H r- 3
CU E O O
Q T- .E i- '
3 4-> T- E
2 4-> 3 O
ra
CO 4->
_J CO
LO O LO O
tO LO «^ ^*
III 1
CD LO O LO
LO
CU
*r-
C"
O
03
r
O
^
4->
E
O
O
o
cu
cu
s-
cn
CO
13
^
00
cu
4-^
(O
S-
.c
4-1
3
O
S-
CD
<
E s:
LU
OO
E i.
O O
r- M-
4->
Q. CU
E
3 -Q
oo ra
00 i
ra !-
ra
<+- >
CD ^O
CU CO
CO S-
£Z O3
rO
S- 00
4->
O r
4-> 3
00
CU QJ
3 i-
-Q
CU
O ra
rO U
.a
r CU
r 00
O ra
5 f**>
S- >^
CO
-0 E
00
a> oo
C714J
E 00
ra O
C£ O
ra
*
1
ea;
Qi
Cj
O
00
r
O)
>
cu
r
o
3
4->
13
rO
00
O-
( )
I
CU
4->
r
E
^*
r
A
is
"s^,
1 t
A
co
o
c/}
2:
«\
Q.
C_)
>
:>
U_
cu
s-
ra
00
cu
s-
~^
00
ro
ai
E
n^
0
S-.
4->
E
0
o
T3
r
H-
l~
4->
c
cu
a
1 1
00
ro
r
r
cu
3
00
ra
00
00
0
O
CU
o
E
ra
cu
4J
c
r
rO
E
T3 C
E o
rO !
r
CDr
C !-
r- E
ra o
i- 0
CU LO
o
4-J
4-> 00
cu cu
C i-
rO
4-) d)
E C
0 O
U 4->
o
ro T3
CU
O T3
4-> C
C 3
r- O
i.
Cn
C 00
r- 4->
.^ 00
rO O
4-> O
F^ C
CO CU
en E
r C
r-
C rO
rO T-
3 Q.
C ra
c o
ro
T3
CU CU
S- N
rO -r-
r
00 ro
[ ' ^3
00 E
O C
C_) (O
(O
E "O
O CU
r r
4-> r
r- O
T3 S-
"O 4-^
rO E
O
CU O
-E
O
+-* -1-^
E
CU i-
00 CU
CU Q.
S-
Q.O
CU O
S- LO
00 *&
4«>
00 O
0 4->
O
o
T3 O
CU O
4-J r
rO *A
E
r- <4-
4-> O
00
ai cu
4_>
CU ra
^ E
1 t ,._
j . «f
00
a>
T3
CU 4-^
r- 00
>+- O
r- O
-4 >
C 00
CU T3
TD E
i 13
O
E JO
CU 1
CU S-
_a cu
3
4-> O
0 i
c
rO
CU
> >^
ro _a
TD
4-> CU
E -r-
CU r
E cu
E -r-
ra
4-> 3
40 E
ra
"O
r CU
rO S-
4-^ 'r
Q -3
4-> cr
cu
S. S-
o
*4 CT
0
00 T-
CU 4->
U O
rs 3
00 T3
ro CU
CU S-
E
E
r O
O -r-
S_ 00
4-J 00
C -
0 E
cj> cu
1-11
-------
ratio of ambient non-methane hydrocarbons to nitrogen oxides. Thus,
the results from EKMA would be more accurate if NMHC/NOx ratios were
available for every AQCR. Based on a typical nationwide ratio of 9.5:1,
EKMA predicts that less than 10 AQCRs could attain a .08 ppm standard by
1987. Even at .10 ppm, less than half of the AQCRs could attain the
standard even assuming low growth and high control, while at .12 ppm
about 40 AQCRs would still be in violation in 1987 without applying
further controls.
1.3.2 Implications for Further Control
This analysis confirms that most areas of the country will have to
apply extensive control measures in order to attain any of the reasonable
levels of the standard. Areas will have to place controls on new and
modified sources, aggresively apply RACT to existing sources, institute
effective inspection/maintenance programs for in-use vehicles, and implement
various degrees of measures to reduce vehicle miles traveled in the urbanized
areas. While higher levels of the standard will mean that some areas will
come into attainment automatically or will require less emission reduction,
most major urbanized areas will not be able to relax control efforts,
particularly in the near-term. Because ozone is such a pervasive and intense
air pollution problem, any modest relaxation of the standard will not result
in any significant changes to existing or planned control strategies for
these areas. The major urban areas will need to continue applying all
reasonably available control measures.
The results of this analysis also indicate the need for the identifi-
cation of miscellaneous solvent uses as well as miscellaneous industrial
1-12
-------
processes and the assessment of the applicability of control measures
for these sources. Current data indicate that these sources represent a
significant source of emissions which will need to be controlled if many
areas are to attain the alternative levels of the standard.
1.3.3 Costs of Control
As Table 1-1 indicates, the annual cost in 1987 of applying reasonable,
identified control measures will range from three to four billion dollars
for all levels of the standard under consideration. The cost difference
between alternative standards is not as great as might be expected for two
basic reasons. First, the vast majority of the total annual costs ($3.0
billion) result from the FMVCP and new source controls, which are assumed
to be the same regardless of the level of the standard. Secondly, the
RACT control costs do not differ greatly from alternative standards since
many areas will have to apply full RACT regardless of the level of the standard.
Since many AQCRs will not attain alternative levels of the standard by
applying the reasonably available control measures considered in this
analysis, additional control measures will be needed in some areas in
order to reduce emissions even further so that the standards can be attained.
These measures could include restrictive transportation control measures that
significantly reduce vehicle miles traveled in urban areas, control of
stationary sources for which RACT has not been identified or defined, tighter
controls on new and existing sources other than those achievable with RACT
and restrictions on growth. While the costs of such measures have not been
estimated, on the whole they are believed to be more costly than current measures.
1-13
-------
Even though the costs are not precisely known, it is still useful to
estimate the cost of attainment in some manner in order to better indicate
the cost differences between alternative standards. To do this, a cost
estimate of $1,000 per ton of emissions controlled is assumed for the
additional emissions reductions required for each AQCR to attain the
alternative standards. Table 1-1 summarizes the estimated costs of
total attainment for the alternative standards. The cost differential
between standard levels is greater, with the 1987 annual costs for the
alternative standards under consideration ranging from $3.5 to $6.5
billion using the results from the linear rollback model. Based on the
results from EKMA, total attainment costs could range up to $9.0 billion.
1-14
-------
2.0 BASELINE AIR QUALITY AND ALLOWABLE EMISSION LEVELS
This section examines the current ambient levels of air quality in
non-attainment areas in order to establish a basis for determining the
levels of emission control required to meet alternative levels of the
ozone standard. Based on these air quality levels, techniques for
establishing allowable emission levels in individual areas are discussed.
2.1 EXISTING AMBIENT AIR QUALITY LEVELS
Many areas of the country are experiencing levels of ozone and
oxidant concentrations well above the present standard. In this analysis
90 Air Quality Control Regions (AQCRs) are considered which have validated
data that indicate ambient concentrations at or above the present standard
of .08 ppm. While other areas of the country may currently violate the
standard, validated data were not available for additional AQCRs. None-
theless, the AQCRs considered contain all the major urbanized areas with
populations greater than 200,000.
The baseline ozone design values for the 90 AQCRs are based upon a
listing contained in Appendix A. These values reflect an approximation
of the ambient concentration that on the average will be exceeded once
per year in each AQCR. Using the analytical methods for the statistical
form of the ozone NAAQS, these values were obtained from an analysis of
data for the years 1975 to 1977. Using three years of data the design
value would be expected to fall between the third and fourth highest daily
maximum values. In selecting the design values, the fourth highest
daily value over a three year period was used, unless the difference
between the third and fourth highest values exceeded .01 ppm, in which
case the third and fourth highest values were averaged.
2-1
-------
These approximate values were derived only for analytical purposes
for use in this assessment. In the SIP revisions submitted to EPA,
the states will calculate the actual design values which will be used
for attainment determinations and for planning purposes. The values will
be calculated based on guidance contained in reference 1.
Figure 2-1 shows the distribution of the design air quality values
among the 90 AQCRs for which there are data. The adoption of alternative
levels of the standard in the range of .08 ppm to .14 ppm would have an
immediate effect of bringing many of the AQCRs into compliance, depending
on the level of the standard. For a .08 ppm standard, two AQCRs would
immediately come into compliance, while for standards at .10 ppm or .12 ppm
nine and 21 AQCRs, respectively, would come into compliance. Thirty-two
areas have design values above .16 ppm, while only 12 are above .20 ppm.
2.2 ALLOWABLE EMISSION LEVELS
For each alternative level of the standard, the maximum allowable
emission levels are calculated for each AQCR based on baseline air quality
levels discussed earlier, 1975 total emissions of non-methane hydrocarbons
(NMHC), the level of the alternative standard, background ozone levels of
zero and assumed relationships between hydrocarbon emissions and ambient
ozone concentrations. The maximum allowable emission levels indicate,
based on the parameters listed, the emissions which will permit attainment
and maintenance of the standard. The 1975 emission levels for each AQCR
were obtained from the NEDS Emission Summary Report (NE204), corrected to
exclude methane.
Two modeling approaches were used to estimate the allowable emission
levels. The first is the linear rollback model which assumes a proportional
relationship between hydrocarbon emissions and ambient ozone concentrations,
2-2
-------
o
CTi
00
CU
fO
C CO
cnct
r- O
oo cr
cu <:
Q
+->
14- C
o cu
c c
o <-
i- C
4J O
to z
CU
3
CU
CM
CU
o>
L
n
r^.
o
UD
:«=J-
'r
C\J
CM
CM
r
r~N
CM
1 ir> o in |o ILO
[CM CM r- 1,- 1
V
U3
c\
V
.
CM
v
CM
CM
v ]
o
CM
1
V
CO
V
to
r
vi
**
V
CM
*
V
o
r
i
V
CO
o
V
o
Q.
£ v
1 §
ro ->,
> r-*
-1 ^
oo ^:
CU 01
Q !-
CU
£= T3
0 E
NJ O
O 0
Ol
00
T3
CU
-(->
u
O)
Q.
X
QJ
f\
<
X
r
-o
c
O)
Q.
Q.
<
C
O
T3
QJ
oo
rO
DQ
fO
2-3
-------
with appropriate adjustment for natuural background levels of ozone. The
relationship is given by the following equation:
Allowable emissions = (Standard level minus background)
Current emissions(Current ambient concentration minus background
The alternative approach is the Empirical Kinetic Modeling Approach
(EKMA), which employs isopleths based on the results of smog chamber
experiments to relate various concentrations of NMHC and nitrogen oxides
(NOx) to resulting concentrations of ozone. In employing the model it
is necessary to make assumptions as to:
1. the prevailing 6-9 a.m. ratio of ambient NMHC to NOx.,
2. the relative degree to which NMHC and NOx emissions will be
controlled, and
3. ozone background levels.
While the NMHC/NOx ratios for various cities will vary widely, it is
beyond the scope of this analysis to determine the appropriate ratios
for every city. However, best estimates of typical 6-9 a.m. ratios, based
on an examination of data from a number of monitoring sites, indicate
2
a median ratio of 9.5:1, which is used in EKMA for purposes of this analysis.
In addition, it is assumed that NOx emissions remain constant between 1975
and 1987.
This analysis presents the results from both modeling techniques
since both are considered viable approaches for relating ambient ozone to
organic compounds and oxides of nitrogen. No judgement is made as to the
relative effectiveness or accuracy of the alternative approaches. For a
full discussion of the technical basis, uses and limitations of the approaches,
consult reference 2.
In essence, EKMA is a rather complex model that has been primarily
2-4
-------
validated against data from smog chambers, which represent a simplification
of the urban atmosphere. In addition, the absolute positions of the
standard isopleths represent an approximation since their position depends
upon a number of underlying assumptions concerning meteorological
2
conditions and emission patterns.
Rollback, on the other hand, is a rather simple approach that contains
the assumptions that ozone concentrations are proportional to NMHC emissions
and that the amount of organic emission controls needed to attain ambient
ozone levels is independent of the prevailing NMHC/NOx ratio. However,
smog chamber experiments suggest that the lower the ratio, the more
effective the hydrocarbon reduction is in reducing the maximum ozone
formed. Thus, at very low NMHC/NOx ratios, linear rollback may underestimate
the effectiveness of organic controls, while at high ratios estimates may
be overly optimistic. Nonetheless, under the range of NMHC/NOx ratios
believed to prevail in most U.S. cities, rollback appears to be useful
in serving as lower bound for estimates of hydrocarbon controls needed to
2
attain the ozone standard.
Application of the two modeling approaches gives somewhat different
results. Table 2-1 illustrates the comparative control levels required
2 3
for .08 ppm, .10 ppm, and .12 ppm standard levels utilizing both approaches. '
With the exception of cases having very low NMHC/NOx ratios, needed NMHC
reductions estimated with rollback are almost always less than those
obtained with the standard isopleth version of EKMA.
For the purposes of this analysis, the impact of both transport and
and natural background is ignored, so that the background is in essence
assumed to be zero. This approach is most feasible for an analysis of
this sort and is consistent with the alternatives included with guidance
2-5
-------
OO ^
LU O
OO II
LU I
Q <
Z 00
< I _l
LU LU
^ LU 5»
o s: LU
I
O I
O Q
OO LU
CC O
a. n:
O U-
o o
I
CM
(D
J3
rO
*
c
o
u
T3
O)
o:
o
IE
C
cu
u
s-
cu
o_
a
O)
fO
£
r
tf)
LU
CM
1!
T3
T3
c~
rO
4_>
oo
o
r
II
7-5
* '
Q
c
-(->
OO
CO
0
II
-a
i.
rO
-o
c~
fO
+->
00
C
cu 01
O (/)
f*j d)
O Q
g
\x
LU
O
TCI
JO
r
O
^
NX
LU
^
O
(O
J3
r^
O
o:
s
LU
(_>
fO
J3
^
0
C£
^- ^
a» s:
3 a.
^ o.
>
oooococncMLor-^r^cnocMCM
oo^j-uotototototorx«i"^tv~
OO^3"LOoOOLOO^-r^OCMLD
< CMOO^-^UOLOLntOtOtO
OLOocn^a-r^coO' CMOOOO^-
oo LO LO to to to r"** p^ r^» r^ r^ r^^
OI^-CT)CO«d-OLOCOCM«d-I^Crii
i cMoo^LOLOiototototor-^
COOOOOLOCTIi CMOO«d-«d-LOlOr-»
OooooOtoo^r^-crii ooLor~»
CMoo^i-LOLOtDtoiotoi^r^r-^r~»
OCM«d-tOCOOCM«*lOCOOCM<-
LO
0X3
S- CD O
4J 2: i-
c -^ cn
O C_> -M
o 3: o
2:
CM OO
a.
3
in
to
2-6
-------
issued by EPA to the states for determining hydrocarbon reductions required
4
to achieve the ozone standard. The preferred procedure for estimating
transport is to use upwind data collected on the same day as the ozone
design value. However, this approach requires suitable monitoring data,
which is beyond the scope of this analysis.
In the absence of such data, other procedures must be used. One
alternative is to assume that both future and present transport is
comprised of only natural background. This procedure is most appropriate
in circumstances where the urbanized area is isolated and not likely to
be subject to significant transport from other urban areas.
For urban areas that are likely to be impacted by transport from
other urban areas, another alternative is to ignore both transport and
natural background. Not considering background partially compensates
for not being able to directly consider transport from upwind urban
areas since background and transport affect control requirements in
opposite manners. Consideration of background increases control require-
ments whereas consideration of transport reduces control requirements.
This alternative has been tested and found to result, in most cases, in
calculated control requirements which are similar to those in which trans-
port and natural background are considered.
Using both rollback and EKMA, allowable emission levels were
calculated for each AQCR. These levels were then compared to projected
1987 baseline emission levels in order to determine the emission reduction
that needs to be effected so that all areas will attain the standard level
by 1987. These emission projections and control requirements are discussed
in the next section.
2-7
-------
2.3 REFERENCES FOR CHAPTER 2
Environmental Protection Agency, Guideline for Interpretation of
Ozone Air Quality Standards. OAQPS No. 1.2-108, Draft.
Environmental Protection Agency, Uses, Limitations and Technical Basis
of Procedures for Quantifying Relationships Between Photochemical
Qxidants and Precursors, EPA-450/2-77-021a, November 1977.
Freas, Warren P., Air Management Technology Branch, MDAD, OAQPS,
EPA, memorandum to Ken Lloyd entitled "Consideration of Transport
and Background in EKMA and Rollback Modeling," September 11, 1978.
Rhoades, Richard G., Control Programs Development Division, OAQPS,
EPA, memorandum, "Clarification of Attainment/Nonattainment Evaluation
Guidance," August 16, 1978.
2-8
-------
3.0 PROJECTION OF EMISSIONS IN 1987 AND REASONABLE LEVELS
OF CONTROL ACHIEVABLE
This chapter summarizes the emission projections and emission reductions
achievable by 1987 for the affected AQCRs. Reasonable ranges of growth rates
are applied to 1975 emissions in order to determine 1987 baseline emissions
without additional stationary source and mobile source controls other than
mobile source standards mandated by the Clean Air Act. Stationary source
controls in the form of new source standards and reasonably available control
technology (RACT) for existing sources as well as additional mobile source
controls in the form of inspection/maintenance programs and transportation
control measures are assessed and potential emission reductions quantified.
These projected emission reductions are then compared with the baseline 1987 emissions
and the calculated maximum allowable emission levels to determine if the available
controls result in attainment of the alternative levels of the standards. If
the standards are not attained, the additional emission reduction required to
reach the allowable emission levels is indicated.
For purposes of this study, RACT is broadly defined as technology
readily available for application in categories of sources which will lead
to adequate levels of control based solely on technical considerations.
Because of unique circumstances with individual sources, the reasonableness
of technology for particular sources will be determined by state and local
agencies based on technical guidance issued by EPA and depending upon
economic and energy feasibility.
Projections are made for each individual AQCR based on its emission
inventory and national estimates of growth rates and control efficiency.
3-1
-------
National totals of baseline emissions and emission reductions are obtained
by summing the totals for the individual AQCRs. A sensitivity analysis is
conducted which indicates the variability of the results depending on
assumptions concerning growth rates, control levels achievable, and the
relative proportion of emissions from source categories.
The assumptions and methodology used in this analysis were chosen
specifically to arrive at national emission and cost estimates. For planning
in any specific AQCR, more detailed analyses will be required. The procedures
to be followed for preparing detailed plans are presented in the document,
"Control Strategy Preparation Manual for Photochemical Oxidants.
3.1 MOTOR VEHICLE EMISSIONS IN 1987
Mobile sources include light-duty vehicles, other highway vehicles,
and non-highway vehicles (such as aircraft, barges and vessels, railroads
and earth moving equipment). The Federal Motor Vehicle Control Program
(FMVCP) for highway vehicles will significantly reduce emissions from these
sources (see Appendix B). This reduction is counterbalanced by an estimated
two to three percent growth per annum in total miles traveled for most areas
of the country, though VMT growth in some areas will range from one to six
2
percent. For light-duty vehicles, a total reduction of emissions of 60 to
66 percent is projected by 1987 taking into account both growth and tailpipe
controls. Similar reductions for other highway vehicles are expected to be
35 to 45 percent through the FMVCP. These factors are included in the
estimated baseline emission projections for 1987 in Table 3-1.
Even though the FMVCP will reduce total mobile source emissions by 40
to 50 percent, this will not be adequate to offset the growth of stationary
3-2
-------
eo
2:
O
h~H
oo
oo
K-*
s:
UJ
o
re
s
^
^L_
2: s-
o re
03 to CU
o: a: >>
< o ^--
o cr to
o < c
C£ O
O O 4-»
>- cn
rn 4-
Q£ O
UJ O
«^. U_ to
«=C C
rc r^ o
( 00 T-
uj cn p
1 T-
Z.-Z.-SL
O i i -
2:
Q
1 1 1
J
o
UJ
"3
O
o:
CU
p
CO
cu
J3
re
cu
JZ 0
-t-> s-
3 0
OO
to
c 3
O CU
p* "Z.
to
to -O
E re
UJ
ex.
(*« O
00 >
cn s
i U_
rM
0
4-*
C
O
O
o
VJ
c
o
CO
to
p*
E
LU
LO
r~*
cn
>,
C
o
cn
cu
4->
re
CJ
cu
o
S-
3
O
oo
cn cn «3- r-^ co CM oo
O «* to r~» «j- cn
CM O P O O O O
i i i i i i i
CM co oo uo r»« i P-.
00 ^f CO P ID * CO
r-^ O r^ CD 0 0 CD
to 10 CM co r «a- p
co r>x CM i r». vo to
in o r o O O O
to
cu
to
to
to en cu
CU CO
i (/J !- tO O
to u cu s- cu s>
0) -r- P 3 !- Q.
p jz o 4-> i.
O CU T- O 4->
p- . > jz c re to > > -P- 3 T3 S-
> re 4-> c c 4->
3 >> (/> re i " to
>> jz re 3 E 3
4-> cn 3 jo E TH
3 <- JZ E 3 C
o jz cn o re cu i i
1 - O O P
4-> i- JZ -P- O S-
JZ CU 1 E S- CU
cn jz c cu cu 4-> jz
p- 4-> O 3 JZ CU 4->
_l O 2: U- 0 Q- O
o
o
1
CO
LO
d
00
r-.
o
to
c
o
p~
4->
re
4->
oo
(U
u
P*
>
s-
O)
oo
cu
c
r
o
to
re
0
co
to
o
i
P
to
o
CM
CO
O
4J
S-
o
Q.
to
c
re
s-
h-
-o
c
re
cu
cn
re
s_
o
4->
oo
E
CU
p
o
s_
4->
CU
a.
*
CO
r_!
'
<*
CM
r
1
«*
-'
C
O
4->
re
S-
o
Q.
re
>
UJ
4->
C
cu
>
r
O
oo
P
re
r-
1.
4->
to
3
o
C
1 1
o
LO LO
to
e-f*
^^
«d- r^ cn p
CM «d-
i i
0 0
vo co
CO CM
^f co
to 53- cn cn
«d- co <* r^
LO O O CO
P'""
c
o
4->
re
s-
o
Q.
re
>
UJ
to
4-> 3
c a» o
CU 4-> CU
> CO C 1
p re re ct
0 3 i 1
00 P- O
"0 CU I
re -p- o
CU r to
i. a -r-
CC ^1
i/j
Q
UJ
-^r
*^-_
cu
JZ
4->
s-
4-
"0
CU
C
p
re
4->
J3
O
o;
CJ
cr
=£
JZ
u
re
cu
s_
0
4
to
c:
O
CO
to
E
CU
tl
v^_
O
E
3
to
cu
JZ
4->
to
4->
C
CU
to
cu
s_
a.
cu
S-
>>
S-
o
4->
c
cu
>
c
r-*
to
C
o
P
to
to
1
E
CU
to
p-
JZ
H-
re
CO
^
cn
p
S-
a>
JTJ
o
4->
t_>
o
*
s^.
«=J-
o
CM
LU
^
^
4->
5-
0
CL
a>
o:
>>
s_
re
E
E
3
OO
E
O
p
I/)
to
i
E
UJ
CM
CO
d
o
*r~
4"'
u
cu
00
c
r~
a
a>
to
to
3
O
to
-a
to
cu
4-5
re
l
>
JZ
4->
^
O
S-
cn
re
o
*r~
4-^
re
E
4-
O
CU
cn
E
re
i_
re
4-
O
E
O
p~
4->
re
o
i
p*"
Q.
