United States Office of Mobile Sources EPA 460/3-83-001
Environmental Protection Emission Control Technology Division
Agency 2565 Plymouth Road
Ann Arbor, Ml 48105
Air
¦sfrERA Controlling Emissions from
Light —Duty Motor Vehicles at
Higher Elevations
-------
EPA 460/3-83-001
Controlling Emissions from Light-Duty
Motor Vehicles at Higher Elevations
A Report to Congress
Prepared by
U. S. Environmental Protection Agency
Office of Air, Noise and Radiation
Office of Mobile Sources
Emission Control Technology Division
Standards Development and Support Branch
in cooperation with
U.S. Environmental Protection Agency
Region VIII
Air Programs Branch
Approved
February 1983
-------
Table of Contents
Page
Executive summary ES-1
I. Introduction ES-1
II. Control Scenarios Based on Current
Standards ES-2
III. Control Scenarios Based on
Revised Standards ES-13
IV. Conclusions and Recommendations . ES-16
Chapter I.
Chapter II.
Chapter III
Chapter IV.
INTRODUCTION 1-1
I. Air Quality in High-Altitude
Areas 1-1
II. Past and Present Standards
Affecting High-Altitude Vehicles 1-1
III. Requirement to Assess and
Implement Standards for 1984
and Later Model Years 1-3
IV. Organization and Scope of the
Report 1-6
IDENTIFICATION OF THE HIGH-ALTITUDE
CONTROL SCENARIOS II-l
I. Introduction II-l
II. The Basis for the Control Scenarios II-l
TECHNOLOGY ASSESSMENT FOR GASOLINE-
FUELED LIGHT-DUTY VEHICLES III-l
I. Introduction III-l
II. The Effects of Increasing
Altitude on Exhaust Emissions . III-2
III. Effects of Increasing Altitude
on Evaporative Emissions .... III-1G
IV. Technologies Necessary for
Complying With the Scenarios . . 111-10
V. Effects of High-Altitude Stan-
dards on Low-Altitude Control
Technology 111-38
VI. Effect of High-Altitude Control
Technology on Fuel Economy . . . 111-39
ENVIRONMENTAL IMPACT FOR GASOLINE-FUELED
LIGHT-DUTY VEHICLES IV-1
I. Introduction IV-1
II. Total Emissions IV-2
III. Air Quality IV-19
IV. Sensitivity Analysis IV-28
V. Summary IV-33
i
-------
Table of Contents (cont'd)
Page
Chapter V. ECONOMIC IMPACT FOR GASOLINE-FUELED
LIGHT-DUTY VEHICLES V-l
I. Introduction V-l
II. Costs to Manufacturers v-l
III. Costs to Users V-23
IV. Aggregate Costs V-28
V. Socioeconomic Impact V-31
VI. Sensitivity Analysis V-40
Chapter VI. COST EFFECTIVENESS FOR GASOLINE-FUELED
LIGHT-DUTY VEHICLES VI-1
I. Introduction VI-1
II. Methodology VI-1
III. Analysis VI-1
Chapter VII. COMPARING THE ALTERNATIVE CONTROL
SCENARIOS FOR LIGHT-DUTY GASOLINE-
FUELED VEHICLES VII-1
I. Introduction VII-1
II. Methodology VII-2
III. Review of the Economic Impact,
Environmental impact, and Cost
Effectiveness VII-3
IV. Ranking of the Alternative High-
Altitude Scenario VII-9
V. Evaluation of the Alternative
Scenarios Based on Their
Sensitivity to the Assumptions
of the Analysis VII-10
VI. Comparison of Scenario 2
to the Base Scenario VII-15
chapter VIII. CONTROL OF LIGHT-DUTY TRUCKS VIII-1
I. Introouction VIII-1
II. Technology Assessment VIII-1
III. Economic impact VIII-1
IV. Environmental Impact VIII-7
V. Cost Effectiveness VIII-16
Chapter IX. CONTROL OF LIGHT-DUTY DIESEL
GASEOUS EMISSIONS IX-1
I. Introduction IX-1
II. Technology Assessment IX-1
III. Economic Impact IX-5
IV. Environmental impact IX-6
V. Cost Effectiveness IX-11
ii
-------
Table of contents (cont'd)
Page
Chapter X. LIGHT-DUTY DIESEL PARTICULATE
EMISSIONS X-l
I. Introduction X-l
II. Controlling Particulate
Emissions from High-Altitude
Light-Duty Diesels X-l
III. Summary X-6
Chapter XI. HIGH-ALTITUDE STANDARDS AS A CONSEQUENCE
OF REVISED LOW-ALTITUDE STANDARDS . . XI-1
I. Introduction XI-1
II. Definition of Control Scenarios . XI-1
III. Technology Assessment XI-7
IV. Economic Impact XI-16
V. Environmental Impact XI-23
VI. Cost Effectiveness . XI-40
VII. Summary XI-40
Chapter XII. CONCLUSIONS AND RECOMMENDATIONS .... XII-1
I. Effects of the Alternative
Scenarios Based on Current
Standards XII-1
Appendix I. PERSPECTIVE ON THE INTERIM HIGH-ALTITUDE
STANDARDS API-1
I. Costs to Manufacturers API-1
II. Cost to users API-1
III. Impact on High-Altitude Dealer . . API-3
IV. Aggregate Cost to the Nation . . . API-3
V. Air Quality improvements API-3
VI. Cost Effectiveness API-3
Appendix II. INFORMATION FOR THE SUPPLEMENTAL
ENVIRONMENTAL ANALYSIS APII-1
Appendix III. SUPPLEMENTAL INFORMATION FOR THE
ECONOMIC ANALYSIS APIII-1
I. Development Costs APIII-1
II. Certification APIII-1
iii
-------
Executive Summary
I. INTRODUCTION
This report evaluates various strategies for controlling
high-altitude emissions from light-duty motor vehicles. It was
prepared in response to section 206(f)(1) of the Clean Air Act,
as amended (the Act). That section requires all light-duty
vehicles (passenger cars) made during or after the 1984 model
year to meet the requirements of section 202 of the Act
regardless of the altitude at which they are sold. Section 202
establishes the current gaseous exhaust emission standards for
light-duty vehicles (LDVs): 0.41 gram per mile (g/mi) for
hydrocarbons (HC), 3.4 g/mi for carbon monoxide (CO), and 1.0
g/mi for oxides of nitrogen (NOx). This section was also used
to promulgate evaporative emission and particulate standards
for these vehicles. The evaporative standard for
gasoline-fueled LDVs is 2.0 g/test HC and the particulate
standards for diesel-powerea LDVs are 0.6 g/mi beginning in the
1982 model year and 0.2 g/mi beginning in the 1985 model year.