Q.
re
E
O
i-
4
to
4->
P""
3
CO
a>
s_
-------
source emissions for most areas. Only in the unique circumstance where
an area with a high proportion of mobile source emissions is close to the
standard will the FMVCP alone permit attainment of the standard. For a
.10 ppm standard level, less than 10 of the 81 nonattainment AQCRs would
meet the standard in 1987 with the FMVCP alone while at .12 ppm l,ess than
15 of the 69 nonattainment AQCRs would attain the standard. Of course, fewer
AQCRs would attain by 1982 with FMVCP alone. Only two of the nonattainment
AQCRs could attain a .10 ppm standard and less than 10 additional AQCRs could
meet a .12 ppm level.
3.2 PROJECTION OF UNCONTROLLED EMISSIONS FROM STATIONARY SOURCES
In order to determine the total reduction in emissions required in 1987
for each AQCR to achieve the allowable emission level, growth in the current
emissions inventory has to be taken into account. In Table 3-1, the 1975
emissions inventory for the 90 AQCRs is indicated, broken down into three
mobile source categories and 10 stationary source categories,, The inventories
for each AQCR come from the NEDS Emission Summary Report (NE 204), current
as of October 1978. The breakdown of source categories represent the
segmentation of sources contained in the summary report.
For the 90 AQCRs as a whole, emissions from mobile source categories
comprise almost 40 percent of the total emissions, mobile source emissions for
individual AQCRs range from 15 to 70 percent of the total, though most AQCRs
fall in the range between 40 and 60 percent. These categories include
light-duty automobiles and trucks, heavy duty trucks, motorcycles, and
off-highway vehicles such as vessels and aircraft.
3-4
-------
Stationary sources, which include a wide variety of sources, make up
the remaining 40-60 percent of the total emissions. While petroleum industries,
chemical manufacturing, oil and gas marketing, and industrial surface coating
are significant sources, almost half of the stationary source emissions come
from a category called area source solvent evaporation. This category includes
such sources as solvent metal degreasing, drycleaning, and cutback asphalt
paving, but for the most part the actual sources are not well defined. Since
much uncertainty exists as to what is exactly included in this category and
since the category constitutes such a large proportion of emissions, this
report includes a sensitivity analysis whereby 50 percent of the area solvent
evaporation emissions are excluded when calculating emission reductions and
costs. The results of this analysis are discussed later.
Uncontrolled emissions are projected in 1987 by applying appropriate
growth rates for each source category. This analysis employs a range of
representative growth rates for the categories which take into account a
number of factors. The growth rates are based both on estimates of national
growth rates for the affected industries from various sources as well as the
2-5
range of growth rates that occur in specific AQCRs. In the analysis
individual growth rates are not used for each AQCR, rather the range of
national growth rates is used. Table 3-2 summarizes the range of growth
rates used in the analysis.
The growth rates for the industrial source categories are based on
estimates of production growth for the affected industries. For the industrial
solvent evaporation category, the growth rate reflects a composite rate
for all manufacturing categories since solvents are used in a wide range of
industries. Projected increases in consumption of gasoline and petroleum
3-5
-------
Table 3-2. RANGE OF GROWTH RATES FOR MOBILE AND STATIONARY SOURCES4
(Compounded Percent per Year)
LOW HIGH
Mobile Sources 2 3
Fuel Combustion 1 2
Chemical Manufacturing 4 6
Petroleum Industries 2 3
Other Industrial Processes 3 4
Gasoline Service Stations 2 3
Petroleum Storage and Transport 2 3
Industrial Solvent Evaporation 3 5
Area Solvent Evaporation 1 2.5
Solid Waste -2 -2
Miscellaneous 0 0
3-6
-------
products serves as the basis for the growth rates for gasoline service stations
and petroluem storage and transport. Finally, population growth rates are
used as the basis for emission projections for the area solvent evaporation
category.
3.3 EMISSION PROJECTIONS WITH CONTROLS ON NEW STATIONARY SOURCES
In non-attainment areas of the country, the recent Clean Air Act Amendments
require new and modified stationary sources to achieve the lowest achievable
emission reduction (LAER) before the source can be located in an area or
modified. Hence, this requirement will result in a significant reduction
in emissions from new and modified stationary sources. In order to estimate
the reduction in uncontrolled emissions that can be achieved, assumptions
as to the efficiency of new source controls have to be made. Since EPA
has not yet determined what constitutes LAER for stationary sources, this
analysis assumes that new and modified sources will have to achieve a minimum
level of control equal to RACT. The underlying assumptions in the determination
of RACT control levels will be discussed in the next section. The assumed new
source control level for applicable sources are summarized below:
o Chemical Manufacturing 80%
o Petroleum Industries 95%
o Gasoline Service Stations 90%
o Petroleum Storage and Transport 80%
o Industrial Solvent Evaporation 80%
o Area Source Solvent Evaporation 75%
Table 3-1 summarizes the baseline emissions in 1987 assuming statutory
motor vehicle controls and minimum controls on new and modified stationary
sources. These stationary source estimates include the normal replacement
3-7
-------
of existing sources (due to obsolescence) with new, modified, or reconstructed
sources that would be subject to control under LAER. As shown in the table,
even with growth in emission sources, new source controls on new and
modified sources in conjunction with motor vehicle controls will result
in an aggregate 20 to 30 percent reduction in hydrocarbon emission levels in
the 90 AQCRs by 1987.
3.4 REASONABLY AVAILABLE CONTROL MEASURES FOR EXISTING SOURCES
3.4.1 Estimates of RACT for Stationary Source Categories
For areas that fail to attain the standard with mobile source and new
and modified source control, existing sources will be required to install
reasonably available control technology (RACT) in order to attempt to
attain the standard. RACT will vary among industries and may well vary
among sources within an industry. RACT is defined as the lowest emission
limit that a particular source is capable of meeting by the application of
control technology that is reasonably available considering technological
and economicc feasibility. Since economic feasibility is basically source
specific, RACT may vary among sources in an industry. Since this study
considers broad categories of sources and does not consider the feasibility
of RACT on individual sources, RACT is broadly defined for purposes of
this study as technology available for application in categories of sources
which will lead to adequate levels of control based solely on technical
considerations. In actual practice, the reasonableness of technology for
particular sources will be determined by state and local agencies based on
technical guidance issued by EPA and depending upon economic and energy
feasibility.
3-8
-------
EPA is currently developing guideline documents for sources of VOC
which will assess the technology available for these sources. Based on
studies underway in support of these documents, estimates have been made
as to the efficiency of RACT for many industries which constitute major
sources of VOC. The industries and individual emission sources are
summarized in Table 3-3, which also presents the estimated efficiency
of RACT for these sources.
While these sources represent the primary sources of VOC, other
sources exist which have not been studied by EPA. These sources are
primarily in the solvent evaporation category. In many cases, the specific
sources have not been identified and control technology and efficiency
have not been assessed, though these sources will be analyzed in detail
in the future.
Since the emission inventory for each AQCR from NEDS is segmented
according to the 10 broad source categories, the average emission
reduction that could be achieved with the application of identified RACT
has to be determined for the categories. To do this, it is first necessary
to derive the weighted average of the efficiency of RACT for the sources
in each category for which RACT has been estimated. This weighted
average takes into account the relative contribution to current national
emissions and the efficiency of RACT for each source. After this calcu-
lation, the total emission reduction in each general source category is
determined by multiplying the weighted efficiency of RACT for applicable
sources by the percentage of emissions in the general source catetory
3-9
-------
Table 3-3. ASSUMED EFFICIENCY IDENTIFIED OF RACT FOR
STATIONARY SOURCE CATEGORIES?^
Source Category and Sources Included
1. Chemical Manufacturing
6 Organic Chemical Manufacturing Industry
- Process Emissions
- Fugitive Emissions
- Storage and Loading Emissions
- Secondary Emissions
o Pharmaceutical Industry
o Paint Manufacture
o Rubber Industry
Weighted Average of RACTb
Current Emissions Affected by RACTC
Emission Reduction in Source
Category Achievable with RACT
2. Petroleum Industry
o Gas and Crude Oil Production
o Petroleum Refining
- Vacuum Jets
- Waste Water Separators
- Miscellaneous Sources
- Process Unit Turnaround
o Natural Gas and Gasoline Plants
Weighted Average of RACTb
Current Emissions Affected by RACT0
Emission Reduction in Source Category
Achievable with RACT
3. Industrial Solvent Evaporation
Auto and Light Duty Truck Manufacturing
Flatwood Products
Paper Coating
.Fabric Coating
Wire Coating
Can Coating *
Metal Furniture
Industrial Machinery
Commercial Machinery
Coil Coating
Fabricated Metal Products
Efficiency of RACT (%)
90%
80%
90%
75%
95%
75%
79%
TBIT
100%
95%
91%
98%
96%
95%
95%
90%
90%
80%
85%
80%
80%
3-10
-------
Table 3-3. (continued)
Source Category and Sources Included3 Efficiency of RACT (%)
Large Appliances 85%
Small Appliances 80%
Weighted Average of RACTb 80%
Current Emissions Affected by RACTC __75%
Emission Reduction in Source Category
Achievable with RACT 60%
4. Area Source Solvent Evaporation
Dry Cleaning 65%
Vapor Degreasing 55%
Cold Cleaning 50%
Graphic Arts 80%
Adhesives 80%
Cutback Asphalt Paving
Weighted Average 75%
Current Emissions Affected by RACT 39%
Emission Reduction in Source Category
Achievable with RACT 30%
5. Petroleum Storage and Transport
Gasoline and Crude Oil Storage 75%
Gasoline Bulk Terminals 95%
Gasoline Bulk Plants 76%
Weighted Average 80%
Current Emissions Affected by RACT 100%
Emission- Reduction in Source Category
Achievable with RACT 80%
6. Gasoline Service Stations
Storage 90%
Refueling 90%
Weighted Average of RACTb 90%
Current Emissions Affected by RACTC 83%
Emission. Reduction in Source Category
Achievable with RACT 75%
7 . Fuel Combustion 0%
8 . Other Industrial Processes 0%
9. Solid Wasted " 0%
10. Miscellaneous 0%
Sources included are those for which screening studies or guideline documents
are being prepared. Many solvent evaporation sources and other industrial
processes are not inlcuded because the sources have not been identified or
control levels have not been defined.
Represents weighted average of RACT for sources listed based on current
national emissions for each category.
Represents the proportion of current national emissions for which RACT can
be.applied. Excludes emissions from source categories for which RACT control
levels have not been identified.(based on reference 7) and the residual
emissions from sources which have already controlled to the RACT level.
RACT has not been identified for these source categories.
3-11
-------
which would be affected by RACT. The emissions affected by RACT include
only those sources for which RACT has been identified and which have not
already controlled emissions to the RACT level. Thus, this excludes
sources not covered by RACT and those already controlled since these
7 8
latter sources already achieve RACT control.
As Table 3-3 indicates, emissions from existing sources can be
reduced by 65 percent from the chemical manufacturing industry, 90
percent from the petroleum industry, 80 percent from petroleum storage
and transport sources, 75 percent from gasoline service stations, and 60
percent from industrial surface coating operations. However, controls on
identified area source solvent evaporation sources will reduce emissions
from the total category by only 30 percent, since over half of the emission
inventory results from sources which have not been identified and for which
control levels have not been estimated.
While the control of identified sources serves as the basis for this
analysis, a sensitivity analysis is included which assumes a moderate level
of control on the emissions from the unidentified solvent sources and other
industrial processes. This assumed level of control amounts to 65 percent
of all emissions in each category. Throughout the remainder of the document,
the control of identified sources only is termed "identified" RACT while the
assumed control of the entire solvent evaporation category as well as
other industrial processes is termed "advanced" RACT. For the "advanced"
RACT case, the control of emissions from the other source categories remains
the same as under "identified" RACT.
3-12
-------
3.4.2 Estimates of RACT for Mobile Source Categories
There are measures in addition to tailpipe standards which are
available to reduce emissions from mobile sources. Reasonably available
measures for mobile vehicles include inspection/maintenance programs as
well as traffic reduction measures such as transit improvements, parking
management, and traffic management. The emission reduction potential for
each of these measures will vary among areas dependent on the applicability
of control measures, the relative contribution of each mobile source category,
existing traffic patterns, and other factors.
In this study, it is anticipated that an inspection/maintenance
program for light-duty autos and trucks could achieve an average reduction
of 30 percent of 1987 emissions from the LDV category, with the FMVCP as
910
baseline, though the effectiveness could range from 20 to 50 percent. '
In addition,, for purposes of this study, traffic reduction plans are
estimated to reduce emissions by two percent compared to the projected
baseline in 1987. Hence, total RACT for light duty vehicles is estimated
to represent a 32 percent reduction. Similarly, the emissions from other
highway vehicles are assumed to be reduced an additional two percent through
traffic reduction measures.
As was the case with stationary sources, an "advanced" RACT case is
assumed whereby emissions of highway vehicles are reduced by a total of
three percent due to traffic control measures.
3.5 TOTAL EMISSION REDUCTIONS AND ADDITIONAL REDUCTION REQUIRED
Table 3-4 summarizes the total emission reduction that can be achieved
in the 90 non-attainment AQCRs by 1987. As indicated previously, motor
vehicle tailpipe controls in conjunction with new source controls can reduce
3-13
-------
CO
LU Q
oo
cu
.a
re
^~
r
Q.
Q.
CM
T3 i
i-
ra
-o
re
oo
4-
o
r O
CU r
cu
CO
o
o> cc
^ **.
00 **
-o
c
re
D-
O
^>
^~
u_
-E i
4-> O
r- S-
3 -M
10 to O
C CO
o o
r- -i- CU
to to o
tO tO 5-
'E E o
LU LU OO
in r-. 3
l~- CO CU
CT> OS ^
1 1
r
to
CM
oo
»
CM
r
"^"
^F
|Z
fs^
V£>
O
1
0
a:
3 "0
cu cu
2: -1-
4-
CX -M
o c
> cu
s: TJ
U_ t <
^r "o
i Ji ^
r- re
^
r"
to O
C S-
O -!->
r- C
tO O
CO 0
i cu
LU U
i.
r^^ 3
CO O
C7> OO
i
CM
CO
fv.
px.
o
<^J-
^J-
en
c
t
c~
r*
re cu
-4-> O
4-> S-
I
cu 2: o
ILf^ 1 1 L ^£
£ Qi
^ -^*
^ -p -a
r- C
cu 2 re
-t_>
re T3 «
E S- r-
r- re o
x -a s-
o c -M
s- re c
CX4-> O
Q.OO O
f-
-!-> tO
o a:
3 O
T3 Cf
cu <
ae
o
c en
o
i r
to i T3
to re &.
r- re
E s- -a
LU O C
M- re
r~ -f-^
re -o oo
c cu
o s- c
p~ *^ "r
4-^ 3 re
r- CT-M
T3 CU 4->
-a o: <
<=c
CM
1
OO OO
1 OO
CU OO Q
r 1 i 1
r LAJ
jQ CU "SI
re r
i .a E
re o
0 E H- S-
S- O 4
CU S- £
N 4- O >>
i- 5-
tn to 4- O
i CU 4->
re 4-> i -^ c
3 re o o cu
cr s- ec re >
cu a: .a c
o 4-» ~a i
c 3 cu o to
3 O T- &. C
0 S- 4- O
S_ CO-i- S- <-
en 4-> re to
^ ^: c cu to
0 0) CU C -r-
re -p- T3 -i- E
ca re > i i LU
CM OO * LD
to
C
o
r
4^J
Q.
E
3
to
to
-------
1975 emissions by around 23 percent, taking into account growth. The sum of
emissions in 1987 in the 90 AQCRs can be reduced by an additional 20 to
25 percent with the application of RACT.
This reduction with RACT does not take into account the control of
replaced sources that would also be controlled to the LAER level (see
Section 3.3). Considering the control of replaced sources through new
source controls, existing stationary source emissions can be reduced from
1975 to 1987 by an average of 40 to 45 percent with identified RACT and 60 to
65 percent with advanced RACT. Of course, the reduction in individual AQCRs
will vary depending on the mix of sources and amount of control needed in
the area.
Even with full application of RACT to identified source categories, many
AQCRs will not be able to attain alternative levels of the standard. Table
3-4 summarizes the attainment status for the AQCRs for the alternative standard
levels. As the table indicates, as the standard level increases more AQCRs will
come into attainment. Less than 20 AQCRs will be unable to attain standard
levels of .10 ppm or greater with identified measures.
Table 3-4 also indicates the total additional emission reductions needed
for the non-attaining AQCRs to eventually attain the alternative standard levels.
This emission will have to come from currently undefined control measures which
will likely be localized in nature.
3.6 VARIATION OF RESULTS
In order to determine the variability of the results with differing
input assumptions, a sensitivity analysis was conducted which varied the
assumptions regarding growth rates, level of control achievable, proportion
of solvent evaporation emissions, and the modeling technique. The results
of this analysis are presented in Table 3-5.
3-15
-------
CO
^
o
=c
CD
00
I
O
>-
LT)
<=C
E
LU
M- OJ
O OJ
s_
C JC
r- Ul
-P t/> C
fO =, E
C O -
-Q > O
E
o c
on c
-o v
(J T-
rs E
OJ
ce
oo
1
o
*x
Qi
O
OJ
U
§
>
X3
«£
1/1
O)
_C
O
3:
o
puoi ^.onpsy
I CUO L 3. L pptf
Rill 111 "D1 T W
/. cvi'^hui
^aJUV
SUOLSSIUIJ
Z86l"
i PUOII nnw
filj 1 III PI 1H
Z s' j.
aJUs
^ s J
Z861
guononpay
1 PUO LT 1 DDW
L . +. PHu
fiU 1 U 1 PT TW
Z sjjohu
""
SUOLSSIU13
[euoiitppv
BuLULB^t|.y
syDbv
SUOISSLUIJ
£861
nUOLlDnpBH
[EUQUlppy
BlUUlBnv
Z _' V
aJUu
gSUOLSSllU^
Z861
*0
TD r £
C QJ Q.
CO
LO
^.
OJ
0
CD
0
in
CO
in
^
CO
0
CM
ID
CO
O
Ol
CO
O
0)
CO
ai
in
r~
ID
1
CO
ai
CO
n3 > Q- O
«*
CO
o
o
in
LO
o
CO
CO
o
CO
CO
,
o
CD
cn
CM
O
O
CO
o
ID
O
Ol
O
'
^
^
CO
ID
O
0
10
CM
LO
O
O
*~
CO
CD
in
CO
i
01
CO
en
o
|5
CM
CM
O
CM
in
o
CO
00
CM
Ol
ID
O
O
CO
LO
Csj
CO
CO
-'
CO
,^_
in
"~
CM
d
CO
CO
LO
Ol
ID
O
CM
CO
o
CD
O
CO
O
ID
CO
ID
OJ
CO
in
0
in
CO
^r
^
o
E
OJ
i.
IO
o
01
^ to
O <1J
H- O
O3 i- (/I
- O- C
S- O. O
Q. rt> -i-
O r
s- -a i
Q- C -
CL « E
O r C i-
r O O ro
4J i- CJ "O
CL+-> C
E C QJ ro
Z3 O L) -t->
rO O) O C
O ut-r-
i- i- ro
O 3 3 *->
<+- O OJ 4->
ra
2
O
i-
Dl
3
O
o
i.
CM
ftJ Q_ U_ CJ
^i o cr
(_) > ^I et
^ -P
QJ U. -r- O
QJ 3 CT*
_
O =C
H- C£ US
01
!-> = S-
O.-O (O
O) Ol
n =
t/1 _
_
-4-) -o
^: ^-^ c s-
uj &. ni o
fO 4-> 4-
4- QJ I/)
o >> -o
en aj
S- l/l ro OJ
o c +-> s-
4- O -t->
000
s- x x -a
M- E QJ
i/l O
03 !-
01 0)
-------
3.6.1 Growth Rates
Growth rates impact the results of the analysis since a high level of
growth for new sources will mean that existing sources will have to reduce
emissions to a greater extent to offset the growth in emissions. This
analysis employs a range of national growth rates which for the most part
emcompass the growth rate projections for individual AQCRs and urbanized
areas. The degree of sophistication introduced by using individual growth
rates for each AQCR considered in the analysis was not warranted for the
level of this analysis, thus national growth rates were used. The base
case in the analysis utilizes the high end of the range of the growth
estimates for each source category in order to be on the conservative side
of emission projections and emission reductions required.
As can be seen from Table 3-5, use of the lower growth rates for the
source categories affects the attainment status for only a few AQCRs. While
at a .08 ppm standard 17 additional AQCRs are projected to attain the
standard based on the low growth rates, the difference is only four and
three AQCRs at .10 ppm and .12 ppm respectively. For these seven affected
AQCRs, their individual growth rates from OBERS fall within the range of
national growth rates used in this analysis, except in a few instances.
The population growth estimates, which serve as the basis for projected
area source solvent emissions, range from 1.0 to 1.8 percent for these areas,
while total manufacturing growth estimates range from 3.4 to 4.3 percent.
Likewise, annual VMT growth projections range from 1.4 to 4.6 percent, with
o
four of the areas having rates greater than 2.9 percent.
3-17
-------
3.6.2 Control Levels Achievable
While RACT has been identified for a significant number of sources of
VOC, there is a sizeable portion of the emissions inventory for which control
measures have not been identified. These sources are primarily in the
solvent evaporation category, where even specific sources have yet to be
fully identified. It is not known what control levels can be achieved from
these sources or even if many of the sources are amenable to control in the
first place.
Nonetheless, a sensitivity analysis is performed where the emissions
from the solvent evaporation categories as well as other industrial processes
are reduced by 65 percent through the application of identified and
"advanced" RACT by 1987. This assumption makes a big difference for the
attainment picture at a standard level of .08 ppm since nearly 60 percent
more AQCRs can attain the standard. The differences are not so great as
the standard level increases since the increased control makes a difference
of ten and six AQCRs at .10 ppm and .12 ppm respectively.
3.6.3 Reduced Solvent Emissions
Since there is much uncertainty surrounding the composition of much
of the area source solvent evaporation category, a sensitivity analysis was
conducted whereby 50 percent of the area source solvent evaporation emissions
was deleted from the inventories of each AQCR. As Table 3-5 shows, such a
reduction results in little difference in the attainment picture. Only at
a .08 ppm standard do additional AQCRs come into attainment. The reason for
no discernable change is that the area source solvent evaporation category
still grossly outweighs all other categories even when 50 percent of the
3-18
-------
emissions are deleted. Since linear rollback is based on proportional instead
of absolute relationships, little difference could be detected in the reduction
required for areas to meet the standard unless the category was deleted entirely.
3.6.4 Low Growth Rates, Advanced RACT, and Reduced Solvent Emissions
The most optimistic case from the standpoint of attainment of the
standard occurs with low growth rates, advanced RACT, and reduced area source
solvent emissions. With this case, nearly all AQCRs could attain all standard
levels by 1987. Even at .08 ppm, only 15 AQCRs could not attain the standard,
while only seven and five areas could not attain levels of .10 ppm and .12 ppm
respectively. Furthermore, the additional emission reduction needed for
attainment of the alternative standards is reduced by nearly two-thirds.
3.6.5 Empirical Kinetic Modeling Approach (EKMA)
An alternative modeling approach to linear rollback is EKMA, which was
discussed in Chapter 2. A sensitivity analysis was conducted substituting
EKMA for rollback in the base case. As can be seen from Table 3-5, the use
of EKMA with only general and not area-specific assumptions results in
significantly less optimistic results. Only seven AQCRs are projected by this
approach to attain a .08 ppm standard, while 40 and 51 areas are projected to
attain .10 ppm and .12 ppm levels, respectively. The emission reductions needed
to meet the standard levels are significantly greater since the allowable
emission levels are around 25 percent lower.