Section 206(f)(2) requires the U.S. Environmental
Protection Agency (here after referred to as EPA or the Agency)
to report to Congress regarding the economic impact and
technical feasibility of the above "all-altitude" requirement,
in addition to the technical feasibility and health
consequences of proportional high-altitude emission standards
that reflect a percentage reduction in emissions comparable to
that achieved in low-altitude areas. For 1982 and 1983 model
year LDVs, EPA established proportional high-altitude standards
of 0.57 g/mi HC, 7.8 g/mi CO, 1.0 g/mi NOx, and 2.6 g/test
evaporative HC at 5,300 feet above sea level. No proportional
diesel particulate standard was promulgated, but this report
examines the possibility of such a standard.
One problem in developing this report was that the exact
emission control requirements of the all-altitude passenger car
provision are not clear. The statute and accompanying
legislative history can plausibly be interpreted in various
ways. The two basic areas where interpretation of the statute
is necessary are:
1) the altituoe or altitudes where compliance with the
standards of section 202 is specifically required; and
2) whether exemptions from the all-altitude requirement
are permissible.
The Agency will formally establish the exact requirements
of the statute in the future. This report responds to the
Congressional mandate by analyzing various control scenarios
which encompass possible interpretations of the all-altitude
provision. The report also explores some alternatives not
-------
ES-2
currently allowed in the statute, but does not evaluate all of
the many possible options. High-altitude emission control
alternatives are directly influenced by several factors, and
probably most importantly by the low-altitude standaros. There
is also significant uncertainty in the prediction of in-use
emission factors from new technologies only recently
introduced, and this uncertainty bears directly on the Agency's
ability to make firm conclusions regarding the relative merits
of some alternatives. Certain conclusions can be confidently
made and have been. Other conclusions are highly uncertain and
have been qualified as such. Hopefully, however, the options
analyzed provioe an indication of the complexity of the various
interactions and a frame work for further analysis.
This document also investigates other areas of interest.
The most important of these other areas include:
1) the consequences of controlling light-duty trucks
(LDTs) in addition to LDVs; and
2) the consequences to high-altitude emission control
strategies of possible revisions in the current statutory
(low-altitude) standards.
II. CONTROL SCENARIOS BASED ON CURRENT STANDARDS
A. Analytical Methodology
For this report, EPA evaluated seven emission control
scenarios representing a wide range of options (see Table ES-1
for a summary). Six of these control strategies were analyzed
in detail for their economic and environmental impacts relative
to the base scenario (i.e., the current fixed-point
proportional standards). Therefore, the six control scenarios
are evaluated as alternatives to continuing the base scenario.
The economic analysis included the costs of the capital
investment, the hardware, and changes in fuel economy and
maintenance. The environmental impact was measured through
projected changes in lifetime emissions from motor vehicles
sold during the first 5 years of the regulation, and through
estimates . of future ambient air quality in several
high-altitude cities.
The analysis of alternative emission control strategies
was conducted in two parts. The primary analysis was conducted
for light-duty gasoline-fueled vehicles (LDGVs) and was used to
reject unacceptable emission control strategies from the
various alternatives. The secondary analysis further
considered the most desirable control scenario for its effects
on light-duty gasoline-fueled trucks (LDGTs) and light-duty
diesel-powered motor vehicles (LDDs) before a final assessment
of the desirability of an^ alternative scenario was made.
-------
Table ES-1
Summary of the Seven Control Scenarios Evaluated in the Report
Compliance Strategy
Standards!a] Continuous[b] Fixed-Point[c] Exemptions[d]
Scenario Statutory Proportional lUf200 Ft 6,000 Ft 5y3UU Ft Yes No
la
X
X
X
lb
X
X
X
lc
X
X
X
2
X
X
X
3a
X
X
X
3b
X
X
X
Basel ej
X
X
X
[a] See Table ES-2 for numerical values.
[bj Continuous strategies require that a vehicle automatically comply with the
appropriate standards at all elevations up to a certain maximum altitude.
[c] Fixed-point strategies allow a modified or recalibrated vehicle to be sold above
4,000 feet. Compliance with the appropriate standards must be demonstrated at
the single elevation of 5,300 feet.
[d] Exemptions or waivers from the high-altitude requirements would be available for
some vehicles to reduce the economic burden of the standards or to prevent
negatively affecting model availability at all altitudes.
[e] Scenarios 1 through 3 are evaluated as additional requirements to continuing the
1982-83 proportional high-altitude standards.
-------
ES-4
B. Description of the Control Scenarios
1. Base Scenario
The base scenario is primarily a continuation of the
high-altitude requirements currently in place for 1982 and 1983
model year vehicles sold above 4,000 feet. Collectively, these
requirements are termed a "fixed-point proportional strategy"
because compliance with the proportional high-altitude
standards must be demonstrated only at 5,300 feet (Table
ES-2).* Such a standard allows vehicles to be modified to
meet the high-altitude requirements after production if they
will be sold above 4,000 feet.
Exemptions to these current high-altitude standards are
available to reduce the economic burden of compliance. At
present they are granted for certain low-power LDVs which are
expected to perform unacceptably at high altitude and which may
have technical difficulty in meeting the standards cost
effectively. such exempted vehicles cannot be sold at high
altitudes.