3-19
-------
3.7 REFERENCES FOR CHAPTER 3.0
1. U.S. Environmental Protection Agency, Control Strategy Preparation Manual
for Photochemical Oxidants. OAQPS Guideline Series No. 1.2-047,
January 1977.
2. McCurdy, Thomas, Strategies and Air Standards Division (SASD),
OAQPS, EPA, memorandum, "FHWA Projected Increases in Highway Mileage
and VMT." April 6, 1978.
3. U.S. Water Resources Council, 1972 QBERS Projections: Economic
Activity in the U.S., Volume 1, April 1974.
4. U.S. Department of Commerce, Bureau of Economic Analysis, Projections
of Economic Activity for Air Quality Control Regions, August 1973.
5. Data Resources, Inc., (Lexington, Mass.) U.S. Long-Term Review,
Winter 1978.
6. U.S. Environmental Protection Agency, Guidelines for Air Quality
Maintenance Planning and Analysis, Volume 7: Projecting County Emissions,
second edition, January 1975, EPA-450/4-74-008.
7. Walsh, Robert T., Emission Standards and Engineering Division, OAQPS,
EPA, Memorandum to Bruce Jordan, "Estimated Reductions in Volatile
Organic Compound Emissions Which Could be Effected at Stationary
Sources," August 12, 1977.
8. Walsh, Robert T., ESED, OAQPS, EPA, Memorandum, "Current VOC Emission
Inventory," November 30, 1977.
9. U.S. Environmental Protection Agency, Mobile Source Emission Factors,
EPA-400/9-78-005, March 1978.
10. U.S. Environmental Protection Agency, Appendix N to 40 CFR Part 51:
Emission Reductions Achievable Through Inspection and Maintenance of
Light Duty Vehicles, Motorcycles, and Light and Heavy Duty Trucks,
May 1977.
11. U.S. Environmental Protection Agency, Air Quality Impacts of Transit
Improvement, Preferential Lane, and Carpool/Vanpool Programs.
EPA-400/2-78-002a, March 1978.
3-20
-------
4.0 NATIONAL COSTS OF CONTROL FOR ALTERNATIVE LEVELS OF THE STANDARD
In the previous sections of this report, estimates were made of the
amount of non-methane hydrocarbon control needed to meet alternative
levels of the ozone standard and control levels achievable with available
control measures. A comparison will now be made between alternative
levels in terms of the cost of applying available controls for various
mobile and stationary sources. The average cost per ton of emissions
controlled will be established for general categories of sources and
combined with estimates of emission reductions achievable in order to
obtain total costs of control. The costs presented represent the costs
incurred in the 90 non-attainment AQCRs under consideration in the previous
section.
The costs presented in this section represent annual costs of control
which will be incurred in 1987, when it is assumed that the control measures
will be applied fully to the applicable sources. The annual costs include
operating, maintenance and administrative costs as well as annualized
capital charges that take into account depreciation and interest costs.
No estimates are made of the total capital costs that will result from
application of the control measures outlined. In addition, secondary
costs and benefits associated with control measures are not estimated.
4.1 COSTS FOR THE FEDERAL MOTOR VEHICLE CONTROL PROGRAM
The Federal Motor Vehicle Control Program (FMVCP) includes tailpipe
emission standards for hydrocarbons (HC), carbon monoxide (CO), and nitrogen
4-1
-------
oxides (NOx) from motor vehicles. Appendix D presents an analysis of the
costs of the program for light duty autos, light duty trucks, heavy duty
trucks, motorcycles, and aircraft. Based on the information in the
Appendix, Table 4-1 summarizes the total annual costs of the FMVCP in
2
1987. In that year, the total annual cost of tailpipe controls currently
mandated will amount to approximately $5.8 billion.
A judgement has to be made concerning the fraction of the total cost
which should be allocated to HC control as opposed to CO and NOx control.
Because of the overlap in control among the various system components, it
is exceedingly difficult to allowcate costs precisely among pollutants.
In this analysis, the costs are allocated equally among all pollutants which
1 2
are affected by control systems. ' In some instances, the total costs for
various components are divided by three when all three pollutants are affected,
while in other cases the costs are divided by two since only CO and HC are
affected (See Appendix D for details).
Taking into account this allocation approach, the costs for HC control
under the FMVCP are presented in Table 4-1. The total annual costs in 1987
amount to approximately $2.2 billion, of which almost 75 percent is due to
light duty vehicles. The cost of the FMVCP will be the same irrespective
of the level of the standard.
4.2 COSTS FOR NEW AND MODIFIED STATIONARY SOURCES
Costs for new and modified stationary sources result from the costs
of controls on new sources as well as replaced existing sources. To
estimate new and modified source costs, the emission reductions from these
sources, by source category, were obtained from the previous chapter.
4-2
-------
TABLE 4-1. FMVCP ANNUAL COST SUMMARY >
($ BILLIONS)
Total Cost Total Costs
for Allocated tof
All Pollutants Hydrocarbons
Light Duty Autos
Hardware 2.9
Maintenance (0.8)
Fuel Consumption (2.2)
Unleaded Fuel Differential 1.8
Evaporative Emission Control 0.2
High Altitude Control 0.4
Subtotal ~O~
Light Duty Trucks
Hardware 1.1 0.4
Maintenance 0 0
Fuel Consumption 0 0
Unleaded Fuel Differential 0.9 0.3
Evaporative Emission Control 0.1 0.1
High Altutude Control 0.2 0.1
Subtotal 2.3 ~O~
Heavy Duty Trucks
Hardware, Maintenance, Fuel
Consumption, Unleaded Fuel
Differential and Evaporative
Emission Control 1.1 0.5
Motorcycles
Hardware 0.1+ 0.1
Aircraft
Hardware <0.1
-------
Next, the emission reductions were multiplied by an average cost per ton
of pollutant controlled for each source category. This average cost per
ton, the derivation of which will be explained in the next section, was
obtained from Table 4-2, which summarizes the average cost of RACT for
the source categories. While the costs in Table 4-2 pertain to control
of existing sources, these costs were nonetheless used as estimates for
new sources in the absence of similar cost data for new sources. It is
likely that costs for new sources will be somewhat less than costs for
existing sources since controls can be incorporated into new plant con-
struction without retrofit considerations.
Based on this approach, the total national costs of new source control
amount to annual costs of over $800 million. This total represents the
sum of the costs for all 90 AQCRs based on the high range of growth rates
discussed previously. It is assumed that the same level of new source
control will be required regardless of the level of the standard.
4.3 COSTS OF APPLYING RACT FOR STATIONARY SOURCES
4.3.1 Cost-effectiveness of Reasonably Available Control Measures
Appendix C presents a discussion of many of the stationary sources of
hydrocarbons and the control techniques that are available. An analysis
of the costs of these techniques for each source of emissions has been made
and summary is also presented in Appendix C. These costs are current as
of July 1977.
4-4
-------
In order to estimate the average cost per ton of hydrocarbons
controlled for the general source categories for which emission projections
were made in Chapter 3, the individual sources contained in Appendix C
were segmented into the source categories in a manner consistent with
the categorization contained in Table 3-3 of the previous chapter. The
average cost per ton controlled for each source category represents a
weighted average based on the cost per ton and the net reduction in
emissions avhievable with RACT for each source. The average costs for
the source categories are summarized in Table 4-2.
4.3.2 Analytical Methodology for Estimating Control Levels Needed
For each alternative standard, maximum allowable 1987 emission levels
are calculated for each AQCR based on 1975-1977 expected air quality levels,
1975 total emissions of non-methane hydrocarbons, the level of the
alternative standard and assumed relationships between NMHC emissions
and ambient ozone concentrations using both the linear rollback model and
the Empirical Kinetic Modeling Approach (EKMA). The allowable emissions
for each AQCR are then compared to the projected 1987 emission levels,
based on the national estimates of growth and retirement rates and new
source emission levels for each source category over the 12-year interval.
If projected emissions are greater than allowable emissions, existing
stationary source emissions (i.e., those that have not been replaced)
are reduced appropriately due to RACT and RACT costs are computed. The
costs are obtained by multiplying the average cost per ton for RACT for
each general source category (see Table 4-2) by the emission reduction
due to RACT for each category.
4-5
-------
TABLE 4-2. AVERAGE RACT CONTROL COSTS
Control Costs ($/Ton)
SOURCE
Stationary Sources
Fuel Combustion
Chemical Manufacturing
Petroleum Industries
Other Industrial Processes
Gasoline Service Stations
Petroleum Storage and Transport
Industrial Solvent Evaporation
Area Source Solvent Evaporation
Solid Waste
Miscellaneous
Transportation Sources
Light Duty Vehicles
Inspection/Maintenance
Transportation Control Measures
Other Highway Vehicles
Non-Highway Vehicles
Identified RACT
Alone
100
275
160
200
100
635
1000
1000
Identified RACT
plus
Advanced RACT
100
500
400
160
200
300b
635
1000
1000
Costs are offset by product recovery
b Based on a cost of $100/ton for identified sources and $500/ton for additional
sources.
4-6
-------
If controlled 1987 emissions are less than allowable emissions for
alternative standards, the analysis assumes that source categories are
partially decontrolled in a manner which minimizes total control costs.
This involves applying controls to the source categories in order of the
most cost-effective first.
After the costs for each individual AQCR have been determined, the
AQCR costs are then summed in order to arrive at national costs. The
costs presented are only for those existing sources which have not been
replaced by 1987 and does not include the costs of controls on replaced
sources, which are included in the new source costs presented earlier.
If additional emission reductions are needed in any of the AQCRs,
RACT for mobile sources is then applied. Stationary source controls are
considered first since on the whole they are more cost-effective than
mobile source measures. While this may not be true for all individual
stationary sources, the average cost-effectiveness of controls for the
broad stationary source categories is more attractive than the average
costs of broad mobile source measures. The cost of mobile source measures
will be discussed in a later section.
4.4 COSTS OF APPLYING RACT FOR MOBILE SOURCES
4.4.1 Cost-effectiveness of Reasonably Available Control Measures
Reasonably available control measures for mobile sources include
inspection/maintenance programs as well as traffic reduction measures such
as transit improvements, parking management and traffic management. Appendix
E presents a discussion of the cost and effectiveness of I/M programs,
which have a cost of around $635 per ton of HC controlled.
4-7
-------
The cost of transportation reduction measures are extremely variable
and will depend on the local circumstances. Most traffic reduction measures
are interrelated and require a coordinated program in order to be an effective
means for reducing hydrocarbon emissions. The maximum emission reductions
from transportation measures will result from coordinated measures designed
to discourage low occupancy auto use and to encourage transit and carpool use.
It is expected that reasonable aspects of these measures, such as carpool
programs with preferential highway lanes for high occupancy vehicles, can
result in a two percent reduction in light duty vehicle emissions at an
approximate cost of $1000 per ton of HC reduced, though this cost could vary
3
from area to area.
4.4.2 Analytical Methodology for Determining Total Cost of Mobile Source RACT
Areas which do not attain the standard by 1982 with FMVCP, new source
controls, and identified stationary source RACT will be required to implement
I/M programs in urban areas, as well as transportation control measures
and additional stationary source controls in order to attain the standard
by 1987. To determine the cost and extent of I/M programs, an estimate
was made as to the number of areas which would not attain the alternative
levels of the standard by 1982 without additional controls past identified
stationary source RACT. The number of vehicles in the urbanized areas which
would be affected by an I/M program were then estimated. Lacking precise
information on the numbers of vehicles in the areas, a rough estimate was
made based on population in the area, the national average of vehicles per
household, and the average number of people per household. Based on data
4-8
-------
from several sources, it is estimated that in urbanized areas there are
456
450 to 500 vehicles per 1000 population. ' ' This was then compared to
the population in the areas to determine the number of vehicles affected.
Table 4-3 summarizes the estimated number of areas and vehicles which would
be subject to I/M programs.
In order to determine the cost of I/M, the number of vehicles affected
is multiplied by an average cost per car of $4.12, which was derived in
Appendix E.
In areas that need additional control of mobile sources beyond I/M,
emission reductions up to two percent from transportation measures are then
included in the emission projections. The tons of hydrocarbons removed are
multiplied by the estimated cost of $1000 per ton in order to determine
the total cost of the measures.
4.5 TOTAL COST OF APPLYING RACT FOR ALTERNATIVE STANDARDS
4.5.1 Base Case Costs
The total annualized costs in 1987 of applying RACT in the 90 AQCRs
are presented in Table 4-4. The costs range from $675 million for a .08 ppm
standard to $305 million for a .14 ppm standard. Approximately 68 percent of
the total cost of RACT in the 90 AQCRs is due to stationary source controls
while about 30% results from I&M. The remaining costs are due to other
mobile source controls.
The costs differ for alternative levels of the standard since as the
level becomes less stringent more areas can attain the standard without
applying full RACT. The difference in costs between alternative standards
is not as great as might be expected because many of the larger AQCRs with
4-9
-------
TABLE 4-3. ESTIMATED NUMBER OF AQCRs AND VEHICLES AFFECTED
BY I/M PROGRAMS 4,5,6,7
Standard
Level
Approximate
Number of AQCRs
Affected
Estimated Number of
Vehicles Affected
(Millions)
.08
.10
.12
.14
70
45
20
15
43
37
25
19
- 48
- 41
- 28
- 21
4-10
-------
to oo
o; ^
o o
to
r O
LO O
co en
LO CO
LO
o
CO
LO LO o
CM i l
LO
C_> LU
< 00
CtL <
O
cs
z: LU
t i oo
>- <
_> CO
CL,
Q.
< I
C£
O OO
U -- 1
UJ
r-- >
CO LU
CTt _J
i-. <
Q
OO ~ZL
K- <
00 H-
o oo
O I-H
o: h-
o o:
o LU
LU
_l
CO
c u
o c
r- tO
4-> C
o at
a> 4->
D. S=
(/» -r-
C "3
O O LO
O f-» '
CM i i
E
O
4-> cC
oo OL
o o>
OO
c
o
o o
LO LO
«a- co
LO
CM
U- -0
0 S-
T3
T3
O) C
> «
0) -!->
_J OO
E
CL
CL
00
O
E
CL
Q.
O
i
E
Q.
CL
CM
r""
e
CL
CL
^j-
i
«
to
O)
tj
f ^
Ol
co
|-> Q)
O
i. O
0) S.
(T3
oo
o
i/l
O
S-
3 fO O U O)
CT S- < (O >
CD C£ J3 C
J= r -i-
-o +-> -a t
c 2 a> o w
3 O T- S- C
O i- 4- O
i. CD-I- S_ -i-
cn 4-> ro ifl
j^: .c c OJ tn
O CD (D C "
TO -r- -a T- E
CQ n: > i _i LU
CM
LO
c
o
4-11
-------
S-
o
OJ
LO
c^
s:
\s
LU
O LO LO LO
oo r^ o o
^"s" ^O t-f*> Lf)
*
00
2:
O to
i i c: 3
1 00
_J T- -I-
i I 4-3 > «^~
2! (O CD CD
C i. CD
-I Q. S_
N Q c*~
E 1
OO O 4-
z. o o
O LO O LO
<£> *t CFl CQ
r-^ LO co CM
O
1 1
1
Q-
s:
oo
oo
E
ZL CD C O
1 1 (J CO T-
>- 3 > to
rv -a i to
< CU O T-
> o: oo E
LU
o
LO O LO LO
o co LO r~^
IO c£
^r^ "X^J
<: =c
LO O O O
*d- LO co «3-
r- CO LO ^J-
r
_i
cC
1
O
r"""
[_|_
0
>- r
1 .c to
I t ^ 4J CD
> O S 4->
I I I O K3
H- S_ C£.
1-1 CD
OO
O O O O
O ID <* LO
to «d- oo CM
UJ
oo
C
E
O
S-
4-
CD
C\J O
i E
CO CD
$ ' "*
CD CO
i CD c:
-Q S- O
ft3 (T3 *t~'
(_ 4->
tO CL
E C E
0 0 =!
S- ! tO
4- to f)
to fO
to -p-
CD E i-
+-> CU O
(T3 4-
i- 4-*
C C\l
.c: cu
i ^ ^> ^_
s ai
O O 4->
i. to CL
CT> to
CD c"-
S 0 C_i
O 5-
r- Z3 ai
H- o ai
s- o to f i
O =C -
4- Qi fO
CD <.
Q--O fO M
CD CD U.I
U 0 4-
X C O 4-
O) "3 O
> "&^
^- -o o ai
1 ro LO f)
CD i- 13 CD S-.
r O J= > 0
f*^ (4 4_> o 4
rO r^>
h~~ | * 4_) fO 4-1
Q. Q. CL
E OJ Cl) CO CU
O O O U
i- x x -a x
4- cu a> c: ai
ro
to to to to
c: c: E c:
O O O CM O
1J3 '4J '4J «'^Z.
CL CL CL- CL
E E E E
to to to o to
to to to i/i
o
CD CD cu T Q.I
tO tO tO 4-> tO
ro ro ro ro ft'
u u o c o
CU CD CD J23 CL1
c/> to to E f i
fl3 T5 fO O *"O
CQ CQ CQ O CO
LO
UJ
1
CQ
O
CD
_J
Standard
CL
Q.
00
o
Q.
Q.
0
Q.
CL
CXI
r*
CL
CL
oo
1
O
z:
CM CO
4-12
-------
the most serious problems are not expected to achieve any of the standard
levels. Since these areas have significant levels of emissions and have
to apply full RACT regardless of the level of the standard, this results in
the same cost of control of the areas.
4.5.2 Sensitivity of Costs to Varying Assumptions
The items discussed in the sensitivity analysis in Chapter 3 also
impact upon the costs of RACT since the emission reductions required are
different. Table 4-5 presents the total costs of RACT utilizing low growth
rates, advanced RACT, reduced solvent emissions and EKMA in the base case
analysis.
Low Growth Rates
Lower growth rates used in projecting emissions affect the costs since
less control of existing sources is needed to offset the growth in emissions
from new sources. For standard levels less than or equal to .12 ppm, the
total RACT costs using low growth rates are 10 to 15 percent less than
when the higher growth rates are assumed. At a .14 ppm level, the difference
is greater than 20 percent.
Advanced RACT
For "advanced" RACT, the costs are naturally greater since more sources
are being controlled at a higher cost. The costs range from $440 million
at a .14 ppm standard to $1145 million at .08 ppm. In comparison with the
identified RACT case, the costs at .08 ppm with advanced RACT are 70 percent
greater, but 60 percent more AQCRs (26) are able to attain the standard with
the measures. At higher levels of the standard the costs are 45 to 60 percent
greater but only four to ten additional AQCRs are able to attain the standard.
4-13
-------
Reduced Solvent Evaporation Emissions
The effect of reducing the area source solvent evaporation emissions by
50 percent from the baseline inventory is comparable to the impact of
the lower growth rates on the results. Even though the reduction in emissions
are significant, the change in costs is not greater because there is not a
large reduction resulting in this category in the first place.
Combination of Low Growth Rates, Advanced RACT, and Reduced Solvent Emissions
When the previous three assumptions are combined, the result Is that
the costs are close to the base case costs. The increase in costs due to
advanced RACT are for the most part offset by the decreased costs of low
growth rates and reduced solvent emissions. At a level of .08 ppm, the
costs of this case are about 15 percent greater, while at .14 ppm the costs
are less than 10 percent less.
Empirical Kinetic Modeling Approach (EKMA)
With the general assumptions included in EKMA in this analysis, the
use of the approach results in greater emission reductions required for
areas to attain alternative standards. This naturally translates into
increased costs due to RACT since a higher level of control is required.
However, the differences in costs between EKMA and rollback increase
significantly as the level of the standard becomes less stringent. Costs
based on EKMA are less than 10 percent greater for a .08 ppm standard
since most areas are going to have to apply full RACT regardless of the
modeling approach. However, the implementation of full RACT allows more
areas to attain the standard using rollback than is the case for EKMA.
4-14
-------
LLJ
OO
T-
to M to
O -r- O)
O -l-> 1-
C 3
r a; to
(O T3 fO
+> >' OJ
o 2:
r^
CO
in
*
CO
oo
co
CO
et
o;
o
CtL OO
'ZZ O
O >-H
LU I I
C_> CQ
a:
=D«A
o---
oo
a.
0
t|_ O)
o -
Q i
H 1^
tO «r- <_3
4-> 4-> ef
to c a:
o a>
c_j -a
1 H
CD
it- O i
O S- O
3 S-
t/) O 4->
+-> oo c
to o
O S 0
0 0)
r^ LO *j- co
*
o o o o
v»
OO 00 00 00
o o o o
*f>
o
o
o
OO
oo
o
CJ
tO O
o
O
00
CM
CM
CM
CM
CM
CM
to
01
(O
0)
C
a>
3
O
S-
oo
10
I
=!
-a
r S_
o -a
oo
a.
CO
a.
a.
Q.
CM
-------
As the level of the standard increases, significantly more areas can attain
the standard with reduced control under rollback than under EKMA. At .10 ppm
the cost difference is almost 25 percent while at .14 ppm the difference is
close to 70 percent. The reason for this sizeable cost difference is that
under EKMA a majority of the areas will have to apply full RACT regardless
of the level of the standard.
4.6 SUMMARY OF COSTS FOR IDENTIFIED CONTROL MEASURES
Table 4-6 summarizes the total costs of the identified control measures
discussed thus far. This includes the FMVCP, new source control, and identified
RACT measures. As can be seen, the total costs of the measures considered
vary relatively little with alternative standards. The costs range from
$3.7 billion at the level of .08 ppm to $3.3 billion at .14 ppm.
The reason for the relatively small difference between standard levels,
when compared to the total cost magnitude, is twofold. First, the vast
majority of the total costs ($3.0 billion) result from the FMVCP and
new source controls, which are assumed to be the same regardless of the
level of the standard. Secondly, the RACT control costs do not differ
greatly from alternative standards since many areas have to apply full
RACT regardless of the level of the standard. One point to remember,
though, is that these costs do not represent the cost for al" 90 AQCRs to
attain the standard. AQCRs which fail to attain the alternative standard
with the identified control measures will have to apply additional control
measures. This will result in a larger difference between the costs
associated with the alternative levels of the standard.
4-16
-------
<
LU
1^.
*
CO
CO
CO
IT)
oo
cc.
Q
S
u_
OO
u_ ^
o o
oo _i
OO I I
O CO
co
4&
1
-
1
c£
z.
>-
1
1- 1
>
1 1
h-
1 1
oo
^
LU
oo
^3-
LU
-J
<:
i
C 00
O 3
r- O
4-> -i O)
fo > ai
c CD s_
-a a. i
r
O >4-
C_) O
l£> CO CM
* * *
OO OO CO CO
*ft
oo
-l->
oo
-o +-> c
CD C O
o a; -r-
3 > 00
"O i 00
OJ O <-
c£. oo E
LU
l& LT> *tf- OO
*
oo oo oo oo
o^
~o
O)
0 1
C 0
(13 c^
> ("V
-a
e^
< VD a»
O 3 -4->
i o *O
S- Di
CD
^f OO r i
* »
oo oo oo oo
{J)
r- S_
CD ^4 ^
> o -o
cu c
_1 "3
OO
E E E E
CL a. CL CL
Q. CL CL CL
OO O CM **
O r- r- r
oo
o
u
cu
o
^
a
a;
c
O 00
i- CU
^ "§
E 'u
O '-
O
r CO
O C
00 O
O) 00 !-
S- OJ 4->
ro +-> CL
OJ "3 E
C S- 3
00
O -C 00
"2
T3 O S-
O) S- O
T3 CDM-
3 S LO
O O 1
S- r <3-
ai s- co
S- O r-
ns
oo oo 1
OO OO O)
O O CO
c_> o oo
i CM CO
oo
LU
H-
o
zz
4-17
-------
The results of the sensitivity analysis for total costs are presented
in Table 4-7. The costs for the various cases vary not more than $400 million
from the base case costs for any of the alternative standard levels.
4.7 ESTIMATED COST OF ATTAINMENT
As discussed in Chapter 3, many AQCRs will not attain alternative levels
of the standard by applying the control measures outlined in this report.
Thus, additional control measures will be needed in some areas in order to
reduce emissions even further so that the standards can be attained. These
measures could include restrictive transportation control measures that
significantly reduce vehicle miles traveled in urban areas, control of
stationary sources for which RACT has not been identified or defined, and
tighter controls on new and existing sources than those achievable with RACT.
The cost-effectiveness of these measures has not been estimated, though
on the whole they are believed to be somewhat more costly than current
measures that have been defined. While some measures can no doubt be
implemented at costs approaching the level of RACT costs identified
earlier, many measures are likely to be much more costly. For example,
traffic reduction plans are extremely area specific and it is likely that
application of such measures could cost up to several thousand dollars
per ton of emissions controlled, though by the same token they could cost
must less. Likewise, the control of the many solvent evaporation sources
that have yet to be identified is likely to be more costly since these
sources tend to be relatively small and difficult to control from a
technical standpoint. For instance, the cost to control small coin-op
0
drycleaners is estimated to be about $5000 per ton of solvent controlled.
4-18
-------
Finally, the cost of applying more stringent controls in place of RACT on
new and existing stationary sources is likely to be high due to higher
marginal costs of control. Moving to more stringent levels of control
for particular sources will involve exponentially higher costs for a
relatively small reduction in emissions. For example, the marginal cost
of achieving 95 percent control at service stations versus 90 percent
Q
control is likely to be greater than $5000 per additional ton removed.
Even though the costs of additional control measures are not known,
it is still useful to estimate the cost of attainment in some manner in
order to better indicate the cost differences between alternative standards.
To do this, a cost-effectiveness estimate of $1,000 per ton controlled is
assumed for the additional emissions reductions past "identified" RACT
required for each AQCR to attain the alternative standards. While there
are likely to be many measures that are more costly than this, it is also
quite possible that many measures may be applied at cost levels approaching
projected RACT costs discussed earlier in this study. The $1000 per ton
figure represents a compromise between the extreme estimates. A sensitivity
is also included whereby the cost is increased to $2500 in order to see the
impact on the results.
In addition to the assumed cost-effectiveness of additional controls
making the cost differences conservative, the differences may also be
somewhat understated due to the fact that more stringent levels of the
standard are likely to require more sophisticated and extensive, and thus
more costly, controls. The approach used in this analysis does not take into
account such differences.
4-19
-------
TABLE 4-8. ESTIMATED COST FOR ALL 90 AQCRs TO ATTAIN ALTERNATIVE
STANDARDS - BASE CASE ($ BILLIONS)
Level
of
Standard
.08 ppm
.10 ppm
.12 ppm
.14 ppm
Total Cost of
Identified
Measures a
$3.7
3.5
3.4
3.3
Cost of Additional
Reduction
Required b
$2.5
1.6
1.0
0.7
Total
Cost of
Attainment
$6.2
5.1
4.4
4.0
Identified measures include FMVCP, new source control, and identified RACT.
Emission reduction from Table 3-4 multiplied by $1000/ton, which is a lower
bounds estimate.
4-20
-------
To determine the total cost of attainment in all 90 AQCRs, the
additional emission reduction required (from Table 3-4 for the base case),
is multiplied by $1,000 per ton, with the resulting costs added to the
costs of the identified control measures. The estimated costs of attainment
for the base case are presented in Table 4-8. The costs range from $6.2
billion for a .08 ppm standard to $4.0 billion for .14 ppm. As can be
seen, the difference in costs between alternative standards is greater,
ranging from a $400 million difference from .12 ppm to .14 ppm to over
$1.0 billion from .08 ppm to .10 ppm.
Table 4-9 summarizes the variation in total attainment costs with the
various assumptions discussed in previous sections in addition to a
different assumption on the cost of the additional emission reduction
required past identified measures. In this latter sensitivity, the cost
per ton is increased to $2500 per ton in order to indicate the impact on
the total costs if these advanced control measures are more costly.
It should be recognized that these advanced control measures will not
be implemented in the near-term, rather they are strategies that are likely
to be implemented closer to 1987. Prior to 1982, the RACT measures which
will be implemented are only those identified stationary source measures.
In addition, inspection/maintenance and transportation control measures will
not come about until after 1982.
4-21
-------
ro
OO
H-
00
O
h-
z:
LU
s:
i i OO
<=c 2:
1 O
1 1
cC 1
i
1 1-H
«=C CO
1
O iO
I x__^
-^-
t 1
^
o
> 1
1
1 1
o:
-^
T-
CU -i- 4->
01 "O O
rO T3 3
CU =t TD
S- CU
u D;
11 O
Is-. Lf) G"> i
CO r^- LO LO
&-$
<£
s:
LU
O ' CM OO
o~> r^ (& to
to
C 01
0 3
i- 0
4_> (- CU
ro > CU
c cu s-
i- s- .C
JD Q_ 1
E
0 <4-
o o
r~-^ oo 0*1 r**-
^j* «cj" oo oo
^o
01
"O 4-> C
cu c o
(_> O1
"D i 01
CU O T-
cc: oo E
LU
CO CO OO O
LO ^J" *d" **^
^/Y
-o
cu
o:
-o
OJ i *d" O
*
l£) LO *xj" ^J"
^i(^
_Q
.E 01
3 -<-> CU
O 2 4->
i o ro
i. C£
< S_
CU ro
> t) ~^
cu o c
1 ro
.j_>
OO
CM CO CO LD
*
LO ^j- co oo
-fc/"}
E E E E
Q. Q. Q. Q.
Q. Q. Q. Q.
CO O CM *3-
0 r r
* » *
E
S- 0
o s-
1 1 14
A
C C
0 0
T3 -i- 4->
O) -l-> ^s.
S- O O
r- 3 O
3 "o LO
CT CU CM
Q) S-i&
i. CU
C "O 01
C O C rO
O !- ro O
r- 01
-l-> 00 C CU
o !- o to
3 E -*-1 ro
"O CU "">. -Q
cu o
S- >>O CU
.a LO ^:
r- r +J
rO CZ ^^
CO 01
O 4-> >, rO
r ""^ o
4-> O CU
r- 0 T3 E
T3 O CU ro
-O r -r- 01
CL CU
4- OVr- .C
O C +-> -M
^ ^~
+-> >> 3 01
on E T-
0 Q.
O -r- 01 +->
4-> -r- C
to i cu
rs rs c E
i E 0 c
Q. !- -r-
^> -f-* ro
t^. ^2 O 4->
i rs +->
QJ CU
CU C S- H-
i 1- O
JD ro C
ro -i-> O -t-5
1 _d -r- oo
O to O
E oo 0
O !
S- -i- E r-
<+- OJ ro
+-> -U
01 01 CU O
-j_> o c" ^J
01 U -!->
O CU
o s- «3 s:
CU 4->
a> -t-> -a
-a +J c >
3 ro ro CM
r r
O LO C
c cu E
r- _C 01 3
-)-> C r
to E 0
4-> > 3 U
01 LO i
0 0 S-
u -a u o
c: u_
+-> ro S-
C O
CU «>LL.
E oo >>
C r
i > CU
ro i LO >
4-> 1 -r-
+J 01 OO -t->
rO C 0
E CU CL)
r 3 r CL
ro i -Q 01
-t-> O ro CU
O 0 h- 5-
I
ro
oo
LLJ
1
O
^
01
c
o
t
4_>
CL
E
rs
01
to
rO
S-
o
t+
LO
1
^*
CU
r^
-Q
ro
1
CU
CU
oo
-Q
1
i
4-^
r
13
E
01
r
C
O
r-
4~>
O
^3
"O
O)
S-
c
o
1
t/)
CO
r~
E
cu
[
ro
C
O
p
^
'-5
-o
ro
CU
.C
+J
-f-)
rO
t"
+J
-( >
CL
CU
0
X
a>
01
c
o
r-
-t-^
a. c
E o
13 4->
01 -^.
01 O
rO O
0
CUr
C/) -O^-
fO
O *4
0
OJ
01 -a
rO ro
-d CU
4_j
O) to
-C C
4-^ 'i
T3 C
CU 0
N 4->
r~ "^^^
r- O
r- O
-l-> LO
3 CM
OO
CU
01 >1
ro r~i
U
"O
CO O)
r- T
^~ i
h- CL
o
4-22
-------
4.8 REFERENCES FOR CHAPTER 4.0
1. Walsh, Michael P., Office of Mobile Source Air Pollution Control, EPA,
Memorandum to Walter C. Barber, "RARG Review of Proposed Revision to
the National Ambient Air Quality Standard for Oxidants," December 19, 1978.
2. Personal Communication from Michael P. Walsh, OMSAPC, EPA, to Kenneth H.
Lloyd, OAQPS, EPA, December 21, 1978.
3. U.S. Environmental Protection Agency, Air Quality Impacts of Transit
Improvements, Preferential Lane> and Carpool/Vanpool Programs,
EPA-400/2-78-002a, March 1978.
4. Motor Vehicle Manufacturers Association, Motor Vehicle Facts and
Figures '77. Detroit, Michigan.
5. Comsis Corporation, Travel Estimation Procedures for Quick Response to
Urban Policy Issues, Washington, D.C.: National Cooperative Highway
Research Program, 1977.
6. U.S. Water Resources Council, 1972 OBERS Projections: Economic Activity
in the U.S.. Volume 5, April 1974.
7. U.S. Department of Commerce, Bureau of Census, Supplemental Report to
1970 Census of Population: Population of Urbanized Areas Established
Since the 1970 Census, October 1976.
8. Environmental Protection Agency, Control of Volatile Organic Emissions
from Dry Cleaning Operations. Draft, April 1977.
9. Lloyd, Kenneth H., Economic Analysis Branch, OAQPS, EPA, "Cost of
Alternative Vapor Recovery Systems at Service Stations, " October 20, 1977,
4-23
-------
5.0 ECONOMIC IMPACT OF REASONABLY AVAILABLE CONTROL MEASURES
ON SELECTED INDUSTRIES
5.1 INTRODUCTION
Control measures to reduce emissions of hydrocarbons and other volatile
organic compounds will affect many industries in the U.S. economy. In
addition to major industries such as chemical manufacturing, petroleum
refining and automobile production, control measures will affect many
small industries which use petroleum products or organic solvents in a
wide range of applications. Countless industries which produce consumer
products use solvents in coating operations, while many industries use
solvents in metal and product cleaning. For examples of the industries
affected by control measures for hydrocarbons, consult Chapter 3.
An analysis of the economic impact of the revised standard on the
numerous industries affected is not possible to complete in the time frame
available for analysis. However, EPA has conducted economic impact studies
for the major emission sources, though these constitute only a portion of
the total number of sources affected. Economic analyses of all affected
industries are being initiated and significant results are expected within
two to three years. The Clean Air Act does not permit the consideration of
costs and economic impact in the setting of health related national
ambient air quality standards. The incomplete character of the economic
impact analysis for all affected sources does not mitigate EPA's legal
burden to promulgate the standard.
5-1
-------
Nonetheless, an attempt has been made to give an indication of the
economic impact of reasonably available control measures on selected
industries. These are industries for which EPA has conducted previous
cost or economic studies and for which quantitative or qualitative
economic judgements can readily be made.
5.2 PETROLEUM REFINING
5.2.1 Industry Profile
Crude petroluem is refined by 150 companies at 266 refineries located
in 40 different states. Production of refined products in the U.S. totalled
over 15 million barrels per day in 1976, or 93 percent of nameplate capacity.
The industry employs 100,000 workers and is heavily concentrated in the
West South Central region of Arkansas, Oklahoma, Texas, and Louisiana.
These four states employ 44 percent of all industry workers and supply 43
percent of all refined products. Refineries tend to be concentrated in
areas of the country that have oxidant levels that currently exceed the
standard.
The petroleum refining industry is somewhat concentrated. The five
leading producers own 36.5 percent of all industry capacity; the top ten,
58.5 percent. These leading producers are integrated, major oil companies
that engage in exploration, production, refining, distribution, and marketing
on the retail level. Other refiners are independent companies that are
typically not integrated into more than one other segment of the industry.
Prices vary little among companies, although there are occasional examples
of price cutting when there is weak demand and an excess of supply.
5-2
-------
5.2.2 Costs of Reasonably Available Control Measures
Petroleum refineries are significant sources of volatile organic
compound (VOC) emissions. The major point sources of VOC emissions from
petroleum refineries are vacuum producing systems (VPS), wastewater
separators (WWS), process unit turnarounds (PUT), and leaks from
miscellaneous sources such as pumps, compressors, and valves. EPA has
analyzed control techniques for these sources, which reduce emissions
by 95 to 100 percent. For vacuum producing systems, control can be
achieved by venting the emissions to a firebox. Large reductions in
emissions can be accomplished for the wastewater separators through
covering the separators and forebays. Emissions can be controlled from
process unit turnarounds by piping the VOC to a flare or to the fuel gas
systems. Finally, emissions from miscellaneous equipment can be substan-
tially controlled by reducing equipment leaks with a monitoring and
maintenance program.
Costs of these control measures have also been estimated by EPA based
on analysis of a model sized refinery with a throughput of 100,000 barrels
2
per day. Costs for individual refineries will vary considerably due to
differences in size, configuration and age of facilities, product mix,
and degree of control.
The capital cost for piping for controlling emissions from vacuum
producing systems will range from $23,700 to $51,600, depending on whether
the system uses surface condensers or contact condensers. However, this
control measure should result in an annual savings of $89,000 to $96,700
5-3
-------
due to significant credits for the value of the recovered petroleum. For
wastewater separators, the capital cost of covers for the facilities will
be approximately $62,800, but again due to the value of the recovered
petroleum products, a net annual savings of $309,700 should result. The
control method for process unit turnarounds has an estimated capital cost
of $97,600 for piping and valves. Petroleum product recovery is not as
readily recoverable from this control measure, though some refineries
currently have facilities for recovering the hydrocarbons. Nonetheless,
assuming no recovery of petroleum products, an annualized cost of approxi-
mately $25,900 will result. The total capital cost for these three sources
will range from $184,000 to $212,000. However, this capital outlay is more
than offset by the annual recovered petroleum credits valued at $432,000,
which does not include the value of recovered PUT emissions.
Finally, the cost of a monitoring and maintenance program to detect
3
and control equipment leaks has also been estimated. For a model 100,000
barrel per day refinery, the capital cost for monitors will approximate
$8,600. The annualized cost of a program, taking into account monitoring
and maintenance labor, materials, and capital charges, will be about
$103,000. This annualized cost, though, should be offset to some degree
by recovery credits of reduced emissions. However, emission reduction
factors are not presently quantifiable and thus recovery credits will be
extremely variable.
5.2.3 Economic Impact of Control Measures
The economic impact of the control measures outlined in the previous
section is expected to be small for several reasons. First, the capital
5-4
-------
costs are not large in comparison with the capital cost of a refinery as
a whole. The capital cost for a model refinery with a throughput of
100,000 barrels per day is estimated to range from $300 million to $500
million.4 A capital outlay of $220,000 for the VOC control measures
represents an insignificant increase (less than 0.1 percent) in the
capital cost of a refinery.
Secondly, the value of recovered petroleum products offsets the
capital costs entirely in the first year. Thus, the controls can be
justified solely on economic grounds. Lastly, the economic impact should
not be large since a significant portion of the industry has already
instituted the controls. Twenty-five percent of the industry already
controls vacuum producing systems, 80 percent controls wastewater
separators, and 40 percent controls process unit turnarounds. Finally,
a comprehensive study of the petroleum refining industry indicates that
EPA's total air and water pollution regulations will result in only a
small impact on this sector.
5.3 RETAIL GASOLINE SERVICE STATIONS
5.3.1 Industry Profile
In 1977, there were approximately 178,000 gasoline stations in the
U.S. Over 48,000 service stations have closed in the U.S. since the
population peak of 226,000 in 1972. This attrition is expected to continue
at least through the early 1980's to a leveling off point of anywhere from
125,000 to 150,000 outlets. The economies of scale of high volume stations
and the shift to self-service operations are a prime factor in shrinking
retail margins. Consequently, the closure of outlets due to market
5-5
-------
rationalization processes will be most severe for those outlets which
have relatively low sales volume coupled with high unit expenses.
Major oil companies and regional refiner/marketers supplied over
half of the retail service stations in the country with the remaining
43 percent supplied by independent marketers. The traditional retail
marketing strategy of major oil companies has been to operate through
lessee dealers. These lessee outlets still represent approximately
two-thirds of the major oil company stations and almost 50 percent of
all stations in the country. The second largest group of outlets are
known as open dealers. In these operations, the onsite dealer actually
owns or controls the investment in his station where he is physically
employed. Open dealers represent over one-third of the retail outlets in
the U.S. They are generally branded (i.e., station operating under the
brand identification of a major oil company) and supplied either directly
by a major oil company or a branded jobber. The other types of retail
operations are direct salary outlets and convenience stores, which are
low expense, low margin operations which account for less than 25 percent
of the total population of gasoline retailers. A summary of the service
station market segments is presented in Table 5-1.
Retail service stations dispense an average of about 40,000 gallons
per month. In recent years, marketing economics have resulted in a trend
toward stations with larger volumes, with small volume operations being
marginal operations that have to rely on other parts of the retail trade,
such as mechanical work and sales of accessories, in order to remain in
business. The high volume stations tend to be mostly direct operations
5-6
-------
(O
-M
O
CO
r^
si-
o
o
in
00
co
o
o
tn
IO
o
o
CO
CM
00
c
o
o
o
o
O
o
CM
z.
0
1 .-,
1
1
_l
~~i
J
Q-
o
CL.
^^
«£_
O
1
1
^
00
UJ
0
5
3
O
M-
0
s-
0)
s
-5*
c:
o
r
4->
rO
s-
d
o
t-
0
cu
Q.
>)
^^
s-
cu
r^
to
cu
Q
cu
Q.
O
VO
in
r
0
o
00
f^
CM
^SJ.
r~
*
r^
o
o
o
ft
CM
CO
si-
O
O
CM
CU CO
0) CM
l/>
" I
<
O
in
*&^
co
ID
"^'
O
O
CD
^^
IT)
o
CM
««-
0
O
«t
^1-
CT>
»
O
r
X.^"
o
0
o>
r
o>
VO
sf
s ^
0
o
A
ro
CO
i
in
cu
co
o
o
co
o
o
ro
cr»
CD
o
CO
CM
O
O
O
A
in
o
CO
o
o
CM
CO
S-
o>
c
*^
0)
a:
i^
OJ
r^
r^
a
Q.
3
oo
,_
rO
C
s- o
O -r-
r-> Ol
C
S
c
-------
which are controlled and operated by the supplier and operate on relatively
low margins. Low volume stations, those dispensing less than 25,000 gallons
per month, are mostly lessee dealers and open dealers supplied by all classes
of suppliers. These low volume stations, which comprise close to 50 percent
of the total number of stations, are the segment of the retail industry that
is most vulnerable to changes in marketing economics as well as external costs
such as vapor recovery costs.
5.3.2 Cost of Reasonably Available Control Measures
Emissions occur from two major sources at service stations - the loading
of underground storage tanks (Stage I) and the refueling of motor vehicles
(Stage II). For Stage I emissions, vapors can be controlled through the
use of a vapor balance system, where vapors are vented by displacement to
an intermediate holding area (usually the tank truck) for ultimate disposal
or recovery at the bulk terminal or bulk plant. Stage II emissions can be
controlled through a variety of systems, the most basic of which is the
balance system where vapors from the refueling operation are displaced by
means of a tight fitting nozzle and vapor return lines to the underground
tank. More elaborate recovery systems create a vacuum where vapors are
drawn from the refueling operation, alleviating the need for a tight nozzle
fit. Vapors are again displaced to the underground storage tanks, with the
excess vapors being incinerated in most cases. There are several variations
of this vacuum assist system which are too numerous and involved to discuss here.
Preliminary estimates indicate that vapor balance systems are less costly
than vacuum assist systems and provide for between 80 to 90 percent control of
the HC emissions. While it may ultimately be necessary to use the vacuum system
in specific locations, as yet these locations have not been defined. Consequently,
5-8
-------
only the cost of the vapor balance system is considered in this analysis.
The capital costs of the balance system varies with the number of dispensers
at the station, the number of underground tanks, and the physical layout of
the station. For a typical nine dispenser, three island station, the capital
costs will approximate $8,800. These costs can range from $4,500 for a
Q
two-dispenser station to over $11,000 for a 15-dispenser facility.
Essentially the only operating and maintenance cost associated with
the balance system is that for nozzle maintenance since the system does not
require any power to operate and there are no moving parts associated with
the remainder of the system. Nozzle maintenance requirements will be
extremely variable depending on a number of factors. However, it is expected
that the nozzle will have to be replaced only once a year or the faceplate
and/or boot repaired or replaced no more than twice a year. This would result
in an annual maintenance cost of about $60 per nozzle, or about $540 for a
nine-nozzle station.
Since the vapor balance system is characterized by a high fixed cost
component, the annualized cost per gallon of throughput is naturally highly
dependent on the volume of the station and the cost of investment capital for
the station. Costs range from about 0.1 cent per gallon for high-volume
direct operations to over one cent per gallon for low-volume open dealer.
Costs for other low volume outlets range from 0.5 to 0.6 cent per gallon
while other medium to high volume outlets have costs ranging from 0.2 to
0.4 cent per gallon.
5-9
-------
5.3.3 Economic Impact of Control Measures
The direct economic effect of vapor recovery at service stations
is to reinforce the existing economies of scale in gasoline marketing.
The competitive position of high volume outlets may be strengthened
since their economics will not be significantly affected. On the other
hand, low volume outlets which are already marginal operations will have
their position eroded even further, even though it is expected that most
of these marginal stations will close in the next five years regardless
of the requirement of vapor recovery due to unfavorable station economics.
While the costs for the balance system are insignificant to the
consumer, the costs are still of appreciable magnitude to the dealer,
who typically has a profit margin of one cent or less per gallon on
gasoline. Thus, in some instances, vapor recovery costs could entirely
wipe out profit margins and in other cases severely reduce the margin by
over 50 percent. In addition, some owners of stations may have difficulty
obtaining the capital necessary to finance the vapor recovery equipment.
Highly leveraged firms may not have the capacity to absorb additional
debt and thus could not obtain loans. This aspect is particularly crucial
to large independent marketers who typically own anywhere from 20 to 100
stations and hence would have to come up with a sizable sum of investment
capital for vapor recovery systems.
It is difficult to segregate the marginal stations which will
eventually survive in the marketplace but would have to close with the
requirement of vapor recovery. In a study conducted for EPA and OSHA by
Arthur D. Little, Inc., an attempt was made to estimate the number of
5-10
-------
closures nationwide which would result due to market forces, due to
capital availability constraints for vapor recovery investment, and
finally due to the impact of the vapor recovery costs on the profitability
g
of stations. The analysis indicated that over 20 percent of the current
population could close by 1981 due solely to market forces, with over
75 percent of closures resulting in the lessee dealer segment of the
market. Vapor recovery requirements, on the other hand, would result in
additional closures representing about six percent of the current population,
or just over 10,000 stations nationwide. Around 12 percent of these vapor
recovery-induced closures would be in the large independent segment of
the market where companies with large numbers of stations would be unable
to obtain the required investment to finance vapor recovery systems at all
their stations. The remainder of the closures would be open dealers for
whom the increased costs would severely reduce or eliminate profit margins
and make staying in business unattractive. The closures in this segment of
the market would represent about 17 percent of the total open dealer
stations.