Future regulations could, of course, include other types
of exemptions, or perhaps waivers that allow some vehicles
certified only at low altitude to be sold at high altitude.
Performance-based exemptions appear to work satisfactorily,
however, and they are the only type of exemptions considered in
this study.
As previously stated, the six remaining control strategies
in this report were examined as increments to the base
scenario. The Agency believes that these regulations were
justified during the final rulemaking process and that little
benefit would result from again presenting the detailed
supporting analyses. Therefore, the primary purpose of this
report is to determine which alternatives to the current
standards warrant further consideration.
2. Scenario 1
This alternative scenario is termed a "continuous
statutory strategy" because it would require compliance with
the low-altitude (statutory) standards at all altitudes up to a
maximum elevation (Table ES-2). Every vehicle sold in the
nation would have to automatically meet the standards) (i.e., no
modifications are allowed). The variations of this scenario
are:
Technically, any place over 4,000 feet is currently
considered high altitude. Compliance is demonstrated at
5,300 feet because that is the altitude of Denver,
Colorado which has appropriate test facilities.
-------
ES-5
Table ES-2
Standard
Type
Proportional
Proportional and Statutory standards for
High-Altitude Vehicles at 5,300 Feet
(based on current low-altitude standards)
Gaseous standards
HC CO
Vehicle[a] (g/mi) (g/mi)
LDV
LDT
0.57
1.0
7.8
14
Evap.
NOx HC[b]
(g/mi) (g/test)
1.0
2.3
2.6
2.6
Particulate
Standards[c]
(g/mi)
[d]
[d]
Statutory
LDV
LDT
0.41
0.8
3.4
10
1.0
2.3
2.0
2.0
0.2
0.26
TaT
[b]
[c]
Id]
Light-duty vehicle.
Light-duty truck.
Although not required by the
standards were included in this report
Act,
for
statutory LDT
completeness.
Evaporative standards
vehicles.
Particulate standards
vehicles.
Exact level to be determined in the future. In this report,
we estimated that a proportional particulate standard would
be approximately 50 percent more than the corresponding
low-altitude standard.
apply only to gasoline-fueled motor
apply only to aiesel-powered motor
-------
ES-6
1) la - Compliance is up to 10,200 feet without
exemptions;
2) lb - Compliance up to 6,000 feet with exemptions; and
3) lc - Compliance up to 6,000 feet without exemptions.
Exemptions can significantly reduce the economic burden of
the standards. Under control strategies such as scenario 1
where vehicles must meet emission standards without
modification, however, they also appear important in preventing
an adverse impact on model availability at all elevations. For
many low-power, high fuel economy LDVs, compliance with the
statutory standards at high altitude may degrade performance to
such a degree that these vehicles may actually become unsafe to
use at higher elevations. Rather than market such potentially
unsafe vehicles, manufacturers would likely decide to remove
them from the national market. This would affect model
availability and fuel economy throughout the nation. Exempting
these low-power vehicles for sale only at low altitude would
generally not affect model availability at high altitude, since
these vehicles would normally not be sold in these areas
because of their poor performance, even in the absence of
high-altitude standards.
3. Scenario 2
This is termed a "fixed-point statutory strategy" and is
similar to the base scenario except that it would require
vehicles to meet the statutory (low-altitude) standards at
5,300 feet rather than proportional standards (Table ES-2).
Modifications to vehicles sold above 4,000 feet would be
allowed and exemptions would be available.
4. scenario 3
This is termed a "continuous proportional strategy." At
1,800 feet, the standards are the low-altitude (statutory)
standards. At 5,300 feet, the standards are the proportional
standards shown in Table ES-2. In between these elevations and
up to 6,000 feet the standards would vary linearly with
altitude. Scenario 3, like scenario 1, would require all
vehicles in the nation to meet the standards without
modification. The variations of this scenario are:
1) 3a - Exemptions; and
2) 3b - No exemptions.
C. Comparison of the Alternative Control Scenarios
Each alternative scenario was analyzed in detail for LDGVs
to determine its emission control technology, economic, and
-------
ES-7
environmental impacts beyond the base scenario (i.e., the
current fixed-point proportional standards). (See summary in
Table ES-3). Based on this analysis, one alternative was
further evaluated for LDGTs and LDDs.
1. Control Technology
The costs of each alternative scenario vary with the
technical requirements they impose for controlling exhaust
emissions. The emission control technology requirements for
reducing gaseous pollutants are primarily based on three
factors:
1) the maximum elevation for which control must be
demonstrated;
2) the extent of exemptions, if any; and
3) the level of the standards.
Emission standards based on scenario la represent the most
stringent interpretation of the Act's requirements. Requiring
compliance up to 10,200 feet with no exemptions, these
statutory standards would require the use of sophisticated
electronic equipment and turbochargers, and are the most
technically difficult alternative. Standards based on scenario
2 would require the least complex emission control hardware.
These statutory standards would require the use of either
specifically calibrated control modules, or an expansion of the
capabilities of existing units on high-altitude vehicles
equipped with electronic control systems. Less sophisticated
aneroids (pressure sensing devices) would be required on
high-altitude vehicles equipped with non-electronic control
systems. Standards based on the remaining alternatives would
fall between the hardware estimates for scenarios la and 2.
2. Economic Impact
Scenario 2 is by far the least costly of the alternative
scenarios (Table ES-3). Even without the estimated
fuel-economy benefit, the incremental cost of this control
strategy is only $17 million during the first 5 years of the
regulations. In comparison, the incremental costs for the
other alternatives range from $187 million to $4.97 billion.
The primary reason for this is that scenario 2 is a two-car
strategy, so only those vehicles sold at high altitude need
additional emission controls. The other scenarios are one-car
strategies and require high-altitude emission control hardware
on all vehicles nationwide, regardless of where they are sold.