5.4 GASOLINE BULK PLANTS
5.4.1 Industry Profile
Bulk gasoline loading plants are typically secondary distribution
facilities which receive gasoline from bulk terminals by trailer transports,
store it in above-ground tanks, and subsequently dispense it via account
trucks to local farms, businesses, and service stations. Bulk plants may
be owned by a major or independent petroleum refiner, an independent jobber,
or an individual operator. Although operation and ownership of bulk plants
5-11
-------
include cooperatives and salaried employees, the predominant types are
the independent jobber and the commission agent who operates the plant
for a larger refiner but owns his own delivery trucks.
Currently there are less than 20,000 bulk plants in operation in
the U.S. with almost half having daily throughputs less than 4,000 gallons.
This represents a decline of nearly 4,000 stations facilities from the
population in 1972. This trend is expected to continue as major oil
companies dispose of their many bulk plants as they decline in importance
in gasoline distribution and become less profitable. While bulk plants
served useful pruposes in years past, their role in the distribuiton chain
have declined since more stations are receiving deliveries directly from
bulk terminals. Bulk plants are being bypassed since economies of labor
and capital can be realized if the transport truck can deliver directly
from the terminal to an account, thus reducing the cost of gasoline to
the station by an appreciable amount. Another important factor in the
decline of bulk plants has been and will continue to be a decline in the
customer population served by bulk plants. Small stations and commercial
accounts which once depended upon bulk plants are also undergoing a
significant attrition in the retail market. Even commercial accounts
once served by bulk plants are receiving deliveries directly from terminals,
No estimates are available which indicate what the bulk plant population
will be in the next five years or once the market rationalization process
is completed, though it is expected to be significantly less than the
current population.
5-12
-------
5.4.2 Cost of Reasonably Available Control Measures
Control of breathing, working and miscellaneous losses resulting from
storage and handling of gasoline at bulk plants can be accomplished through
submerged fill, balance systems, vapor processing systems and control of
truck loading leaks. Vapor processing systems have not been applied to
bulk plants, but have been used to recover hydrocarbon vapors at bulk
terminals during truck loading.
By changing from top splash loading to submerged fill, vapors generated
by loading tank trucks can be reduced by about 58 percent. Submerged fill
decreases turbulence, evaporation, and eliminates liquid entrainment. The
cost to install submerged fill at the typical plant with three loading arms
is less than $1,000. This cost is more than offset by the cost savings that
result from the elimination of the generation of vapors during loading.
The vapor balance system operates by transferring vapors displaced from
the receiving tank to the tank being loaded. A vapor line between the truck
and storage tanks essentially creates a closed system permitting the vapor
spaces of the two tanks to balance with each other. In addition, vapor
balancing of incoming transport trucks displaces vapors from the storage
tanks to truck compartments, with the emissions ultimately being treated at
the terminal with a secondary recovery/control system. The vapor balance
system can reduce emissions from the bulk plant by around 90 percent.
The capital costs for vapor balance systems at bulk plants will vary
depending upon a number of factors, such as the configuration of the plant,
age and condition of tanks, and requirements for additional equipment due
to local regulations such as fire laws. In areas where these regulations
are less restrictive, vapor balance costs are substantially lower. Based
5-13
-------
on an analysis of costs from various sources, EPA estimates the capital costs
for converting to submerged fill and installing the cheaper vapor balance
systems to be around $4,000 for a 4,000 gallon per day (gpd) facility
and $5,000 for a 20,000 gpd plant. Except to a minor extent, these
costs are not a function of throughput of the bulk plant since the
number of tanks is relatively independent of throughput and the number
of delivery trucks serviced is small. The annualized costs for these
model facilities are offset by a credit for gasoline recovery. The
credit, which includes only the savings for the emissions which are not
generated in the first place as opposed to the vapors which are returned
to the bulk terminal for processing, is naturally a function of throughput.
The large bulk plant has a sizeable gasoline recovery credit.
In areas where local regulations prohibit use of the cheaper vapor
balance systems, a more complete and expensive balance system will have
to be installed. Capital costs for converting to submerged fill and
installing the complete balance system would range from about $23,000
for a 4,000 gpd facility to close to $26,000 for a 20,000 gpd plant.
The annualized costs for these model facilities amount to around $3,500 for the
small plant (or about 0.3 cents per gallon) and about $750 for the large
plant (less than 0.1 cent per gallon).
5.4.3 Economic Impact of Control Measures
The economic impact on bulk plant operators due to vapor recovery
requirements depends on the system which can be installed at the plant.
If the cheaper balance systems can be used, the capital costs are not
of such magnitude as to cause a significant impact in the industry.
5-14
-------
An economic impact analysis is being conducted by EPA which will quantify
the potential impacts which could result from the range of vapor recovery
system costs.
The magnitude of the costs for the more expensive balance systems
will likely have a more significant impact on bulk plants. A capital
outlay of around $23,000 represents a significant investment for small
plants. Some sources have reported sales prices for bulk plants that
have been sold by major oil companies which range between $45,000 and
$65,000. Thus, investment for the expensive vapor balance systems
represents one-third to one-half of these transaction prices. A
preliminary economic analysis prepared for EPA concludes that bulk plants
having throughputs less than 4,000 pgd are either unprofitable or only
marginally profitable and could possibly be unable to cope with an
expenditure of this magnitude. EPA's current analysis will better
determine the extent of the potential impacts of these costs on small
bulk plants.
When assessing impacts on bulk plants or potential closures of
plants, it is important to consider several points. First, until SIP
revisions are submitted, enforcement discretion is being utilized to
prevent closures from occurring due to pollution control requirements.
Guidance has been developed and provided to the states in developing
the SIP revisions. Secondly, not all bulk plants in the country will
be affected by vapor recovery regulations. Only bulk plants in
5-15
-------
non-attainment areas will be required to install controls. Since bulk
plants are concentrated in rural areas, it is likely that a large number
of bulk plants would not be required to install vapor recovery systems.
Finally, most of the bulk plants that could be closed by vapor recovery
requirements are likely to go out of business or dispose of their gasoline
operations in the near future due solely to market forces. Hence, vapor
recovery requirements could only accelerate closures that will take place
even in the absence of such requirements.
5.5 AUTOMOBILE ASSEMBLY PLANTS
5.5.1 Industry Profile
In the model year 1976, over 8.5 million automobiles were sold by
U.S. automakers. They represented a significant rebound from 1974 and
1975 sales levels when, respectively, only 7.3 and 6.7 million cars were
sold. General Motors and Ford Motor Company dominate the industry as the
two companies accounted for 58 and 24 percent of the autos produced in
1976, respectively. The other two major automakers, Chrysler and American
12
Motors, accounted for 16 and three percent, respectively.
There are currently 46 auto assembly plants in the U.S., though this
number can vary due to temporary shutdowns and switchovers to light-duty
truck assembly. GM has 22 of the plants while Ford has 14. The remainder
of the plants are owned by Chrysler and AMC as well as Checker Motors and
Volkswagen, which is opening a new plant in Pennsylvania. The locations of
13
these plants are indicated in Table 5-2. Essentially all of the plants are
located in non-attainment areas for oxidants.
5-16
-------
Table 5r2.
Manufacturer
American Motors
Chrysler Corp.
Ford Motor Co.
General Motors
Checker Motors
Volkswagen
U.S. AUTOMOBILE ASSEMBLY PLANTS
Location
Kenosha, Wisconsin
Toledo, Ohio
Belvidere, Illinois
Hamtramck, Michigan
Detroit, Michigan
Newark, Delaware
St. Louis, Missouri
Atlanta, Georgia
Chicago, Illinois
Dearborn, Michigan
Kansas City, Missouri
Lorain, Ohio
Los Angeles, Calif.
Mahwah, New Jersey
Metuchen, New Jersey
St. Louis, Missouri
San Jose, California
Twin Cities, Minnesota
Wayne, Michigan
Wixom, Michigan
Arlington, Texas
Baltimore, maryland
Detroit, Michigan
Doraville, Georgia
Fairfax, Kansas
Flint, Michigan
Framingham, Mass.
Fremont, California
Janesville, Wisconsin
Lakewood, Georgia
Lansing, Michigan
Leeds, Missouri
Linden, New Jersey
Lordstown, Ohio
Norwood, Ohio
Pontiac, Michigan
St. Louis, Missouri
South Gate, Calif.
Tarrytown, New York
Van Nuys, California
Willow Run, Michigan
Wilmington, Delaware
Kalamazoo, Michigan
Pennsylvania
5-17
-------
The earnings of the automakers sagged in 1974 and 1975 due to reduced
sales levels. However, in 1976 earnings expressed as return on equity or
return on assets returned to historical levels, though American Motors is
still experiencing financial difficulties.
5.5.2 Cost of Reasonably Available Control Measures
For the paint coating of auto bodies at assembly plants, numerous
options exist for the control of VOC emissions, with control ranging
from 70 to 95 percent. Options potentially consist of process changes,
such as electrodeposition (EDP) of the primecoat and water-borne topcoats,
and add-on control devices such as carbon adsorption, thermal incineration,
and catalytic incineration. For purposes of this study, the most cost-effective
control options for prime and top coating were chosen that resulted in at
least 80 percent control. Other control options could possibly be chosen
in actual existing plants, but this analysis considers only the least costly
14
option, based on costs furnished to EPA by Springborn Laboratories. The
following control option was chosen:
Prime coating: EDP with water-borne dip and solvent guide coat
Prime and top coat spray booths: Catalytic incineration with
primary heat exchange
Prime and top coat ovens: Catalytic incineration with primary
heat exchange.
Add-on controls in addition to EDP are needed for the prime coating
operation in order to control emissions from the application of the solvent
guide coat.
5-18
-------
Costs for these options have been estimated for a "model" auto
assembly plant producing 211,200 bodies per year. The capital cost of
converting the plant to EDP and adding the control devices is estimated to
be about $20.2 million, with $15.7 million resulting from the conversion
to EDP, $0.8 million from the prime coat add-on devices, and $3.7 million
from the top coat add-on controls.
Annualized costs have also been estimated taking into account operating
and maintenance costs of the processes and devices as well as the depreciation
and interest charges. Only the incremental O&M costs incurred over the
existing base case (solvent-borne prime and top coats with no control) are
included in the estimates. However, the entire capital charges of the new
processes and devices are included since it is assumed that the existing
equipment has no salvage value. Salvage values will vary significantly
from plant to plant and thus it is difficult to generalize on an appropriate
value. Based on these assumptions, the increased annualized cost of control
is estimated to be almost $34 per car.
5.5.3 Economic Impact of Control Measures
E-PA is conducting but has not completed a formal study of the economic
impact of these controls on the automobile industry. However, tentative
conclusions can be drawn.
Currently, about 60 percent of the assembly plants employ EDP to
apply prime coats. Since this process change contributes almost half of
the annualized cost per body, many of the existing plants will be able to
5-19
-------
achieve the required additional control for about $18 per car. In
addition, due to the fact that such a large portion of the industry has
already moved to EDP for economic and technical reasons the economic impact
of such a switch for the remainder of the industry should not be unduly
burdensome.
The impact on sales of automobiles is not expected to be significant.
Rough estimates based on the Ford Econometric Sales Forecasting Model indicate
that a $34 increase in the cost of the average automobile could result in a
15
reduction in sales of 0.2 percent in 1983. This is not a reduction in
sales from current levels, but rather a reduction in levels that would
otherwise occur in 1983. Such a reduction in foregone sales will have a
negligible effect on the return on investment in the industry.
It is important to remember that these general conclusions are based
on model plants and average conditions in the industry. Though none of
the major firms are expected to experience serious impacts, some individual
plants will experience more costly conversions to alternative processes
which could affect their viability. A determination of the individual
plants which have the potential to be severely impacted has not been
determined and is beyond the scope of this analysis.
5.6 METAL FURNITURE INDUSTRY
5.6.1 Industry Profile
The metal furniture industry consists of about 1600 firms employing
nearly 100,000 people and producing around $3.4 billion in metal furniture
5-20
-------
shipments in 1975. The industry is highly fragmented, including the
following categories of products: household metal furniture, office metal
furniture, public building furniture, and metal partitions and fixtures.
Around 500 firms manufacture household furniture, another 500 manufacture
partitions and fixtures, over 400 produce public building furniture, and
200 firms engage in office furniture manufacture.
The industry is characterized by relatively small manufacturers.
Whereas single-unit firms, those with one establishment for both manu-
facturing and administration, account for only 19 percent of the value of
shipments for all manufacturing establishments listed by the Census of
Manufacturers, such firms account for over double the average for household
furniture, public building furniture, and partitions and fixtures. Only
the office furniture segment of the market is consistent with the overall
industry average. In addition, over 85 percent of all metal furniture
establishments employ less than 100 people. In fact, more than 50 percent
of the establishments in segments other than office furniture have less
than 20 employees, according to the 1972 Census of Manufacturers. The
significant number of small firms indicates that no economies of scale
are evident which prohibit small manufacturers from competing, especially
in regional markets where low labor productivity may be overcome by lower
distribution costs.
The manufacturing markets of metal furniture facilities vary. Some
plants manufacture furniture to be sold directly to consumers through
5-21
-------
retail stores. In contrast, job shops, which produce furniture on a
contract basis, apply coatings on many different furniture pieces according
to the customer's specifications.
Metal furniture plants are located throughout the U.S. However, the
states of Illinois, California, Michigan, New York, and Pennsylvania
contain over 50 percent of the establishments in the industry.
5.6.2 Costs of Reasonably Available Control Measures
Measures to reduce volatile organic emissions from metal furniture
coating operations consist of process changes as well as exhaust gas
treatment with add-on control devices. Applicable process changes include
conversion to waterborne coatings, high solids coatings, powder coatings,
and electrodeposition (EDP) of waterborne coatings. Add-on control devices
include carbon adsorption and incineration. While each of these measures
achieves reasonable levels of control, the option chosen by an individual
plant will depend upon circumstances specific to the plant.
EPA has estimated costs of the alternative measures on the basis of
18
model plants in order to indicate the relative costs of the alternatives.
These models are one-color lines and are sized based on the annual product
coverage rates for the coating lines.
For electrostatic spray lines, the most feasible control option appears
to be conversion to high solids coatings in order to reduce solvent emissions.
The reduction can range from 50 to 90 percent depending upon the type of
coating used previously. For a three million square feet per year coating
line, the capital cost to convert the line is approximately $15,000, while
5-22
-------
for a large plant (48 million square ft/yr) the capital cost will approximate
$62,000. These costs represent a five to six percent increase in the invest-
ment in the existing line. However, in both cases, the increased capital
costs appear to be justified on economic grounds due mainly to the savings
in lower applied film cost when compared to conventional solvent coatings.
For the larger plant, the cost savings in the first year offset the capital
cost entirely, while for the smaller plant the savings represent a return
on investment approaching 20 percent.
Conversion to waterborne coatings, appears to be the most feasible option
for dip coating lines. Switching to waterborne coatings would entail a
capital investment of $3,000 for a smaller facility (seven million square
feet/year) and $5,000 for a larger plant (22.5 million square feet/year).
These costs represent an increase in investment of two to three percent.
There is an increase in operating and maintenance costs due to higher
materials costs, resulting in an increase in coating costs of seven percent
for the smaller plant and four percent for the larger facility.
5.6.3 Economic Impact of Control Measures
An analysis of the economic impact of these costs on the segments of
the metal furniture industry has not been conducted by EPA, thus no
definitive conclusions can be drawn. However, it appears from the model
plant analysis that conversion to high solids coatings for electrostatic
spray lines is justified from an economic standpoint and would be of
benefit to that portion of the industry utilizing this coating application
method.
5-23
-------
On the other hand, conversion to waterborne coating for dip coating
lines is more difficult to assess since one has to consider both the capital
investment requirements as well as the increased cost of coating on an
annual basis. With regard to this latter point, the 1972 Census of
Manufacturers indicates that the cost of coating materials comprise
19
0.8 to 1.4 percent of the value of shipments for metal furniture. An
increase of four to seven percent in coating costs resulting from the
conversion to EDP will affect the final selling price for the metal furniture
by an insignificant amount (less than 0.1 percent). In addition, a capital
investment of $3,000 to $5,000 does not appear to be burdensome for most
establishments. Some small, marginal facilities may find such an investment
level unjustified, but the extent of this is not known.
5-24
-------
5.7 REFERENCES FOR CHAPTER 5
1. Energy and Environmental Analysis, Inc. Estimated Cost of Benzene
Control for Selected Stationary Sources, February 27, 1978, p. 3
2. Environmental Protection Agency, Control of Refinery Vacuum Producing
Systems, Wastewater Separators, and Process Unit Turnarounds,
EPA-450/77-025, October 1977, p. 4-10.
3. Environmental Protection Agency, Control of VOC from Petroleum
Refinery Equipment, Draft, April 1978, p. 4-7.
4. Energy and Environmental Analysis, p. 10.
5. Environmental Protection Agency, Control of Refinery Vacuum Producing
Systems, Wastewater Separators, and Process Unit Turnarounds,
EPA-450/77-025, October 1977, p. 5-3.
6. Sobotka and Company, Inc., Economic Impact of EPA's Regulations on
the Petroleum Refining Industry. EPA-230/3-76-004, April 1976.
7. Arthur D. Little, Inc., The Economic Impact of Vapor Recovery Regulations
On the Service Station Industry, Report prepared for EPA and OSHA,
March 1978, p. 6.
8. Lloyd, Kenneth H., "Cost of Alternative Vapor Recovery Systems at
Service Stations," Economic Analysis Branch, OAQPS, EPA, October 20, 1977.
9. Arthur D. Little, p. 131.
10. Sobotka and Company, Inc., Bulk Plant Vapor Controls Economic Impact,
August 15, 1977, p. 3.
11. Pacific Environmental Services, Inc., Economic Analysis of Vapor
Recovery Systems on Small Bulk Plants, September 1976, p. 2-2.
12. Motor Vehicle Manufacturers Association, Motor Vehicle Facts & Figures '77,
pp. 8 and 9.
13. Springborn Laboratories, Inc., Study to Support New Source Performance
Standards for Automobile and Light-duty Truck Coating, June 1977,
EPA-450/3-77-020, p. 35.
14. Ibid, pp. 8-42 and 8-43.
15. Personal communication from Tom Alexander, OPE/EPA to Ken Lloyd
OAQPS/EPA, March 13, 1978.
5-25
-------
16. Springborn Laboratories, Inc., Study to Support New Source Performance
Standards for Surface Coating of Metal Furniture. EPA-450/3-78-006,
April 1978, p. 8-1.
17. Springborn, p. 8-15.
18. Environmental Protection Agency, Control of Volatile Organic Emissions
from Existing Stationary Sources - Vol. Ill: Surface Coating of Metal
Furniture, December 1977, pp. 3-8 and 3-9.
19. Springborn, p. 8-22.
5-26
-------
APPENDICES
-------
APPENDIX A
Ozone Design Values for 90 Air Quality Control Regions
(Estimated)
-------
OZONE DESIGN VALUES FOR 90 AIR QUALITY CONTROL REGIONS
(Estimated)
AQCR NUMBER
AQCR NAME
EXPECTED SECOND
HIGH DAILY
VALUE3
(PPM)
EXPECTED SECOND
HIGH HOURLY
VALUEa
(PPMj
002
004
005
013
015
016
018
022
024
025
028
029
030
031
033
036
038
041
042
043
045
047
048
049
050
052
055
056
060
062
065
067
069
Columbus-Phoenix, GA
Metropolitan Birmingham
Mobile-Pensacola, AL-FL
Clark-Mohave, AZ-NV
Phoenix-Tucson, AZ
Central Arkansas
Metropolitan Memphis,
AR-MS-TN
Shreveport, LA
Metropolitan Los Angeles
North Central Coat, CA
Sacramento Valley, CA
San Diego, CA
San Francisco, CA
San Joaquin Valley, CA
Southeast Desert, CA
Metropolitan Denver, CO
San Isabel, CO
Eastern Connecticut
Hartford-New Haven-
Springfield, CT-MA
New Jersey-New York-
Connecticut (NJ-NY-CT)
Metropolitan Philadelphia
NJ-PA
National Capital
(DC-MD-VA)
Central Florida
Jacksonville-Brunswick,
(FL-GA)
Southeast Florida
West Central Florida
Chattanooga, TN-GA
Metropolitan Atlanta, GA
State of Hawaii
Eastern Washington-
Northern Idaho
Burlington-Keokut, IA
Metropolitan Chicago, IL
Metropolitan Quad Cities,
IL-IA
.15b
!l5c
.15
.12
.14C
.13
.14
.14
.38
.11
.14
.24C
.17
.15
.25
.17
.09
.23
.27
.24
.26
.19
.10
.13
.12
.14
.11
.14
.06
.08
.11
.20
.11
.15b
.15
.16
.15
.14
.13
.14
.15
.38
.12
.19
.24
.19
.19
.25
.17
.09
.23
.32
.27
.30
.21
.10
.13
.13
.14
.11
.15
.08
.08
.12
.26
.13
A-l
-------
070 Metropolitan St. Louis,
IL-MO .23C .23
073 Rockford-Janesville-
Beloit, IL-WI .14 .17
078 Metropolitan Louisville, KY .17 .22
079 Metropolitan Cincinnati,
KY-OH .17 .20
080 Metropolitan Indianapolis,
IN .17 .17
081 Northeast Indiana .17° .17&
082 South Bend-Elkhart-
Benton Harbor, IN-MI .16b .16b
085 Metropolitan Omaha-
Council Bluff, IA-NE .10 .10
092 South Central Iowa .10 .11
094 Metropolitan Kansas City,
KS-MO .10 .12
099 South Central Kansas .17C .17
106 Southern Louisiana-
Southeast Texas .19 .19
113 Cumberland-Keyser, MD-WV .12C .12
115 Metropolitan Baltimore, MD .23 .25
118 Central Massachusetts .16 .16
119 Metropolitan Boston, MA .16 .17
120 Metropolitan Providence,
MA-RI .19 .19
121 Merrimack Valley-Southern
New Hampshire, MA-NH .17 .17
122 Central Michigan .13 .17
123 Metropolitan Detroit-Port
Huron, MI .18 .23
124 Metropolitan Toledo, MI-OH .14° .14
125 South Central Michigan .09 .09
128 Southeast Minnesota-La
Crosse, MN-WI .13 .16
131 Minneapolis-St. Paul, MN .09 .12
151 Northeast Pennsylvania-Upper
Delaware Valley, PA-NJ-DE .23 .23
152 Albuquerque-Mid Rio Grande,
NM .14C .14
153 El Paso-Las Cruces-
Almogordo, NM-TX .16 .16
158 Central New York .12 .12
160 Genesse-Finger Lake, NY .13 .13
161 Hudson Valley, NY .14 .14
162 Niagara Frontier, NY .16 .18
167 Charlotte, NC .14 .17
173 Dayton, OH .18 .18
174 Greater Metropolitan
Cleveland, OH .19 .19
176 Metropolitan Columbus, OH .16 .16
A-2
-------
178 Northwest Pennsylvania-
Youngstown, OH-PA .19 .21
184 Central Oklahoma .11 .12
186 Northeastern Oklahoma .14 .18
193 Portland, OR-WA .13 .16
195 Central Pennsylvania .13 .15
196 South Central Pennsylvania .18 .19
197 Southwest Pennsylvania .18 .20
199 Charleston, SC .14b .14b
200 Columbia, SC .14 .15
208 Middle Tennessee .17 .17
?12 Austin-Waco, TX .12 .13
214 Corpus Christi-Victoria, TX .11 .14
215 Metropolitan Dallas-
Forth Worth, TX .17 .19
216 Metropolitan Houston-
Galveston, TX .26 .27
217 Metropolitan San Antonio, TX .15 .16
220 Wasatch Front, UT .15 .16
223 Hampton Roads, VA .15 .15
225 State Capital, VA .17 .20
229 Puget Sound, WA .13 .14
230 South Central Washington .15C .15
239 Southeastern Wisconsin .23 .25
240 Southern Wisconsin .12 .13
The fourth highest average value over the 3-year period (1975-77) was used
unless the difference between the third and fourth highest values exceeded
.01 ppm (20 mg/m3), in which case the average of the third and fourth highest
value was used.