-------
Table ES-3
Costs and Beneifts of Alternative LDGV Control Strategies[a]
Incremental
Aggregate Costs Total Reductions Cost Effectiveness
(5-year total)Lb] (5-year total) (dollars/metric ton)[cJ
(1Q6 dollars) (103 metric tons) HC CO
Scenario
LOW
High
HC
CO
Low
High
LOW
High
la
4,151
4,974
24.6
-425
101,000
121,000
[ tii
Id]
lb
524
1,010
24.6
-425
12,600
24,500
Id]
Id]
lc
1,384
1,907
24.6
-425
33,600
46,200
[d]
Id]
2
Uncertain[e]
17.9
331
neg.
575
neg.
30
3a
187
224
1.1
18 .0
95,000
115,000
b,20U
7 , 5UU
3b
449
636
1.1
18.0
230,000
325,000
14,900
21,100
[a] Each control scenario is examined as an increment to continuing the
1982-83 proportional high-altitude standards. For comparison/ over the
first two years of the regulation, these standards will cost the nation
about $24 million (1981 dollars discounted to 1982), and reduce HC and
CO emissions by about 33 and 1,200 tons, respectively. The cost
effectiveness of these standards is $365 per metric ton for HC and $10
per metric ton for CO. (These costs and benefits are relative to the
total absence of high-altitude regulations.)
[b] 1981 dollars discounted to the effective date of the regulations (1984).
[c] Cost effectiveness was determined on a per vehicle basis. For scenario
2 the high-cost estimates exclude the estimated fuel economy benefit.
[d] The cost-effectiveness values for CO under scenario 1 were not
presented in the table since emissions of this pollutant may increase
under this strategy. However, it is also possible that CO emissions
may decrease by about the same total amount listed for both scenarios 2
and 3. Using this assumption, the cost effectiveness would range $535
to $5,100 per metric ton for CO.
[e] A net savings can result if the incremental purchase price increase
(about $15 per high-altitude vehicle) is offset by a potential fuei
economy benefit (about $25 per high-altitude venicle). If the
potential fuel economy improvement is excluded for scenario 2, the cost
would range up to $17 million. The estimated fuel savings is tentative
at this time because of the limited data base.
-------
ES-9
3. Environmental impact
Significant differences are possible in the environmental
impacts associated with each of the alternative scenarios.
Scenario 1 (all options) predicts the greatest incremental
reduction in HC emissions, although it also appears to increase
CO emissions in high-altitude areas. These projections are
based on tentative emissions factors, however. In the future
as more complete data are collected, it is possible that
scenario 1 may substantially reduce CO emissions in areas above
1,800 feet instead of producing an adverse environmental
impact. Therefore, a final judgment on the impact of scenario
1 must await additional information.
The incremental emission reductions predicted for
scenarios 3a and 3b are relatively small. Scenario 2 appears
to offer the largest incremental emission reductions of all the
alternatives. Again, the uncertainty in predicting emission
factors discussed above applies here also.
Air quality modeling projections show that scenario 2 has
a positive impact on the ambient air quality in high-altitude
regions, although the affect is relatively small. This is not
surprising, however, because scenario 2 is an incremental
control strategy and future improvements in air quality will
only come by combining several incremental controls, each
having a small benefit of its own. More detailed analysis is
needed before a firm conclusion can be made regarding whether
these more stringent controls are needed, or if those provided
by the base scenario are sufficient to attain the NAAQS in
particular areas.
4. Cost Effectiveness
Table ES-3 shows that scenario 2 is predicted to be the
most cost effective of the alternatives analyzed. The
incremental cost effectiveness of reducing HC emissions ranges
from less than $0 to $575 per metric ton, compared to $12,600
to $325,000 per metric ton for scenarios 1 and 3. The
incremental cost effectiveness of reducing CO emissions under
scenario 2 ranges from less than $0 to $30 per metric ton,
compared to $535 to $21,100 per metric ton for the other
alternative scenarios.
Scenario 2 is the only alternative scenario with
cost-effectiveness values comparable to those for other mobile
source emission control regulations. Even with high-cost
estimates, scenario 2 appears to be a cost-effective approach
to reducing high-altitude emissions from LDGVs, using the
assumptions previously discussed.
-------
ES-10
5. Rationale for the Consideration of Scenario 2
The complexity of analyzing alternative high-altitude
standards for 1984 and later model year motor vehicles requires
the use of simplifying assumptions. However, these assumptions
may affect the consideration of scenario 2 as an alternative to
continuing the current standards. it is, therefore, important
to examine the sensitivity of scenario 2 to the underlying
analytical assumptions.
Five potentially important assumptions underlay the
economic and environmental impact chapters of this report:
1)
the
estimated fuel economy
saving;
2)
the
number of exemptions;
3)
the
levels of the emission
standards;
4)
the
use of low-altitude
vehicles in high-altitude
areas; and
5) the fleet mix of electronic (feedback) and
non-electronic (nonfeeaback) systems.
The first three assumptions are most important and are briefly
discussed below.
a. Estimated fuel economy savings. Scenario 2 is
sensitive to changes In fuel consumption caused by the use of
high-altitude emission control hardware. In fact, the
incremental net cost is so sensitive to this aspect of the
analysis that the lower limit of the possible fuel economy
benefit (i.e., no improvement) was included in the previous
discussion of cost and cost-effectiveness values. If the upper
limit of 4 percent improvement were used in the analysis, the
potential net savings would be even greater. Because of the
sensitivity of the analysis to projected changes in fuel
economy, this subject area should be carefully reevaluated as
additional information becomes available so that the total cost
associated with this scenario can be more accurately determined.
b. Number of exemptions. The desirability of scenario
2 may depend on the number of vehicles needing exemptions.
Desirable high fuel-economy vehicles, which theoretically coula
be sold in the absence of scenario 2 (or any scenario providing
exemptions), might easily become unavailable at higher
elevations. Such vehicles could represent as much as about 10
percent of the market, possibly reducing model availability in
these areas. On the other hand, the absence of the exemption
provision may well prevent the availability of such vehicles
nationwide.
-------
ES-11
Further technical achievements, however, may reduce the
need for exempted vehicles. Also, as better information
becomes available, the exemption criteria may be refined to
resolve apparent problems regarding the number and t^pe of
exempted vehicles. Finally, other options could include
waiving the high-altitude requirements for some vehicles to
allow their continued sale at higher elevations, or allowing
these vehicles to meet a somewhat less stringent high-altitude
standard. Therefore, the desirability of scenario"2 depends on
including appropriate exemptions or waivers to mitigate any
adverse effects on model availability.
c. Level of emission standards. The incremental costs
and benefits of scenario 2 are based on the assumption that
statutory standards would be promulgated at high altitude. The
previous discussion in this section demonstrated that
high-altitude regulations, as with any requirement, must be
chosen to moderate or eliminate a complex mixture of
potentially adverse impacts (e.g., environmental, economic,
energy, and model availability). The most efficient standards,
therefore, may be at some level other than the statutory
standards. While it appears that more stringent control beyond
the proportional standards is feasible perhaps down to the
levels of the statutory standards, less control may provide the
majority of the needed environmental benefit in a more
cost-effective manner. Only further study can identify the
optimum level of control.
D. Application Scenario 2 to LDGTs and LDDs
This report assumed that scenario 2 should also be applied
to gasoline-fueled light trucks and diesel-fueled vehicles. As
with LDGVs, the analyses for these other vehicle types were
conducted to determine the incremental costs and air quality
improvements of scenario 2 beyond those achieveable by
continuing the current standards (base scenario) for 1984 and
later model year vehicles.
1. LDGTs
a. Control technology. The technical requirements of
meeting the statutory LDGT standards in scenario 2 are
essentially the same as for LDGVs equipped with non-electronic
(nonfeedback) emission control systems.
b. Economic impact. The incremental economic impact
under scenario 2 should not be substantial. Depending on
whether the estimated fuel economy benefit is included, this
scenario would either result in a net savings or a cost of $7
million for the first 5 years of the regulations. As with
LDVs, this potential fuel economy benefit should be carefully
reevaluated as additional data become available.
-------
ES-12
c. Environmental impact. The air quality projections
show a positive impact from adding LDT control to that for
LDVs. Controlling LDGTs under scenario 2 would provide a 50
percent greater reduction in HC and a 40 percent greater
reduction in CO from vehicles sold during the first 5 years of
the regulations than if LDGVs were the only class of motor
vehicles subject to more stringent high-altitude standards.
d. Cost effectiveness. The addition of LDGTs to
scenario 2 makes it a slightly more cost-effective HC control
strategy than controlling only LDGVs, but does not change the
cost effectiveness of reducing CO under worst case assumptions.
2• LDDS
a. Control technology. Under the statutory gaseous
emission standards of scenario 2, light-duty diesel-powered
vehicles (LDDVs) would probably require a recalibration of
their exhaust gas recirculation (EGR) system by changing the
electronic control module at high altitude. Similarily,
light-duty diesel-powered trucks (LDDTs) would require a
recalibration of their non-electronic EGR systems at high
altitude.
The analysis of both proportional and statutory
particulate standards for light-duty diesel-powered vehicles
and trucks shows that a proportional particulate standard, when
definitively determined, will be readily achievable (there are
currently no such standards). in addition, the application of
gaseous emission control may achieve further reductions in
particulate with little or no addition effort. However, it is
not possible to determine if the 1S85 low-altitude particulate
standards will be fully achieveable at high altitude due to the
severe limitations of the data base. Additional data is needed
before a conclusive judgment can be made regarding the
technical feasibility of the 1985 statutory particulate
standards at high altitude.
b. Economic impact. The incremental cost of complying
with gaseous emission standards or proportional particulate
standards at high altitude during the first 5 years of the
regulations will be small.
c. Environmental impact. Adding control of
diesel-powered motor vehicles to scenario 2 should be
beneficial in further reducing HC emissions and should also
help reduce CO emissions.
a. Cost Effectiveness. Controlling gaseous emissions
from LDDs makes scenario 2 more cost effective than if LDGVs
and LDGTs are controlled separately.
-------
ES-13
III. CONTROL SCENARIOS BASED ON REVISED STANDARDS
Although revisions to the current statutory (low-altitude)
standards remain speculative, EPA has tried to anticipate the
effects of less-stringent standards on the previously
identified alternative scenarios. While there are a number of
different options for low-altitude standards, this study
assumes that if the statutory emission standards (g/mi) are
changed, the levels will be revised:
From To
0.41 HC 0.41 HC
3.4 CO 7.0 CO
1.0 NOX 1.5-2.0 NOX
The corresponding revised proportional standards (g/mi) at
5,300 feet would be:
Fr om To
0.57 HC 0.57 HC
7.8 CO 16.0 CO
1.0 NOX 1.5-2.0 NOX
However, the technical requirements of meeting an 16 g/mi CO
standard appear the same as those for meeting an 11 g/mi CO
standard. The Agency, therefore, assumed proportional standards
(g/mi) of:
1) 0.57 HC
2) 11.0 CO
3) 1.5-2.0 NOX
Revisions to the current
diesel particulate standards are
hence, they were not analyzed in
low-altitude evaporative HC and
not being considered by Congress,
this report.
To be consistent with the previous analysis the control
technologies, costs, and benefits of the alternative revised
high-altitude scenarios were evaluated relative to a revised base
scenario (fixed-point proportional standards based on revised
statutory standards). This approach is valid because compliance
with the revised proportional standards is expected to be similar
to compliance with current proportional standards. In both cases,
leaning the excessively rich fuel/air mixtures at high altitude is
the primary emission control technique.