Estimated value since no data are available for these areas.
c Daily values are not available, thus these values represent the hourly values.
SOURCE: Monitoring and Reports Branch, Monitoring and Data Analysis Division,
OAQPS, EPA
"Note: These are only estimates of design values for use in this analysis.
Actual values which will be used in the determination of attainment
and for planning purposes will be calculated by the state and local agencies
based on the guidance issued by EPA.
A-3
-------
APPENDIX B
MOBILE SOURCE EMISSION FACTORS
-------
APPENDIX B
MOBILE SOURCE EMISSION FACTORS
This appendix summarizes the mobile source emission factors which
serve as the basis for emission projections resulting from the Federal
Motor Vehicle Control Program as well as an inspection/maintenance
program. Table B-l presents the non-methane hydrocarbon (NMHC) emission
factors for various classes of vehicles for base years 1975 and 1987.
The 1987 emission factors reflect the emission standards mandated by the
Clean Air Act Amendments of 1977.
In order to estimate the change in mobile source emissions for an area
between 1975 and 1987, the following equation is used:
1987 Emissions = 1975 Emissions X ^75 Emission Factor X (1 + Annual Growth Rate)12
Estimates have also been made regarding the effectiveness of an
inspection/maintenance program in reducing mobile source emissions through
overall improvement in fleet maintenance. Emission factors for light duty
vehicles for various I/M scenarios are summarized in Table B-2. As can be
seen, the effectiveness of an I/M program depends on the stringency level
and the extent of mechanic training. For a 30 to 40 percent stringency
level, an I/M program can reduce emissions in 1987 by 25 to 44 percent
depending on whether there is mechanic training. For purposes of this
study, an emission reduction of 30 percent was assumed to reflect a mid-range
of the estimates.
B-l
-------
Table B-l. NON-METHANE HYDROCARBON EMISSION FACTORS FOR
MOBILE SOURCES3 (grams/mile)
1975 1987
Light Duty Vehicles
(without I/M) 7.96 1.98
Light Duty Trucks
0-6000 Ibs 8.65 3.30
6000-8500 Ibs 12.17 4.71
Heavy Duty Gasoline Trucks 28.66 12.18
Heavy Duty Diesel Trucks 4.33 3.34
Motorcycles 11.01 1.36
All Modes 9.08 2.71
a Includes evaporative emissions
Source: Environmental Protection Agency, Office of Transportation and
Land Use Policy, Mobile Source Emission Factors, EPA-400/9-78-005,
March 1978.
B-2
-------
Table B-2. EFFECTIVENESS OF INSPECTION/MAINTENANCE PROGRAMS
FOR LIGHT-DUTY VEHICLES (1987 NMHC Emission Factors)
Grams/Mile
% Reduction
from Base
Base
w/o I/M)
1.98
--
30% Stringency Level*
No Mechanic
Training
1.49
25%
Mechanic
Training
1.19
40%
40% Stringency Level*
No Mechanic
Training
1.42
28%
Mechanic
Training
1.13
43%
^Stringency level is a measure of the rigor of a program based on the estimated
fraction of the vehicle population whose emissions would exceed cutpoints for
NMHC were no improvements in maintenance habits or quality of maintenance to
take place as a result of the program.
Source: Based on Appendix N to 40 CFR Part 51: ^Emission Reductions Achievable
Through Inspection and Maintenance of Light Duty Vehicles, Motorcycles.
and Light and Heavy Duty Trucks, May 1977.
B-3
-------
APPENDIX C
Analysis of Costs for Hydrocarbon Control Measures
-------
Appendix C
ANALYSIS OF COSTS FOR HYDROCARBON CONTROL MEASURES
C.I Introduction
This analysis presents the costs for selected hydrocarbon control
measures for stationary and mobile sources. The costs are derived from
numerous EPA reports and documents and represents the Agency's best
estimates of costs as of July, 1977. The sources covered by this analysis
are not the only sources of volatile organic emissions, rather they are
the sources for which cost information is readily available. Some are
described more completely than others, with the extent of coverage
depending on the availability of information for each source.
C.2. Costs for Stationary Source Control Measures
Tables C-l through C-7 summarize the control costs for selected
stationary sources. The methodology for estimating costs for most sources
involved selecting model facilities of a size or sizes considered typical
in the industry. For these model facilities, capital costs for the
control equipment were developed which included equipment costs as well
as installation costs.
The annualized costs for each model facility include direct operating
costs such as labor and materials, maintenance costs, and annualized capital
charges. This latter component accounts for depreciation, interest,
administrative overhead, property taxes, and insurance. The depreciation
and interest are computed by use of a capital recovery factor, the value of
which depends on the operating life of the device and the interest rate
(in most cases, an annual interest rate of 10 percent has been assumed).
C-l
-------
In many instances, the annualized costs also include a credit for
product, heat or steam recovery. These credits are substracted from the
costs of control so that the annualized costs included in the tables are
net costs.
The cost effectiveness of control represents the net annualized costs
divided by the annual tons of hydrocarbons removed. While cost effective-
ness serves a useful purpose as one factor in comparing control measures,
it cannot serve as the only decision-making tool. Cost effectiveness in
itself does not give any indication of the economic feasibility of
alternatives since it does not take into account the baseline economic
or financial conditions of the source or industry. However, such an
evaluation was beyond the scope of this study, so that cost effectiveness
is the only means available by which to compare control measures. None-
theless, the limitations should be recognized.
Finally, there are in many cases several control measures available
to achieve a certain level of control at the various sources. However,
this study considers costs for only one control measure at each control
level. In choosing the measure, the assumption is made that a prudent plant
manager will choose the lowest cost option on an annualized cost basis.
In addition, if a higher level of control can be achieved at a lower
annualized cost than a lower level of control, then the latter is not
considered to be a viable option. Thus some of the control measures
considered may not be as energy efficient as others or recover the end
product, but they are still the least cost options and have therefore
been chosen for inclusion in the tables.
C-2
-------
C.2.1 Oil and Gas Production, Refining, and Storage
Table C-l presents costs of controlling hydrocarbon emissions from
selected sources associated with oil and gas production, petroleum refining,
petroleum storage tanks, and natural gas and gasoline processing plants.
Exact cost figures are not yet available for these sources, but preliminary
consideration of costs indicates that the net costs will be minimal since
the costs of control will be, for the most part, offset by appreciable
savings from product recovery. In fact, much of the control equipment
is already in operation in many plants or fields, indicating that the
controls must be justifiable from a cost standpoint in many cases.
C.2.2 Gasoline Handling and Distribution Operations
The costs for controlling hydrocarbon vapor emissions from selected
gasoline handling and distribution operations are shown in Table C-2.
These operations trace the flow of gasoline and resultant hydrocarbon
vapors from the bulk terminal to the bulk plant to the service stations
and finally to the refueling of vehicles. As can be seen from the table,
relative control costs are much higher at smaller facilities as evidenced
by the significantly greater cost effectiveness numbers.
At service stations there are two sources of vapor lossthe under-
ground tanks (Stage I) and vehicle refueling (Stage II). Stage I can
be implemented alone at service stations but Stage II cannot since vapors
captured during refueling would be lost through the underground tanks that
had no control. Thus, costs are presented for Stage I alone and Stages
I and II in conjunction. The cost effectiveness numbers for Stages I and
II take into account the total amount of vapors collected during the two
stages instead of just the incremental cost of controlling Stage II emissions.
C-3
-------
LU
CD
O
h-
^C
UJ
2;
i
o
a:
»
o
Lu
O
h-
CO
0
0
.
QJ
ro
1
u
c:
QJ
S-
QJ
QJ
CL
r OJ
O S-
s- ra
4-> in
C fO
O QJ
0 S
(
in
in
QJ
QJ C
4-J > O
I/) -r- 4-J
O 4-> -^
O O <^
QJ
tjj_
LU
a
QJ
IM -
lr_ 4_> O
» in o
ro o o
c
r-
4-> 4-J O
r- m o
a. o o
ro c_} -t^-
O -^.
i U
o c
S_ GJ
C U ^S
o *"
LU
i QJ
^
(J LO
^^
*r
-
S-
QJ
O.
O
i
QJ
3^
QJ
Q
O
ro
d
r
1
1
ra OJ
o
c c
O ro
4-1 QJ
O 4J
OJ C
4-1 -
OJ ro
D E
CO
c
s:
«c
^^
OJ
z:
c
o
r-
|
O-
in
ro
CD
c
-
o
1
N«. O
i OJ
CL CL
, .
i i
OO oo
E <- <-
5) QJ QJ
oo oo O O
. >, 00
ro CO
-a o) _-o
C CO !- O OC
ro OJ OJ O Ore
o i -a i- s.
C c ro ra C
O ro O OJ Ol OOO
r- c - 3C C C-i-
4-1 OJ C !- -1-4-1
O 4-J ro QJ 4J 4-J o
QJ C _C S- ro ra O)
4-1 T- O ra O O 4-J
OJ ro OJ i r r OJ
Q E S U- U- U_Q
1 II III
^f
c -a- co
z: ~-^ <_*
i § s
z z: z
r- 00
^
>^ ro
ro "O
-O J3
JD
O
0 0
0 0
O
0
OJ O
CO OJ
00
00 S-
(/) OJ O
OJ O 4J
i- s_ ro
S- 3 TJ J-
O) O C ro
C CO ro CL
t- OJ
<4- 00 C 00
OJ 3 . i-
Oi O ro S-
QJ S- Q)
3 (O ro
OJ i 00 S
r r 00 OJ
O 4-1
S- 0 0 00
4-1 oo o ie
oj >- s_ 3
D- S: Cl-
OU
CO
CL
1
00
if- OJ
o
C 10
OJ O 00
O -r- C
C 4-1 0)
rt3 ro -o
c: s_ c
QJ QJ O
4J CO
C !- I
i- 0 C
ro CO
E >-H c
i
m
CO
OJ
z
z
CTl
CTl
^
ro
-O
.a
0
o
o
o
o
00
I
00
00
Ol
c
o
"g
CL
e
3
O
ro
1
in
OJ
CL
C 00
O S-
o o
I CL
C ro
c
QJ
4-1 <
oo .a
3 ro
XI 00
E £
o -o
1
CO
c
s:
z
z
Ol
-o
c
3
urnaro
4-*
C
in
in
QJ
8
S-
O-
f
CNJ
CL
r
^
C C
0 *
S*o
00 O
s_
£" 0)
ro C
C 4J
O Q- O 4-1 CT
O QJ OJ QJ O)
<_> COS Q 4-1
1 1 1
CO
c
s
§
z
CTl
oo
4-1
C
r ro
ro
S- 0-
do.
ro C
C !-
00
T3 00
C 0)
ra O
0
00 S-
ro CL
Ol
QJ
i C
££
3 O
4-1 00
ro ra
Z 01
^}-
S-
OJ
0-1 >
-t o
-M S
c: s_
g +j
fa. O
₯1
OJ i.
O.
O E
!- 5
4-1 S_
C <»-
O
O OO
cn
OJ C
-C !-
4-J >
ra
«H 00
O
u
§4-1
a>
oo
OJ If-
O 1-
C 0
OO OJ
ro t
E
E'5
E oo
OJ 00
S- 0
ra o
00 I
4-1 O
00 i-
0 4-1
0 C
o
4-1 O
t3
Ol OJ O)
C -r- >
I r OJ
> O) «
ro -Q f*-
00 a)
oo JD
OJ 00
C 4-1 !-
OJ "~ 4-1
ra S-
O OJ "
r- > C
r- 0 4-1
00 T3
OJ "O
oo C ro
OJ S
f Q £2
4-1 c >-<
C -V
QJ
S- 4-1
ro O 00
Q- C T3
« OJ OJ
>1 S- -
t- ro 4-
OJ
> 00 S-
O OJ O
O S-
QJ 3 00
S- 00 4-1
ro C
4-1 OJ ro
3 ErQ.
o "o >,
S- S- C
CL 4J ra
s- o E
O O C
OJ *~
4-1 OJ 00 C
r-i OJ O
o -Q jr -i-
OJ ro 4-1 4-1
!- i ro
O "- S- S-
ro O OJ
QJ > If- CL
"O ra O
3 00
r 4-1 4-1 C
O O 00 T-
C C O
r- 0 >>
II -0
00 4-1 ra
4J 0 QJ
00 ct ra S-
O X *
o z: LU re
OJ CO
o
C;
S
QJ
4-1
t J=
QJ 4-1
3
4-1
0 >,
J= S-
ra
o >
S- t
4J r
|*
i- in
ro f
JD ro
QJ
QJ in
5 >>
C
01 jj5
^ -^
s- s-
QJ D-
O QJ
0 >
.*-£
0 10
4-1 T)
00 C
O ra
i.
QJ OJ
t~ >
4-1 O
S_ 00
O i
ro
S- OJ
QJ VI
00
c >>
OJ S-
T3 ro
c -a
8 0
OJ <1)
O 00
ro
f- f-
S- O
3
00 00
00
(O OJ
C
^ OJ
0? >
4-1
L O
OJ OJ
> H-fJ
i- if-o
OJ QJ3
o no
OJ 4-10
S- 00i_
OCL-
OJ O
l/l 4J ~O
i- ra -DQJ
OJ 00 C&-
> c roO
O OJ 4-1
O T3 >4J1
C S-
If- O OJOJ
0 0 >J=
O O4-1
S_ ra O
OJ4-
01 f- S-O
c o
r- 4JQJ
4-1 4-> OS-
ro oo 33
O O-ooo
1 O OOO
i[ t-OJ
QJ CLS-
O "O CL
4-1 3 QJ
>, O CO
t C QJO-
C "--Cro
0 >
4-"a
c o coj
i- C (O3
0 !_
4-1 00 C4J
!_ OJ O
O- -O4J.C
00 O) 3
cn 4-i-cro
C ra rjjc
- E s-ra
ro 4-J c>,
00 00 O4-1
QJ-i-r-
OJ OOO
oo oo coo
-C -C EO)
r 1 LU>
^t LOLD
C-4
-------
CO
z.
O
Qi
LU
Q.
O
CQ
Ii
C£.
oo
O
CO
CJ3
Q
o:
o
oo
LU
o:
o
I/O
oo
o
CNJ
O)
QJ
C
QJ
QJ
4-
QJ
r- QJ
0 t_
i- i
C ft
0 OJ
0 SI
U1
QJ
Qj'c
*-» > C
0 +J ^,
4-
LiJ
"S__
'fO o O
c -
c
2 -MCD
Q- O C
O C
S- OJ
-t-> -i
c: u »«
o -
O 4-
UJ
s-
r- QJ
U t/1
U.
c
o
4J
to
OJ
a
O
5
4-
QJ
(J
£
CNJ
a.
CNJ
0
(U 4J
cn a.
£ 0
0) -O
CC fO
o
1
in
o
o
o
in
Cn
-o
-Q
^
v>
c
1
0)
c
1
,_"
£^
"
c c
0 O
-M -M
OJ QJ
cn cn
QJ (U
pf Qf
oj in
-. co"
CO
CO OO
<£ CO
cn cn
33 XJ
Q. a.
cn 01
o o
§ §
O 0
m o
oj in
£
>
E
fc 1=
h- -^
3 Wl
C (/")
» O.
o o
t/) J
fO
C3 1
c c
0 0
QJ fc
L. S-
4- «f-
OJ QJ
^3- CO
cn o
cn oj
i OJ
fS S
i esj
t
CO CO
o -o
a. Q-
cn cn
§o
0
o o
o o
in o
oj in
g
I?
3 m
O - -r-
| U_ Lt_
1
oa
|
a.
^
Q. Q.
0 O
!-» +J
O ^ QJ O ^ QJ
+J T- (J +J -r- (J
4- C 4- C
C to C rtj
O T3 i OX)
i- QJ to - OJ tg
ul CTICQ CO O> -Q
QJ QJ S- OJ QJ S_
> E O > E O
C .0 Q. C i Q.
SD tO O 3 tO
Wl > (_) W >
O O O O
oj in oj o
i CO i OJ
r- co cn oo
o m oj oo
oo co vo o
O «j O r-T
oj in
s ds s
a x
cn cn
g g
0 O
in o
CXI
*-» ^~
C t
(O T-
O.
x:
^- (O
3 r
co a.
to
0)
c ex
i- O
o
QJ QJ
r- 3"--
^ S"
4-> cn QJ
t/i I- cn
QJ 13
QJ X) *->
& '
c .
8. ^ tJ
« 'o >>
> u oo
in o O
1^. IO OJ
00 OJ f
r^ in oo
0 ^ ^
^r co cn
*3- CO O
0 0 O
cn cn cn
1 !.» !«
O tfl O E OB
0 CX 0 3 05
O E O O. O O-
o ex o en o CM
CM *£> CM i
1
w oj -a
C 4^ S
to aj
M i. -
-o a> ui
C Q)
3 u cn
o
QJ
C C C i
1
,;
in
a.
r~
VD
s
^
in
<
cn
cn
_^2
3
C
QJ
C
o
ro
ta
in
£ ^,
ving
cond
pear cap
differen
two sys
sist)
ican
he o
vac
not
ost
. O (J
I
? QJ VI
!§
u -g
O t QJ
t O (->
4- S- 01
-*-> >>
O C in
C O
'i £'
o .
1- 1^ - 4-
O (O CTif
QJ O O XI
0 >, r- J^ OJ
3 m c c ro
f r -C O U
O OU'i-
C C 0> T3
T- o -M *-> C
c -
«» -a a> QJ
** 10 S- U QJ
OJ
(J i n
o -^
S£
C-5
-------
C2.3 Surface Coating Operations
Table C-3 summarizes the costs for controlling volatile organic
emissions from selected surface coating operations. Except for the
primer application area and curing oven in automobile and light duty
truck assembly plants, which utilize a process change in applying the
primer, the control measures considered for these sources are add-on
technology to destroy or recover organic compounds from exhaust gases.
Thermal incineration, which destroys organic emissions, is considered
for all of these processes instead of catalytic incineration because of
the variability in the use of the latter. Catalytic incineration is
limited to a more restricted range of applications as a result of problems
with catalyst deactivation, coating with particulates, poisoning, and type
of fuel used (ref. 12, pp. 54-55). Where applicable, though, catalytic
units offer the potential of significantly lower fuel consumption and smaller,
light-weight units. However, it was beyond the scope of this study to
judge the effectiveness of the option in all of the processes considered,
so thermal incineration was used since its use is not as limited, even
though the costs may be higher.
Costs for incineration include savings from the use of primary heat
recovery. Primary heat recovery involves the use of incinerator exhaust
to preheat incinerator inlet air instead of using expensive auxiliary
fuel. Further heat recovery called secondary heat recovery, is possible
in some applications where incinerator exhaust from the primary heat recovery
stage (or from the incinerator directly if there is no primary heat recovery)
replaces energy usage elsewhere in the plant. This energy can be used for
C-6
-------
g
u.
CO
UJ
C£-
Z3
a
OJ
u
c:
i-
OJ
01
01
« OJ
o i.
i- 3
4-> 1/1
O 0)
o s;
to
c c
o> o
*-» > +J
tO '! "X,
O +J *^J-
C_j (j .. .
£
UJ
s
r- +> O
i VI O
ro O O
13 CJ *>=»-
C
!-> -M"C
cx o cr
(O CJ **=»-
(_> -
r U
O C
i_ at
o -
o <+-
UJ
r- Q.
(-) tO
cr
o
rC
at
cx
o
s
o
cu
Lf-
O)
O
L.
e
to
CM
LO
CX
LO
CO CM CO CM
CsJ ro OJ ro
LOLOjLOLO
cx a. cx a
LO LO LO LO
1
C
C +J
O re
-t-> a
4- CX C
s.
E
. -^
i 4-
UJ O
O
cf
8
LO
CSi
O
8
LO
fj
en
E E
CO CO
o o
X X
CO LO
OJ
>> (0
4-> CU
n L.
"O «O
*-> C C
-C o at
01 - >
r- (0
LJ O
"O >)>- C
ro .a cx j»
E CX 13
(U QJ ra U
i LO
!a "" aj *~
o o i. £
CC 4-> 1 1
O C_
t/i c
ra
C re
V*7
O O
^ en
« c\
O O
CN
o o
o c
CM r-
LO LO
CO CTi
E
CO
o
X
CO
OJ
c
o
C '-M
O rd
+j at
a. c
S_ T-
0 0
irt C
a -r-
fd
C (C
s- a
0 ^
^1- -cf
CO CO
t 1
cx cx
LO LO
c c
o o
+3 -M
ro rO
CO OJ
So
o
CM CO
^£
SO
0
LO O
2r
LT) LT
CO CJi
E
«£
'O
O c
X
CO
ai
*
cz
o
+-1
u
cx
cx
+->
rei
o
o
1
CO ^T
OJ
CO O
CO CO
OJ
LO LO
en en
L^ E
(/) U
I/t
o
CD CO
r O
X
X
CO OJ
at
0
en
£1
3
o
0
t
' CO i CO CO
CO CM CO OJ .
1 1 ( 1 1
CM CM OJ CNJ CO
CX CX CX CX O_
LO LO LO UD LO
cr c c
00 0
C: -M c +j 4J
O ra. O rei ro
CX C CX C C.
O O O O O
LO C I/I C C
T3 - T3 '- '-
"0 ra
QE 0 E E
JQ 1_ jzi C S-
i. ai, s- at tu
O h- (_> t t
LO O
CM r
CM CO
CM <
V? >s
O LO
(E
to
CO
o
X
o
LO
LO O LO
o o i en
i
O O
CM
O
X
LO
r
en
c
ra
o
c
LJ
Is:
<
LO
cn
u
LO
CO
o
X
LO
1
u
a>
cr
uT
c:
ai
cn o
fO S-
o ai v>
O -M ro
§a>
1_
<~ O rei
o
O 1
1 1
cx cx
LO LO
c
o
c +->
o ra
+-> CD
cx c:
0 U
to c:
ra
C ro
t- at
V ^
LO LO
CO CO
O LO
tO LO
CM IT)
CO r-
O LO
O
o
X
LO
"O
(t)
ro
U
ra
cn
c c: c
-M -° >
S4-" O
ra
CJ O cn
i C
-Q < -a
ra
U. 1
LO CM CO C'
CM CM . .
I 1 1 CO t
LO LO CO LO CO
CX CX CX CX CX 1
' ro
LD LO 4O « r~~
c: c:
0, 0
en
c
-t-*
d 4-> +J CT O
O *«
>- S-
t -r-
a. c
S- "-
C i rd O
i- ! O I
o o cj <->
CO C C OJ
a -t- -i- e £
c 'rei
51
s- ai
CJ 1
0 LO
«d- co
0 LO
LO LO
O O
CM in
CO i
O LO
cn cn
E
U
to
CO
CD
X
LO
.1
M
O
(J
1
ro i- -Q
ai
_<^
I
LO
en
s
c:
en
(O
Q- S-
QJ
Q- *->
O rO
, , .