A. Comparison of the Alternative Revised Scenarios
Several findings in the earlier analyses remain valid even
under revised standards. All continuous control strategies
evaluated continue to be unreasonably burdensome and are not cost
-------
ES-14
effective primarily because they would either seriously restrict
model availability at high altitude, require expensive and
complicated emission control technology on all vehicles, or
unnecessarily control emissions at elevations which are not
expected to have an air quality problem (above 6,000 feet). Also,
exemptions continue to be valuable in reducing the economic burden
of the standards or preventing model availability problems at all
elevations. Therefore, the fixed-point standards associated with
the revised base scenario (proportional standards) and the revised
scenario 2 (statutory standards) are the scenarios analyzed in
this portion of the study.
B. Evaluation of the Revised Scenario 2
1. Control Technology
Complying with revised statutory standards at low and high
altitude would require essentially the same control hardware as
previously estimated for complying with current statutory
standards. For LDGVs, this hardware includes the use of two
aneroids in addition to the one aneroid that would already be in
place to meet the revised proportional standards, for a total of
three. However, with revised statutory standards, the change in
low-altitude emission control technology may require that the air
pump system, which may be eliminated at low altitude on some
LDGVs, be replaced when these vehicles are sold at high altitude.
This may affect 40 percent of all high-altitude LDGV sales and is
included in the analysis as a worst case assumption. For LDDVs
sold at high altitude, manual adjustments will be needed in
addition to those that may already be needed for proportional
standards to limit the maximum fuel rate further and also to
recalibrate the fuel injection timing.
2. Economic impact
Complying with the revised scenario 2 would slightly increase
the purchase price of an average high-altitude LDGV. The purchase
price of LDDVs should not increase.
The incremental cost of fixed-point statutory standards with
revised levels is greatly influenced by the fuel economy benefit
the Agency expects from the use of high-altitude control
technology. Including this fuel savings in the cost of the
standards would result in a net incremental savings to the nation
during the first 5 years of the regulations. Excluding the
estimated fuel economi benefit from the calculation would cost the
nation up to $40 million during the first 5 years. Because of
this great variability, the fuel economy benefit should be
reevaluated as additional information becomes available.
In comparison to the incremental cost of fixed-point
statutory standards based on the current low-altitude
requirements, implementing the revised statutory standards could
-------
ES-15
be more costly to the nation if the estimated fuel economy
benefits of the standards are excluded. If the fuel benefits are
included, both types of standards will provide a net savings to
the nation.
The incremental purchase price for the average high-altitude
LDV should not affect a dealer's sales or a consumer's ability to
purchase a vehicle. One aspect of the revised scenario 2,
however, may have a significant negative impact on trading between
low- and high-altitude dealers. as discussed earlier, EPA has
made a worst case assumption that 40 percent of the high-altitude
fleet may require the addition of an air pump to meet the
statutory standards in the revised scenario 2. If true, these
particular vehicles may be prohibitively expensive to modify for
high-altitude use after production. This could result in model
availability problelns In some high-altitude areas. However, more
data is needed to confirm this potential effect.
2. Environmental Impact
Implementing statutory standards at high altitude over
proportional standards would reduce HC and CO emissions somewhat
less under the revised low-altitude standards than under the
current low-altitude standards.
There are no significant differences for ozone between the
high-altitude standards based on: 1) the revised low-altitude
standards, and 2) the current low-altitude standards. This is to
be expected, since the emission standards for HC are the same for
the respective statutory (0.41 g/mi) and base (0.57 g/mi) control
scenarios.
The same conclusion can be reached for CO with
Inspection/Maintenance, that is, there is no difference between
the number of violations under the two types of standards.
Without I/h, however, the number of CO NAAQS violations under the
scenarios based on revised standards is generally less than that
under the previous scenarios based on current standards. This
difference in CO NAAQS violations is a function of the assumed
catastrophic failure rates for currently used feedback emission
control systems. If these failure rates are significantly reduced
in the future as more experience is gained with these new systems,
then the observed difference in CO levels between the two types of
standards could be eliminated or even reversed.
For NOx, high-altitude standards based on revised
low-altitude standards will have a small negative impact on NAAQS
violations near the end of this century.
3. Cost Effectiveness
Implementing revised fixed-point statutory standards with
exemptions rather than revised fixed-point proportional standards
-------
ES-16
with exemptions would either provide further HC and CO emission
reductions at no net cost, or cost up to $1,250 per metric ton of
HC and up to $85 per metric ton of CO. The wide range of
incremental cost-effectiveness values is caused by the inclusion
or exclusion of the estimated fuel economy benefit that may
accompany implementating the revised statutory standards at high
altitude. In the worst case analyzed (i.e., no fuel economy
improvement) this scenario is nearly twice as costly per ton of HC
than the most expensive emission control strategy that has already
been implemented. For CO, it is more comparable to the other
strategies. On the other hand (i.e., inclusion of the estimated
fuel economy benefit) the revised scenario 2 is very cost
effective in relation to the other control strategies.
IV. CONCLUSIONS AND RECOMMENDATIONS
The existing proportional high-altitude standards have proven
valuable in improving the air quality of high-altitude areas in a
cost-effective manner. This report has analyzed various
high-altitude control scenarios to determine their incremental
costs and air quality benefits relative both to the current
standards and to less stringent revised standards. The evaluated
scenarios covered control options for both gasoline and
diesel-powered vehicles.
A. Control Scenarios Based on Current Standards
The major conclusions for each vehicle type are presented
separately below.
1. LDGVs
The Agency considered six alternative scenarios to continuing
the current fixed-point proportional standards. The costs of
these alternatives vari with their technical requirements, which
in turn are based on three basic factors:
1) the maximum elevation for which control must be
demonstrated;
2) the extent of exemptions, if any; and
3) the level of standards.
Based on these three factors, EPA concludes that:
1) Continuing the current statutory high-altitude
requirements, as mandated in section 206 of the Act, is extremely
costly, may significantly limit model availability at both low and
high altitudes, and is extremely cost ineffective;
2) there is no air quality justification for controls
above 6,000 feet;
-------
ES-17
3) statutory standards at high altitude can provide a
small incremental improvement in air quality;
4) exemptions, or similar waivers, can significantly
reduce compliance costs, while maintaining acceptable model
availability at higher elevation;
5) exemptions, or similar waivers, can prevent the
potential for adversely affecting model availability throughout
the nation that may accompany implementing statutory standards at
higher elevations in 1984, as required by the Clean Air Act; and
6) fixed-point statutory standards which require vehicles
sold above 4,000 feet to comply with the standards when tested at
5,300 feet and which provide for some exemptions are the most
cost-effective alternative beyond the current requirements, of the
six alternatives analyzed. Of course, there are many other
possible alternatives, one of which may be better than any of the
six analysed here.
2. LDGTS
In the case of LDGTs, EPA finds that controlling these
vehicles in addition to LDVs results in a positive impact on the
ambient air quality of high-altitude areas. Controlling light
trucks to statutory standards, while not specifically required by
the Act, would reduce vehicle emissions of HC by 50 percent more
and of CO by 40 percent more than if LDGVs were controlled alone.
In addition, the Agency finds that controlling light truck
emissions under fixed-point statutory standards is more cost
effective than the same degree of control for LDGVs.
3. LDDs
The Agency analyzed both gaseous emissions and particulate
emissions from LDDs. For gaseous emissions, EPA concludes that
achieving fixed-point statutory standards should be no more
difficult for diesel engines than for gasoline engines, and that
the cost over a 5-year period should be small. The Agency finds
that particulate emissions will be reduced by the same techniques
that reduce gaseous emissions, although it is too early to
determine if the statutory particulate standards can be met with
these techniques alone. Also, controlling the gaseous emissions
from LDDs to statutory high-altitude standards is expected to be
more cost effective than controlling LDGVs.
B. Control Scenarios Based on Revised standards
The Agency also considered the effects of the same six
alternative scenarios under revised standards. The major
difference is that under revised fixed-point statutory standards,
the control options might tend to reduce model availability
-------
ES-18
somewhat more than they would under the current fixed-point
statutory standards. Otherwise, EPA concludes that:
1) the technical difficulties of compliance would remain
about the same;
2) exemptions, or similar waivers, would retain their
positive effects of overall costs and model availability; and
3) fixed-point statutory standards would likely be the
most cost-effective alternative to proportional standards.
Based on the assumption that revised LDV standards at low altitude
are as stated above (i.e., 0.41 g/mi HC, 7.0 g/mi CO, and 1.5-2.0
g/mi NOx), these conclusions would remain valid for both
gasoline-fueled and diesel-powered cars and trucks. Nevertheless,
different conclusions are possible under scenarios which assume
other revised low-altitude standards.
C. Recommendations
EPA recommends that section 206 of the Clean Air Act be
amended to:
1) Provide the Administrator the flexibility to adopt
two-car compliance strategies, and to establish high-altitude
standards, within the range from proportional to statutory, for
any class of motor vehicles that is necessary to attain the NAAQS
for ozone and carbon monoxide after considering the technical
feasibility, impact on model availability, and economic impact of
any such requirements; and
2) Confirm the Administrator's authority to exempt certain
vehicles from the high-altitude certification requirements or
waive the high-altitude standards for certain vehicles, and to
decide on the maximum number of such exemptions or waivers.
-------
Chapter I
Introduction
I. AIR QUALITY IN HIGH-ALTITUDE AREAS
Many metropolitan areas located at hiyher elevations have
significant air quality problems. The automobile is an
important source of air pollution in these regions. For
example, in the rapidly growing automobile-oriented cities of
Denver, Albuquerque, and Salt Lake City, motor vehicles account
for more than half of all hydrocarbon (HC) emissions and almost
all of the carbon monoxide (CO) emissions. In combination with
summer sunlight and stable winter atmospheric conditions, these
emissions cause serious air pollution problems.
II. PAST AND PRESENT STANDARDS AFFECTING HIGH-ALTITUDE VEHICLES
To combat these problems, the U.S. Environmental
Protection Agency (hereafter referred to as EPA or the Agency)
established several programs to control emissions from motor
vehicles in high-altitude locations. As part of these past and
present regulatory programs, EPA has defined a high-altitude
location as any county with most of its land area located 4,0UU
feet above sea level.[1] This description includes much of
Colorado, Utah, Wyoming, New Mexico, and Idaho, and parts of
Nevada, Montana, Nebraska, Arizona, Oregon, and Texas. Though
California has counties above 4,000 feet, it sets its own
emission standards for motor vehicles.[1J
For the 1977 model year, EPA promulgated gaseous emission
regulations requiring all dealerships in high-altitude counties
(i.e., essentially areas above 4,000 feet) to sell only
light-duty vehicles (LDVs) and light-duty trucks (LDTs) that
were certified to meet special high-altitude standards when
tested at 5,300 feet (i.e., the location of test facilities at
Denver, Colorado). These standards were numerically identical
to the applicable emission standards at low altitude. The
numerical values of the 1977 standards were 1.5 grams per mile
(g/mi) HC, 15 g/mi CO, and 2.0 g/mi NOx for LDVs and 2.0 g/mi
HC, 20 g/mi CO, and 3.1 g/mi NOx for LDTs.
During the first model year these regulations were in
effect (1977), many vehicle models and optional engine
configurations available at low altitudes were unavailable at
high altitudes. Manufacturers chose to limit model
availability at higher elevations because the small percentage
(3 to 4 percent) of the market represented by high-altitude
-------
1-2
sales did not justify the costs required to develop
high-altitude emission control capabilities for all of their
vehicle configurations. Thus, high-altitude consumers could
not readily purchase approximately 50 percent of the vehicle
configurations offered at lower elevations.