O 0
CO r
CM cn
CM CO
O 0
CO CO
LO LO
LO '^)
CO 33
S- i-
S- - i-
^ -E
O 0
LO LO
cn r
r- LO
LO
a>
u
(0
'cx
a.
ra
ai
E1
fO
1
i CM co ^ LO LO r .
CO CM
CO
a.
CX CTl
" CO
s
£
o cn c: cn
CO E $_ C
£- -4-* Q J-i
x: o at o
cn u 4-> u
X 3
CO LO CD CM
LO CM rj" LD
CM ^f '
co cn r co
LO OJ CO LO
r LO
LO LO O O
CO CO CO CO
s- s-
S- \ S- ^v
E CT E CT
C
1- O -!->
3 CJ fO
u_ o
«0 S- O. j
MO. -
OJ tO Q
CO
i <
S- -r- r-
> -i- i
i OJ CO
C-7
-------
process heat requirements or for plant heating. However, credits for
secondary heat recovery have not been included in this analysis since
the amount of energy that a plant can recover and use depends on the
individual circumstances of the plant (ref. 12, p. 45).
Carbon adsorption, the other add-on technology considered, separates
and recovers organic vapors from the exhaust stream but is not as efficient
as incineration in general. The annualized costs for this control measure
includes a credit for the recovered solvent at its fuel value, which is
lower than the solvent's market value. Nonetheless, the market value of
the recovered solvent has not been used in general because reuse of the
solvent may not be feasible due to mixture of solvents or breakdown of
single solvents. Distillation is possible, but the complexity and cost are
so variable that it is difficult to generalize (ref. 12, p. 32). Thus,
the fuel value of the solvent has been used as a conservative figure.
C.2.4 Graphic Arts Processes
Costs of control measures for selected graphic arts processes are
tabulated in Table C-4. Once again, the techniques considered are
incineration with primary heat recovery and carbon adsorption with solvent
recovery credited at the fuel value of the solvent.
C.2.5 Degreasing, Dry Cleaning, and Cutback Asphalt Paying
Table C-5 presents control costs for selected sources of evaporative
emissions of organic vapors. One such category of sources is organic
solvent metal cleaning, which includes cold cleaners, open top vapor
degreasers, and conveyorized degreasers. Dry cleaning operations include
neighborhood and industrial petroleum solvent plants as well as coin-op,
C-8
-------
00
r
C£.
cu
(J
c
cu
S-
cu
cu
CtL
i CU
o s~
i- 13
4-> VI
c re
o cu
<_> E:
CO
CO
cu
ai'c
£>£
O 4-> -~>
O O v>
cu .
UJ
o
cu
N --.
rO CO C
3 0**
C O .
C
o o
cu .a
s_ s_
Q U
o o
CM CM
csf ooc
o o
O LO
t-~ 00
0 0
o o
CM r--.
1 1
uo o
en 01
a
CO
§
o
O
[ i
ro
cu
c
* c
o o
i- 4->
CL
CU i.
E 0
4- ro
4-> C
CU -Q
!_ i.
r- re
a <_>
0 0
o
M OO
0 0
LO O
00 r
0 0
CD CD
i OO
uo o
CTi Ol
E
O
o
o
o
CM
01
c
4J
a_
4j
CU
CO
o
JD
CU
3
UO
CL
f
,
fe fe
re re
S- I.
cu cu
c c
u u
c c
cu cu
re re
0 O
c O
CM. OO
O O
LO O
00 r
O O
O O
LO «J-
i OO
LO O
en 01
CO
o
8
o
CM
CO
CO
CU
CL
CU
4-> CD
CU C
1 'r-
4O
fl T
0) t.
3 a.
i CM 00 ^J-
a cu
cu :>
-C o
o
cu
g
"8
CU
g
(J
CU
S-
cu
o
CO
cu
3
r CM 00
C-9
-------
Of
ft
3;
LU
>
o;
o
a-
*/>
Ǥ *~*
ro .*-*
P -P O
r- CO O
a. o o
ro 0*0-
O "
o c
i- QJ
c a **
0 T-
0 <4-
UJ
^.
r- QJ
r N
U CO
ra
o
_
i-
QJ
0.
o
_
Q
U
c
4-
QJ
CJ
o
CO
1
^
CL
^
r-T
P
1
U
ro
*+
QJ
cn
c
s-
o
CM
in
o
m
CM
in
JT -a
-P QJ
5 co
>, CO
-P LO
t- ro
*" >)
o <
4- 'ro
U
QJ -
D"> C
ro rO t-
C -C QJ
r- (J >
ro Qj o
S- E U
Q
^_^
in
CM
10
CM
u>
o
ro
4J
+-> c
c QJ
QJ >
i > >~
rO i O
-M LO O CO
QJ c: co
E _0 ^
4_> 4J CO 4-J -r-
C ro I. M-
QJ S_ QJ M-
> QJ C - 4->
t CL ro -P ra
O O QJ ra >
in r r o
CD C_> O >
(J C >
c c ? cn
ra rO O O >|
en QJ O -J 31
,_
*
CL
r~T
CL>
i
-C C
o en
cu cu cu
> 4-> -o
o ro
0 S- TJ
cu cu
i cn to
ra > o
3 S- r
c <+- o
ra cu c
s: ce QJ
^_ .. ^
LO LO 1
CO OJ
CO
o * to
O CO CO
CO
ro LO o
O ^3 to
1
o LO r-.
CO «3- CTi
ra
o
O-
H-
to
CU
CO
ra
cu
s-
cn
cu
-o
s-
o
CL
^
CL
O
C
cu
o.
o
r^
^
o.
^T
S- i.
cu cu
r- 'i
O U
cu S
4-^ 4->
ro ro
S- !-
cu cu
cn cn
s- s-
4- 14-
CU CU
_,
0 0
OJ r
LO O
ro LO
CO
LO LO
CO 1
0 0
LO LO
ro
o
ca-
n-
to
i.
cu
to
ra
cu
s-
cn
cu
-o
cu
N
'<~ "O
0 -r- C£
>> ro i
cu s- to
> o to
ceo
O O L.
o s: o
o a i i i
CO
3-
o.
00
o
ra
c s_
o ra
f- CL
4J CU
Q. CO
s.
O r
co ra
-O cn
ra 3
<*-
1 5
t_ c
ra cu
0 0
LO O O
ro o ro
LO O LO
ro i <*
co o co
to LO i
i OJ
o o
CO cn
to to
O CU U
r- C CO
4-> ro -r-
ra to cu E
S- 4-> i
cu c o
Q. ro r-»
o^-o^
CL O T- -V
Cn O 4-> U
C E f~ CO 13
C QJ 0 - 1
fO JD S- i- *X
QJ O _C QJ QJ J
s- cn >, -P o
QJ QJ -O '-
>, a. z; LJ_
0
T
**
o.
CO
S-
o
U
CO
c £
ro
Q- i
r- T- u
ra -P 13
- co s:
S- _J
co S- OJ h-
=3 QJ -P O
CD Lu
S- « *
O l
1
CO
T
**
o.
CO
c
o
-P
Q.
S-
o
l/l
-a
c
s
S-
ra
C_J
LC
L?'
CCi
rx>
h*.
UT)
c
o
D-
o
CO
a
j^
o
JD
s_
rO
O
in
in
0
CTi
CM
in
c:
o
O-
s-
o
CO
-a
c:
o
xi
s_
fO
f ^
in
CO
CM
CO
CL
,_r
,
^
lf-
O -P
C ro
o -c:
r- CL
4-> CO
-i- T?
P QJ
CO T-
JD 4
3 T
CO i/l
C
s
t
y
^~
LO
r»«.
in
QJ
OJ +J c: c
>^ f- r i
(j; n-j ,_ ,_
O H-- ro ra
O CO CX O S_
. -P o i_ -P
x: cr i QJ co
u ro cr E 3
1- r- T- E ~D
cu CL o o c:
CL. o o <
i '
^
^
^
cn
c
aj
CL
£
rO
CL
rO
^
U
5
-P
r)
i CM CO
C
rO
CJ
CO M-
cn -i-
c: c:
r- cn
ra co
CO C
QJ CO
QJ QJ
P -P
U S-
r- U
-a c
c o
i o
CO -P
QJ r-
co ra
jz: CL
P CO
QJ *°
S- "O
ra QJ
D- «-
r-
>> _l
CU r^
0 M-
U 0
cu
!- to
cu
4-> O.
ro >,
CU +J
t s
O 4->
4-> CU
3 +->
"8 c
S- Qj
CL QJ
S- -P
O QJ
P QJ
r- CJ
QJ QJ
S- S-
U QJ
QJ M-
^ -o
U QJ
C U
to CL
4->
CO OJ
O _C
(_j |
CM
C-10
-------
commercial, and industrial perchloroethylene solvent plants. Finally,
the control measure for cutback asphalt paving entails the substitution
of emulsified asphalt in order to eliminate evaporative emissions.
C.2.6 Rubber Products Manufacture
In Table C-6 costs are summarized for selected processes in the
manufacture of various rubber products. The manufacture of tires and
inner tubes represents the largest source of emissions in this category
with over 50 percent of the total. Thermal incineration and carbon
adsorption are the primary control measures for most sources. However,
for the sole attachment operation in shoe manufacturing, substitution of
hot melt adhesives accomplishes the control of emissions at a lower cost
than the add-on technologies.
C.2.7 Chemical Manufacturing Processes
Table C-7 indicates the costs for controlling hydrocarbon emissions
from selected chemical manufacturing processes. For all but one process,
two sets of costs and control measures are given for each level of control
efficiency. The difference depends on whether or not the heat from the
control device is recovered and utilized. If heat is recovered, equipment
has to be added to utilize the heat for process requirements or to generate
steam for process requirements. For this option to be feasible, the higher
capital costs have to be offset by the savings for the heat or steam recovery.
This means that the plant has to have uses for the recovered heat or steam
either for in-plant uses or for other plants nearby. Thus, the special
circumstances of individual plants dictate the viability of heat or steam
C-ll
-------
8
E
_J
O
0)
u
c
QJ
i-
QJ
QJ
DC
i OJ
O S*.
O QJ
O E
in
in
0
VI *i- -P
O -P \
o u *»<*
QJ .
4-
UJ
-o
QJ
N
r -P O
fd in o
3 o o
c o &
C ^
'fd -p o
-P tn o
r- O C
fd <
o
. o
o c
i- QJ
C U 5^
c3 £
UJ
I Q
O I/O
fd
U-
c
o
M
(O
S-
g
o
-o
QJ
-p
o
QJ
t
c
u
S-
c
CO
CL CL CL
cn cn cn
c
o
-l_>
fd
i_ QJ
QJ C
o o
u c c:
0 T-
cn 'i
C -P t
i- 0- 10
« o E
O in QJ
u7 «< 1
*t CD LO
Z ,
vn o O
ro r^. LO
CM to
o 'o o
CO O O
« r*^ LO
OOO
co cn cn
'fd u o
ooo
O CD O
O O CD
O r LO
CM "
l/l
QJ
P fd QJ
l/l Q) i-
fd s- fd
uj E cn
S- >> c:
O fd S- -r-
- u_ QJ -C
p > in
QJ ^ 0
.C c: u c:
M fO QJ -r-
C h- Ct: LL_
CO 1 1 >
,
CO
^J-
CL
cn
C
0
C T-
0 4->
Adsorpt
Inciner
o in
cn iX3
CO OJ
o o
CO LO
CO CO
0 0
O 0
CO OJ
o o
co cn
E
**-
o
o
CD
o
in
in
i!
3 Ol
h- C
S- -P
QJ C
C QJ
U
c o
in _d
QJ fd
i- U_
t i
CM
CO
I CO CM
=f QJ
- -Q
CL fd
h-
cn
c
O
-p
fa
is
c
o
c c
0 T-
Adsorpt
Thermal
LO LO
CO <3-
CO CM-
0 O
0 0
m r^.
0 O
o o
CM m
CO r
O 0
co cn
E
4-
O
in
O
0
CD
in
01
c
*>»
fd
h.
in
QJ
t.
QJ
QJ
CD
I
CO
o
CL
cn
C
0
-p
inera
o
c:
1 Thermal
o
o
cn
OJ
o
o
CO
OJ
o
o
r-.
OJ
0
cn
u
in
0
o
0
CM
cn
0
L.
QJ
CL
O
Ol
C
s-
( J '
1
OJ
CM
«*
cn
c
o
p
ITS
| Inciner
LO
r*.
j
o
o
o
o
CM
O
Ol
in
o
CD
0
LO
CD
C
P
C
QJ
E
QJ
(J
-a
fd
QJ
ndertr
ID
I
CM ro
ID O -
i i cn
^- *3- QJ
.Q
CL a. to
cn cn '
0
+j
fd
S-
QJ
O
C C
o -
p »
S- E
0 S-
in QJ
< I
o o
CM O
CM cn
CM
CO CM
O 0
CTi O
r CO
CM
0 0
0 0
^ CM
Oi cn
^ ^
o u
w in
O O
r- LO
CM
C C
OV" CO
C >> T-
«- fd -P
S- -P S- (0
fa c Q. s_
OJ QJ m QJ
-P S QJ 0
o o >
O f- Ol
Lu l_ in c
QJ QJ -r-
i. Jl JZ TJ
QJ JD. -a t
-Q 3 "=E O
(~1 ^ .^ , ^-
o: i i
CO
OJ
CT*
CL
cn
a>
QJ
1
o
O 1-
c: -Q
O 3
P O
o. in
S-
o-o
in c
^
CO
LO
o
in
1
o
cn
E
M-
O
cn
O
O
LO
in
QJ
in
in
OJ
U
O
CL
t-
O
^_)
ra
E
fd
o
QJ
a:
^r
!
y CO
cf
CL
cn
c
o
-p
(O
fc
c
o
(d
OJ
i
LO
CM
O
ID
CO
O
O
CM
O
cn
u
in
O
S
LO
Ol
c:
£
'aj 01
CQ ^C
-o -P
c c:
fd QJ
E
Q) QJ
in o
o
n: o
S- S-
QJ -Q
JD U_
C£ 1
LO
CO CO OJ
O O l£>
CL CL CL
cn cn cn
c c
o o
p -P
fd fa
S- £-.
QJ QJ
C C
O U
C C C
- -r- O
Id 'fa "o.
E g fe
QJ QJ CO
iS i£ S
OOO
O O CM
cn 01 CM
CM CM
CV CM (v,
OOO
O O cn
CO CO r-
CM OJ
CO
ooo
OOO
CM CM
ooo
cn cn cn
o u o
in in i/)
OOO
O CD O
CD 0 0
LO LO i
CM OJ
m
O
to cn
QJ >,
-Q fd
-Q i_
3 CL
CC. in
"O QJ
QJ Ol >
P C 01 *-
fa - c in
U -O *r- QJ
S- O 3 "O
^ y cj ^
LL. 1 1 1
U3
ro ^J
O OJ
CL Q.
cn cn
c:
o
4-}
fa
QJ
c
CJ
c c:
r- O
r~ -P
S- O
QJ m
.c -a
r «=C
O 0
O OJ
cn CM
CM
CM co
O 0
O cn
CO r
CM
O 0
0 CD
CM
O O
cn cn
u u
in in
0 0
o o
0 0
LO r
CM
O)
c
fd
QJ
CO CD
C
'% 1
ra
CL. QJ
01 >
in -i m
a! 2 I
_*: o -o
« s: <
fO
CD 1 i
I
fv%
O
CL
CT?
c:
C)
4J
fU
t.
a>
c:
c
Thermal
S
en
CM
CM
0
CD
CO
CM
0
CD
CM
O
cn
u
o
o
o
LO
OJ
Ol
J-
fO
-D
QJ
S-
in
3
£ c1
J- -r-
QJ t-
4 3
C <_)
o
^ 1
CO
CM CO
ID O
CL CL
cn cn
c
0
'^
OJ
c:
c: c
O r-
P r
CL fO
S- E
O s-
m QJ
-------
u
C
CU
cu
H-
CU
CC.
r CU
0 i-
i. 3
4-J in
C 10
o cu
CJ £:
OO
LU
OO
00
LU
CJ
0
a:
Q-
CD in
« O
a: o £
=5 GJ
_1
ct
z:
Lu
5 "S
Tsl ^^
O T- 4-> O
LU f in O
r 1C O O
LU C
_J C
LU «t
OO
ct
o
LL.
00 r
LU ITS -
o: 4_> 4-> o
o - in O
OO O. O O
<£ ro CJ-faO-
LU O * -
^
1
o:
Z >1
o o
° *o cu
U. S- -r-
O 4-> O &3
C i-
oo o
J3 I N
t U OO
rO
U-
c
o
4->
ro
CU
O.
O
a
o
cu
4-
<
CJ
S-
o
00
o
ro i
. CO
O r 1
;> LU
i CX
i CX
01
" C
S- S- -r-
O O J3
£ "E £ 2 jf
CU ro J3 CU U
c J3 c in
r- S_ 3 -i-
CJ CU S_ (JO
C t O C -
i- -r- I/I . 4->
o in
r- -0 0 . 3
Ecu £> E o
4-» in s-
O> in 3 CU T3
1 3 u 1 <0
o in
OJ r
ro f
r- LO
o r~.
OJ O
~
0 0
in o
o in
OJ i
cr> en
ro
O
x >-
o.
O I
in
ro
T3
S- C
o o
JZ 4->
0 IO
r- C
a i-
s-
.>.
J= X
+-> o
1 1 1 *^*t
1
i
OJ
CJ
CO
O 2C
r O.
l CL
L.
S- r S.
O'"- O
re j5 10
CU 4-* CU
C 10 C
r- CU T-
r- CU -r-
4J
I W I
§S JO
S- S_
0) T3 -
a_
o
j;
r
C
*>>
i-
^^
OJ
t
CsJ
" ro
10 1
CO
1
00'
:> uj
r O.
r CX
c
o
4J
IO
i.
a>
c
o
c
1
o
4_>
,*>
IO
IO
CJ
o
OJ
m
co
ro
in
00
in
co
ro
O
X a.
H-
o
o
^_>
c
rO
"a.
c
X 10
o -o
ai 'x
C 0
^ £
LU 1
ro
c
o
s- £
o cu
4-> C
£ i
C !-
cu
en
IO L.
cu cu
4J J=
00 1
* ~«
LO LO
*^-X |
,- * ^"
i CM
LO O
^J- ro
LO LO
CO CO
V)
0
XQ.
o
r
^
C
r_
a.
o
ra
T3
X
o
d
cu
x"
o
1
^-
ro
ro
O 1
> 00
u.
^ .
r- Q.
in
R)
D) s-
0
C IO
cu s-
> 01
01 T-
in o
3 C
O >-
JZ
s- s- 'g
i C !_
r- S. 0)
O 3 JZ
CO Jd r
in o
OJ i
""
IO
5" OJ
«a- oj
00 OJ
co en
o o
CO CO
ro
o
x a.
1
0
LO
^
>.
IO
O) 4->
D ia
>> 0
JZ
OJ S_
r >
§ '-
t. 00
o
U. 1
'^-
in
LO
, OJ
O J
Ltl
r .
r CX
I. L.
0 O
IO IO
i- S-
13
r-
X
o
"S
QJ
X
1
in
en
C 4->
10 E
in ffl
4-> "Q- J
01 a.
C 3
01
10 01
O 3
r- It-
1
i- 10
in in
CU 10
in 01
Ol
5 ^
C W
OJ 03
s- S
03
O_ (/I
> - -
S_ -r-
O) O Z3
O C
U , OJ
4- S- 4-> U
+J Q> OJ S_
CU O -C O.
^ E 0
O OJ S- V)
s- s_ i- o ro
0 q-
f-1 E >>
4-> V) TO TO S-
U OJ CU OJ CU
3 3 -C 4J >
"O ' W 4-J O
O TO i- i- O
L. > O t. -0 OJ
Cu o OJ i- cn
vd E M~ s. c:
S_ r^- TO O -M *r-
O CT> 0) -t-> TO r
H_ ^- 4^ .. ,
.(-> 0 OJ U
- 4_> i_ i_ TO i- a>
-a o u > O S-
ai -o *+- *+-
l_ OJ CU Ol OJ
u +j +j -o c 4J -a
ro -r- 3 -r- -r- rs
CU "O "D r 4J XJ i
o ex cu o ro cu o
3 3 s- c: cu s- c
i o >- -c u -
u cu
r- TO cu o cu cu o
-o c: T3 ~o c
w t/> rs 3 rs
+J 4-> i trt i r 1/1
in t/> o QJ o O <1J
o o c o c c: o
O C_> " ' .O t < » " Q
r oj ro ^3~ LO us r»«.
C-13
-------
recovery. Existing plants may not have the flexibility necessary to
incorporate this option. As a result, some plants may have to employ
incineration without any heat or steam recovery, making the annualized
costs of control higher than would otherwise be the case.
C-14
-------
REFERENCES FOR APPENDIX C
1. "Air Pollution Control Technology Applicable to 26 Sources of Volatile
Organic Compounds", Office of Air Quality Planning and Standards,
U.S. Environmental Protection Agency, May 27, 1977.
2. Radian Corporation, "Background Information on Hydrocarbon Emissions
from Marine Terminal Operations", Vol. I, EPA-450/3-76-038,
November, 1976.
3. "Control of Hydrocarbons from Tank Truck Gasoline Loading Terminals",
Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, May, 1977, Draft.
4. "Control of Volatile Organic Emissions from Gasoline Bulk Plants",
guideline document in preparation by Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency.
5. Based on internal EPA analysis of comments submitted in response to
November 1, 1976, Federal Register notice concerning the proposal
of Stage II vapor recovery regulations.
6. "Control of Volatile Organic Emissions from Existing Stationary Sources--
Volume II: Surface Coating of Cans, Coils, Paper, Fabrics,
Automobiles, and Ligh-Duty Trucks", Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency, EPA-450/2-77-008,
May, 1977.
7. "Control of Volatile Organic Emissions from Organic Solvent Metal
Cleaning Operations", Office of Air Quality Planning and Standards,
U.S. Environmental Protection Agency, April, 1977, Draft.
8. "Control of Volatile Organic Emissions from Dry Cleaning Operations",
Office of Air Quality Planning and Standards, U.S. Environmental
Proection Agency, April, 1977, Draft.
9. Monsanto Research Corporation, "Identification and Control of Hydrocarbon
Emissions from Rubber Processing Operations", EPA Contract No.
68-02-1411, Task 17, March, 1977.
10. "Control of Organic Solvent Emissions from Adhesive Usage During Shoe
Manufacture", guideline document in preparation by Office of
Air Quality Planning and Standards, U.S. Environmental Protection
Agency.
11. Air Products and Chemicals, Inc., "Engineering and Cost Study of Air
Pollution Control for the Petrochemical Industry", Volumes 2-6,
EPA-450/3-73-006.
C-15
-------
12. "Control of Volatile Organic Emissions from Existing Stationary
SourcesVolume I: Control Methods for Surface Coating Operations",
Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, EPA-450/2-76-028, November, 1976.
13. "Control of Volatile Organic Emissions from Existing Stationary Sources
Volume V: Surface Coating of Large Appliances," Office of Air
Quality Planning and Standards, U.S. Environmental Protection
Agency, EPA-450/2-77-034, December, 1977.
14. "Control of Volatile Organic Emissions from Existing Stationary
SourcesVolume III: Surface Coating of Metal Furniture",
Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, EPA-450/2-77-032, December, 1977.
C-16
-------
Appendix D
Analysis of Costs for the Federal Motor Vehicle Control Program
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
U.S. Environmental Protection Agency
Ann Arbor, Michigan
-------
APPENDIX D
The purpose of this appendix is to present the EPA cost estimates
for the FMVCP based on the most recent information available to the EPA
technical staff. The cost of the following programs will be discussed in
detail:
Light-Duty Vehicles
Light-Duty Trucks
Heavy-Duty Vehicles
Motorcycles
Aircraft
It should be recognized that the cost estimates provided in this
document are all in 1978 dollars and are subject to future revision.
However, due to the need for these numbers in the immediate future, the
best, currently available cost estimates have been used.
Table 1 presents a summary of the FMVCP costs.
I. Light-Duty Vehicles
No estimates of the cost of the light-duty vehicle program have yet
incorporated all of the most recent information. Such an estimate will
be developed in the following discussion.
I.A. Initial Cost of Emission Control Systems
The predominant emission control system which is expected to
be in production for model year 1987 (and emission standards
of 0.41 HC, 3.4 CO, 1.0 NOx) is a three-way plus oxidation
catalyst system using feedback carburation, exhaust gas
recirculation and air injection. Some vehicles, particularly
those at inertia weights of 3,000 pounds and below with low
power to weight ratios, could be certified using current
D-l
-------
Table 1
Cost Summary
(In Billions of 1978 Dollars)
Source
I.LDV
Control
Gaseous
Emissions
for HC
Control
Unleaded
Fuel
for HC
Control
Evaporative
Emissions
for HC
Control
Altitude
for HC
Control
-0.63*-0.75
.59-.86
.272-.322
.025-.473
II.
III.
LOT
HDV
.38-. 800
.532
.30-. 43
Included
under
gaseous
emissions
IV. MC
V, AC
.046
.025
included in
LDV estimate
.016
.10-.189
*Note: the negative sign indicates a net savings
D-2
-------
oxidation catalyst technology. The initial cost to the consumer of
2
these systems are $158-162 for a vehicle using an oxidation catalyst
2
system and $269-285 for a vehicle using a three-way plus oxidation
system. Both ranges of cost are compared to uncontrolled vehicles.
To provide estimates which are on the high side, only the $285
number will be considered further. But first, the $285 figure will
be fully substantiated.
The $285 figure was generated by the EPA technical staff by combining
the costs of emission control components as provided in reference 3. Most
of the costs used by the staff were taken directly from this reference.
One important exception is the cost of an electronic control unit. A
lower cost was used than the one provided by reference 3 for three reasons.
First the EPA technical staff is aware of replacement parts (which are
normally much more expensive than OEM parts) that sell for about the
same price as that quoted by reference 3. And second, the ECU cost
estimated in reference 3 is for a unit controlling more functions than
assumed by the technical staff, and finally, the cost assumed in reference
3 is for analog components not for large scale integrated (LSI) components
as will be used in 1987.
The precise component costs used by the technical staff are shown
in Table 2.
The component costs in reference 3 were generated by Mr. LeRoy
Lingren of Rath and Strong, Inc. Mr. Lingren has been recognized as an
D-3
-------
Table 2
Cost of Components in a Three-Way Plus Oxidation Catalyst System
Component Cost in '78 Dollars
Min. Max.
Throttle Position Sensor - 2
PCV valve 1 1
HEI (less breaker point distributor) 7 7
TVS (spark) - 2
Electric choke 1 1
EFE 4 4
ECR (backpressure) 7 7
TVS (EGR) - 2
Stainless steel exhaust pipe (less
steel pipe) 9 9
Air injection system 30 30
Air switching system 2 2
Feedback carburetor (less open loop
carburetor) 8 8
Three-way plus oxidation catalyst 157 157
ECU 30 30
02 sensor 3 3
H20 temp sensor - 2
Inlet air temp sensor - 2
Engine speed sensor - 2
Crank angle position sensor - 2
EGR pintle position sensor - 2
Evaporative system 10 10
TOTAL 269 285
D-4
-------
expert in the field of cost of emission control systems for a number of
years. In addition to his efforts for EPA, he has done similar work for
the National Academy of Sciences (see reference 4).
Now, assuming the cost of the 1987 emission control system to be
$285, a judgment must be made concerning what fraction of that cost should
be allocated to HC control as opposed to CO and NOx control. Two estimates
of that fractional cost were developed by the technical staff. The high
estimate is $200. A more reasonable estimate is $285/3 (for control of
3 pollutants) or $95.
Assuming 10 million cars are sold per year, the annual cost due to initial
cost increases would be:
10 x 10 (vehicles soldqper year) x ($95 to 200 per vehicle) =
$0.95 to 2.00 x 10y
I.E. Maintenance Costs
The items in Table 3 were considered to be the changes in maintenance
between uncontrolled vehicles and 1987 model year vehicles.
The costs in Table 3 are considered by the technical staff to be a
case of minimal savings as the prices assumed for spark plus changes and
point/condensor changes are probably too low. The increase in oil change
intervals, which is at least in part attributable to the use of unleaded
fuel, is not considered. Also 0^ sensor change intervals may be more than
30,000 miles by 1987 as that interval is being used or exceeded by some
production vehicles already. The $50 miscellaneous cost for repair of
emission control systems is in addition to that amount for repairs
attributable to inspection and maintenance programs.
It is then assumed that a $5 per year savings in maintenance cost
will be accumulated by each vehicle on the road (110 x 10 vehicles)
g
for a total savings of $0.55 x 10 per year.
D-5
-------
Table 3
Maintenance Changes Over 100,000 Miles
Maintenance
Change 02 sensor 3 times
Miscellaneous emission system
repairs
Save 5 plus changes
Save 10 point/condenser changes
Save 1 muffler change
Total
Total Cost of
Maintenance
3 x 15 = $45
$50
5 x 10 - $(-)50
10 x 10 = $(-)100
$(-) 20
Cost Related to
HC Control
1/3 x 45 = $15
1/3 x 50 = $17
1/2 x (-50) = $-25
1/2 x (-100H-50
1/3 x (-20) = $-7
$(-)75 per 10 years
$(-)50 per 10 years
D-6
-------
I.C. Differences in Fuel Economy
Figure 9 of reference 5 shows the composite economy of 1978
California vehicles (which meet a 0.41 HC standard) to be about 7 or
8% better than that of uncontrolled vehicles at a constant weight mix.
It also can be seen from Figure 9 of reference 5 that current Federal
vehicles achieve 19 or 20% better fuel economy than their uncontrolled
counterparts. While it could be argued that fuel economy will continue
to improve until 1987, even with respect to Federal vehicles, the
technical staff hass chosen to further consider only the 7% benefit.
This value represents only a minimal fuel savings.
To determine the annual dollar savings due to improved vehicle fuel
economy in 1987, some value of the fuel economy must be assumed. Two such
values were chosen, and those values are 17.0 and 25.0 miles per gallon.
These values represent a 7% improvement when compared to 15.9 and 23.4
miles per gallon respectively.
Assuming that all vehicles on the road (110 x 10 vehicles) travel
10,000 miles per year (somewhat low) and that gasoline costs 70<£/gal
(again this may be low for 1987) the total savings ranges from 2.1 to
3.1 billion dollars. For HC control only (i.e., divide by 3) this would
be a savings of 0.72 to 1.03 billion dollars.
I.D. Summary of Gaseous Emissions
A range of cost attributable to LDV gaseous emission control was
obtained by summing the initial cost of emission control systems, maintenance
savings, and savings in fuel economy.
D-7
-------
The minimum cost of the program is calculated as follows:
[.95 - .55 - 1.03] x 109 = -$0.63 x 109
The maximum cost of the program is then:
[2.00 - .55 - 0.7] x 109 - $0.75 x 109
I.E. The Use of Unleaded Fuel
Unleaded gasoline is now 4 to 5<£/gal more expensive than regular
gasoline. This trend is partially due to a slightly higher refining cost
of unleaded fuel and partially because of pricing strategies of gasoline
retailers. Unleaded fuel is currently the "lower volume" seller, but
by 1987 unleaded fuel will be the "high volume" product as demand for
regular gasoline will be greatly reduced.
If the cost differential between regular and unleaded gasolines is
heavily influenced by production volume, regular may be more expensive
than unleaded by 1987. However, in the interest of keeping the calculations
in this text conservative, a 4<£/gal penalty is assumed.
Again for 110 million cars going 10,000 miles per year with fuel
economy ranging from 17 to 25 miles per gallon, the total penalty would
be 1.76 to 2.59 billion dollars. For HC control only, the penalty is 0.59
to 0.86 billion dollars.
I.F. Evaporative Emissions (Includes LDV and LOT)
Control of evaporative emissions has evolved from a 2 gram/test
canister standard to a 6 gram/test SHED standard in 1978 to a 2 gram/test
SHED standard for 1981. In 1974 the cost to implement the 6 gram/test
SHED regulations was estimated to be 82 million dollars per year. The
2 grams/test SHED regulations (for 1981) are to cost an additional
12 to 62 million dollars per year.
D-8
-------
A good estimate for the cost of going from uncontrolled to the 2 gram/test
g
canister standard was not available. So an estimate of $10 per vehicle
was added to each new vehicle. The number of new vehicles was assumed to
be 14 million per year.
Thus, the total cost of LDV and LOT evaporative emission standards
was calculated as follows:
$.082 x 109 x (l.l)4 (to go to '78 dollars) + [.012 to .062 x 109]
+ $10 x 14 x 105 (vehicles per year) - $0.272 to 0.322 x 109
I.G. Emission Control at Altitude
Emission control at altitude is not required for model years 1979
through 1980 as a result of the 1977 Clean Air Act Amendments. Altitude
regulations may be in force again by the 1981 model year. The final form of
these regulations has not yet been determined, and a number of options
for emission control at altitude during the 1981 and subsequent model
years are under consideration. In addition the Clean Air Act requires
stringent altitude controls beginning with the 1984 model year. Thus,
the program cost can only be estimated within a fairly wide range. That
Q
range is from $0.050 to 0.945 x 10 as seen in references 9 and 10 respectively.
To consider only HC control the dollar amounts were cut in half to $0.025 to
0.473 x 109.
II. Light-Duty Trucks
11.A. Initial Cost of Emission Control Systems
Initial costs of emission control systems for light-duty trucks were
assumed to be the same as for light-duty vehicles. That is $95 to 200
for HC control compared to uncontrolled trucks. Four million LDT's were
assumed to be sold in 1987. The total cost then is:
($95 to 200) x 4 x 106 (sales) - $0.38 to 0.800 x 109
D-9
-------
II.B. Maintenance Costs
To be conservative, no maintenance savings were credited to light-duty
trucks.
II.C. Difference in Fuel Economy
Fuel economy savings were also not credited to light-duty
trucks, even though savings are expected.
II.D. Use of Unleaded Fuel
The dollar penalty for light-duty vehicles was reduced by the factor
of 0.4 (4 x 10 / 10 x 10 = 0.4) to account for differences in population
and increased by the consideration of the difference between LOT and LDV
fuel consumptions. (The DOE gasoline volume model predicts* the ratio of
LDV to LOT fuel economy in 1985 to be 1.25).
The cost is calculated as follows:
($0.59 to 0.86 x 109) x 0.4 x 1.25 = $0.30 to 0.43 x 109
11. E. Emission Control At Altitude
This was estimated to cost 40% of the cost of LDV emission control at.
altitude. The 40% is to correct for differences in annual sales between
LOT and LDV.
III. Heavy-Duty Vehicles
III.A. Gaseous Emissions
The total cost of the proposed 1983 and subsequent model year
heavy-duty vehicles emission standards (for 90% reductions of HC and CO)
was found to be 2.5065 billion dollars in reference 11 for 1983 to 1987.
*Personal communication between K. Hellman of EPA and B. McNutt of DOE
on October 26, 1978.
D-10
-------
This cost includes first cost and maintenance for hardware to comply
with the emission standards for 1983 and subsequent model years, certifi-
cation testing, modifications to facilities to permit testing, inspection
and maintenance programs for such vehicles, and the switch to unleaded
fuel for gasoline-fueled vehicles. Fuel economy of heavy-duty vehicles
was assumed to remain constant.
An approximate annual cost for gaseous emission control was calculated
as follows:
$2.5065 x IP9 = $0.501 x IP9
5 years year
This number includes an unleaded fuel cost differential which is 5
-------
III.B Evaporative Emissions
Reference 12 states that the cost of heavy-duty evaporative emission
control for the 1982 to 1986 model years will be $56 x 10 . This figure is
currently being reevaluated. Our best estimate is that this amount will not
exceed $80 x 10 . So based on the latter estimate, the cost per year would be:
$80 x 106 = $16 x IP6 = $.016 x 109
5 years year year
IV. Motorcycles
Reference 13 indicates that the cost to the consumer of the 1978 and
1980 motorcycle emission standards will be 104.3 million dollars* over the
1978 to 1982 timeframe (including fuel consumption). Assuming these to be
1976 dollars, the approximate annual cost for HC control would be:
$104.3 x IP6 x (I.I)2 (to go to 78 dollars) x 1/2 (for HC only)
5 years
= $1.26 x 1Q7 = $.013 x 109
year year
Motorcycle emission standards for the 1983 to 1984 model years and the
1985 and subsequent model years have not yet been proposed. However, the cost
of these regulations may be estimated from figures 25 and 28 of reference 13.
The incremental cost of the 1985 standards based on those figures is
roughly estimated to be $100 (in 1978 dollars) per motorcycle over uncontrolled.
This represents a $150 first cost minus a $50 fuel savings. If there are one
million motorcycles sold per year, then the cost of this program would be about:
$100 x 106 (motorcycles) = $.100 x 109 per year
For HC control only, the cost is assumed to be one third of the 0.1
billion dollars or .033 billion dollars as HC and CO could be reduced by
90% and NOx could be reduced by 75% for 1985.
D-12
-------
The total cost of the motorcycle program for HC control would then be:
($[.013 + .033] x 109 = $0.046 x 109 per year
V. Aircraft
A complete discussion of the cost of proposed aircraft regulations is
presented in reference 14. The cost of the program was determined as
shown in Table 4.
Table 4
Cost of the Aircraft Program
Costs Summed in '76 Dollars Location in Reference 14
66,400 x 10? Table 8, p. 20
5,894 x 10, Table 8, p. 20
114,443 x 10, Table 9, p. 21
228,600 x 10, Table 10, p. 22
267,078 x KT Table 10, p. 22
36.7 x 10? Table 11, p. 24
-0.7 x 10; Table 12, p. 25
-88.7 x 10° . Table 13, p. 26
629.7 x 106 for 15 - 18 years
The total of $629.7 x 106 is in 1976 dollars and is for a total of 18 years
of HC and CO control and 15 years of NOx control. Fuel consumption effects
are included.
Approximate annual cost of the program for HC was calculated as follows:
;o '
T5~
[$629.7 x 106 x (l.l)2 (to go to '78 dollars)] x 1/2 (for HC only)
= $2.539 x 107 = $.025 x IP9
year year
g
The $.025 x 10 per year is on the high side as three pollutants are
controlled for 15 years and two pollutants for an additional three. In
any case the aircraft program is small contributor in terms of total program
cost.
D-13
-------
References
! Federal Register. June 22, 1978 (43 FR^ 26962-26985).
2. "Analysis of Technical Issues Relating to: California's Request for
Waiver of Federal Preemption with Respect to Exhaust Emission
Standards and Test Procedures ofr 1981 and Subsequent Model Years
Light-Duty Vehicles", Environmental Protection Agency, Emission
Control Technology Division, March, 1978, p. 21.
3. Cost Estimations for Emission Control Related Components/Systems and
Cost Methodology Description, EPA-460/3-78-002, prepared by LeRoy
H. Lingren of Rath and Strong, Inc., March, 1978.
4. Manufacturability and Costs of Proposed Low-Emissions Automotive
Engine Systems, Consultant Report to the: Committee on Motor Vehicle
Emissions, Commission on Sociotechnical Systems, National Research
Council, Sept., 1974.
5. "Light Duty Automotive Fuel Economy Trends Through 1976," SAE Paper
780036, J.D. Murrell, Environmental Protection Agency, Feb.-Mar., 1978.
6. "Revised Evaporative Emission Regulations for the 1978 Model Year,"
Environmental and Economic Impact Statement, Environmental Protection
Agency.
7. "Revised Evaporative Emission Regulations for 1981 and Later Model
Year Gasoline-Fueled Light-Duty Vehicles and Trucks," Environmental
and Economic Impact Statement, Environmental Protection Agency,
Aug. 7, 1978.
8. Investigation and Assessment of Light-Duty Vehicle Evaporative
Emission Sources and Control, Report No. EPA 460/3-76-014.
9. "Draft Environmental and Economic Impact Statement for 1981-1983
High Altitude Emission Standards", Environmental Protection Agency,
Sept. 29, 1978, Table III, 1977 approach with I/M.
10. "Draft Report to Congress in Response to Section 206(f)(2) of the
Clean Air Act as amended in August, 1977", Environmental Protection
Agency as of Oct. 27, 1978.
11. "Revised Gaseous Emission Regulations for 1983 and Later Model Year
Heavy-Duty Engines," Draft Environmental and Economic Impact Statement,
Environmental Protection Agency, as of Oct. 26, 1978.
D-14
-------
12. "Draft Environmenal Impact Statement for Gasoline-Fueled, Heavy-Duty
Vehicles - Notice of Proposed Rulemaking", Environmental Protection
Agency, as of Oct. 26, 1978.
13. "Exhaust and Crankcase Regulations for the 1978 and Later Model Year
Motorcycles", Environmental and Economic Impact Statement, Environmental
Protection Agency, Dec. 14, 1976.
14. "Cost-Effectiveness Analysis of the Proposed Revisions in the Exhaust
Emission Standards for New and In-Use Gas Turbine Aircraft Engines
Based on EPA's Independent Estimates", Technical Support Report
for Regulatory Action, Richard S. Wilcox and Richard W. Munt,
Environmental Protection Agency, Dec. 14, 1976.
D-15
-------
Appendix E
COST/EFFECTIVENESS OF INSPECTION AND MAINTENANCE PROGRAMS
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
U.S. Environmental Protection Agency
Ann Arbor, Michigan
-------
Appendix E
COST/EFFECTIVENESS OF INSPECTION AND MAINTENANCE PROGRAMS
I. Direct Costs and Benefits
The following costs and benefits are associated with an I/M program.
1. The cost of program implementation is assumed to be negligible
by 1987, on the premise that an annual I/M program begins in 1982.
2. The cost of a program's day-to-day operation is assumed to be
covered by a $5/car inspection fee. This is the fee for an I/M
sticker in Oregon's I/M program. The cost of inspection is expected
to vary somewhat from program to program, but inspection fees in
currently operating I/M programs are in the neighborhood of $5.
3. The cost of repairing failed vehicles was estimated from the
two latest data sets available for 1975-1977 model year cars. The
first data set is the Portland Study, which is a large scale test
program designed to evaluate the effectiveness of an operational I/M
program (Oregon's I/M program). The second data set is the California
Air Resources Board's Light Duty Vehicle Surveillance Test Program,
Second Series CVS [LDVSP-II], in which cars failing the idle portion
of a loaded mode test were maintained by CARB in a manner expected
to be seen in the field. A $50 limit on repair costs was also imposed.
The data are summarized below.
California-!2 cars (all catalytically equipped) - Avg. cost = $16.45
Portland-99 cars (most catalytically equipped) - Avg. cost = $23.35
E-l
-------
A conservative best estimate of repair costs due to I/M is $15, since
voluntary owner repairs will eventually be made to many of the items
which are performed to pass an I/M test.
4. The cost to the consumer in time/gas is ignored in this analysis.
5. The benefit to air quality from an I/M program which employs an
idle test is estimated on the basis of EPA surveillance data and projections
of these data to future years, and a simulation of the I/M process. The
computer program MOBILE1 (available from EPA, Ann Arbor, Michigan) was
used to facilitate the calculations. Assuming that an I/M program
begins in 1982 with an initial failure rate of 30%, the reduction in
hydrocarbons (gm/mi) is approximately 30%.
6. The improved quality of repair which the entire I/M fleet
should experience is ignored in this analysis.
7. The potential for longer vehicle life due to better and more
regular quality of repair is ignored.
8. The average improvement in urban fuel economy for repaired vehicles
was estimated on the basis of the Portland and CARB data sets discussed
under item (3). The data are summarized below.
CARB - 12 cars - 4 percent improvement
Portland - 99 cars - 1 percent improvement
A best estimate is considered to be 1.5%.
II. Cost of I/M per Vehicle
On a per vehicle basis, the costs of I/M are calculated on the
basis of estimates presented in the previous section and also on the
basis of the following assumptions:
E-2
-------
1. A gasoline price of $.70 per gallon in current dollars;
2. A fleetwide average of 10,000 miles per year per vehicle;
3. An average fleetwide fuel economy without I/M of 25
miles per gallon;
4. A failure rate of 30 percent for the program in 1987.
Thirty percent of the vehicles included in the program will fail the
inspection and thus will have to be repaired. The cost per repaired
vehicle will include the inspection fee ($5) and the repair cost ($15)
minus the savings resulting from improvement in fuel economy ($4.20).
The net cost will be approximately $15.80 per repaired vehicle. The
cost for the 70 percent of the vehicles that pass the test will amount
to only the $5 inspection fee.
Only half of the cost of an I/M program is attributable to HC control
since CO is also being controlled. Existing programs are equally stringent
for CO as for HC, hence as many failures result from CO as they do from HC.
III. Cost-effectiveness of I/M
The average cost per vehicle will represent a weighted average of the
cost per repaired vehicle and the cost per passed vehicle, as follows:
($15.80 per repaired vehicle) x (.3) +($5 per passed vehicle)
x (.7) - $8.24 per vehicle
Since only half of this cost is attributable to HC, the cost per vehicle
will be $4.12 for HC control.
The tons of exhaust eliminated per vehicle by I/M based on 30 percent
control will be: (1.98 gm/mi)x(.3)x(10,000 mi/yr) f (908,000 gm/ton) =
6.5 x 10"3 tons/yr
The cost-effectiveness is calculated by dividing the average cost per
vehicle by the tons of HC eliminated: $4.12 - = $635 per ton
6.5x10 tons
-------
TECHNICAL REPORT DATA
(Please read Instructions on tlie reverse before completing)
1 REPORT NO
EPA-450/5-79-002
4. TITLE AND SUBTITLE
5. REPORT DATE
Cost and Economic Impact Assessment for Alternative
Levels of the National Ambient Air Quality Standard
for Ozone
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Kenneth H. Lloyd
3 RECIPIENT'S ACCESSI ON-NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Economic Analysis Branch (MD-12)
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air, Noise and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report presents the results of an analysis of the potential impact of
feasible changes in the national ambient air quality standard for ozone on national
costs of control and the attainment status for various areas of the country. An assess
ment is made for 90 AQCRs of the control requirements for alternative levels of the
standard. For each AQCR this analysis estimates the potential emissions in 1987 and
the allowable emissions for attainment of the standard levels. Next, the analysis
estimates the potential emission reduction achievable with the Federal Mobile Vehicle
Control Program (FMVCP), new and modified source control, application of reasonably
available control technology (RACT) on existing stationary sources and further motor
vehicle controls through inspection/maintenance programs and transportation control
plans. Based on the projected emission reductions» control costs are estimated for
applying technology in an attempt to attain alternative standard levels. While the
analysis considers each AQCR separately, the results are presented in aggregate form
for all 90 AQCRs instead of each individual AQCR.
The analysis does not include a rigorous economic impact assessment on the numeroujs
affected industries or the impact of growth restrictions on the economies of affected
urbanized areas. A summary of the impact on several industries for which EPA has con--
ducted economic impact studies is presented.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Ozone
Photochemical Oxidants
Ambient Air Quality Standards
Hydrocarbon Control
Volatile Organic Compounds
b.lDENTIFIERS/OPEN ENDED TERMS
COSATi Held/Group
13. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (9-73)
------- |