These limitations on model availability affected
high-altitude consumers primarily in two ways: 1) some
consumers were unable to purchase the specific vehicle
configuration they wanted, and 2) if the desired model was
unavailable, the consumer may have had to pay significantly
more for an alternative, certified model. This additional cost
was as high as $500 in a small proportion of cases. A lesser
economic impact on consumers was the incremental cost of
high-altitude emissions control hardware, typically $20-40,
although it was as high as $194 on some imported models.[1]
Limited model availability affected manufacturers and
dealers to a lesser degree. Although it did not reduce total
vehicle sales at higher elevations, some dealers thought it
prevented the expected 10-20 percent growth in sales.
Moreover, some dealers complained that fleet sales were reduced
and that employee morale suffered. Also, near the perimeter of
high-altitude areas there was some difficulty in trading
vehicles between low- and high-altitude dealers.
As a result of these problems, Congress vacated the 1977
high-altitude regulations when the Clean Air Act (the Act) was
amended on August 7, 1977 . The Act also authorized EPA in
section 202(f) to reestablish high-altitude requirements, but
no sooner than the 1981 model year.
In response to these amendments, EPA revised the 1977
standards so that in 1978 and later model years, manufacturers
could voluntarily certify special high-altitude vehicles. In
1978, another voluntary program was also established. Under
this program, manufacturers could provide, with EPA's approval,
instructions explaining how vehicles operated at higher
elevations could be adjusted to improve performance,
significantly reduce emissions, and, in some cases, increase
fuel economy. These instructions were made mandatory on
October 8, 1980, under the authority of section 215 of the
Act. [2]
For 1981 model year vehicles, the year in which the
low-altitude standards became more stringent, EPA established a
voluntary program so that manufacturers could certify their
vehicles to separate "proportional" standards at high
altitude.[3] These voluntary gaseous emission standards were
the same as those promulgated on October 8, 1980, as the
current mandatory standards for 1982 and 1983 model year LDVs
and LDTs.[4]
-------
1-3
According to section 202(f) of the Act, proportional
high-altitude standards require a percent reduction in
emissions from 1970 vehicles at high altitude comparable to
that from the same vehicles at low altitude. in no case,
however, may the standard at high altitude be numerically less
than the corresponding standard at low altitude. This last
requirement is significant for NOx emissions which, unlike HC
and CO emissions, generally decrease with increasing altitude.
Table 1-1 presents the current proportional high-altitude
standards for 1982 and 1983 model year LDVs and LDTs, and the
equally stringent, but numerically smaller, low-altitude
standards for 1981 and later model years.
The current high-altitude regulations are structured to
minimize any negative impact on model availability by requiring
nearly all 1982 and 1983 models to either automatically meet or
be capable of being modified to meet the high-altitude
standards. Thus, each manufacturer's product line can be made
available to high-altitude purchasers if the manufacturer so
chooses. The Agency expects manufacturers will make almost all
models available once they have been certified.
The 1982 and 1983 regulations also reduce the potential
economic impact on the automotive industry by providing
exemptions for certain low-power vehicles, which perform poorly
at high altitude. Controlling the emissions of these vehicles
in a cost-effective manner is expected to be difficult. In
addition, by virtue of their poor performance, these vehicles
are not expected to be in demand by high-altitude consumers.
Therefore, by exempting low-power vehicles and allowing them to
be certified for sale at low altitude only, model availability
is not significantly affected in either low- or high-altitude
areas of the country and the cost of high-altitude regulation
is reduced without reducing the benefits.
III. REQUIREMENT TO ASSESS AND IMPLEMENT STANDARDS FOR 1984 AND
LATER MODEL YEARS
Section 206(f)(1) of the Clean Air Act, as amended in
1977, provides future high-altitude LDV standards by mandating
that:
All light-duty vehicles and engines manufactured during or
after model year 1984 shall comply with the requirements
of section 202 of this Act regardless of the altitude at
which they are sold.
Section 202 contains the current low-altitude gaseous exhaust
emission standards for LDVs. This section also contains the
authority used to promulgate the evaporative emission and
-------
Table 1-1
Current High- and Low-Altitude Emission Standards for
Light-Duty Vehicles and Light-Duty Trucks
Gaseous Standards
Altitude Vehicle
High
LDV[f]
LDT[h]
Year
1982-83
1982-83
Evap.
HC COta] NOx[b][c] HC[d]
(g/mi) (g/mi) g/mi (g/test)
0.57
2.0
7.8
26.0
1.0
2.3
2.6
2.6
Diesel
Particulate
Standards[e]
(g/mi)
[g]
[g]
Low LDV 1981-84 0.41 3.4 1.0 2.0 0.6
1985 and 0.41 3.4 1.0 2.0 0.2
later
LDT 1982-83 1.7 18 2.3 2.0 0.6
1984 0.8 10 2.3 2.0 0.6
1985 and 0.8 10 2.3 2.0 0.26
later
[a] If the CO standard for 1982 LDGVs is waived to 7,0 g/mi at low
altitude, the standard is 11.0 g/mi at high altitude.
[b] If the NOx standard for 1982 and 1983 LDDVs is waived up to 1.5 g/mi
at low altitude, the high-altitude standard is the same numerical
value.
[c] For 1982, American Motors Corporation must only meet an NOx standard
of 2.0 g/mi at both high and low altitude.
[d] Only applies to gasoline-fueled vehicles,
te] Only applies to diesel-powered vehicles.
[f] Light-Duty Vehicle.
[gj No particulate standard was promulgated for high-altitude vehicles,
but such a standard is analyzed in this report.
[h] Light-Duty Truck.
-------
1-5
diesel particulate standards applicable to these vehicles.
These 1984 and later model year standards are summarized in
Table 1-1.
in section 206(f)(2), the Administrator of EPA is required
to report to Congress on the economic impact and techological
feasibility of the "all-altitude" requirements found in
subparagraph (1) of that subsection. The report is also to
evaluate the technological feasibility and the health
consequences of separate proportional emission standards for
light-duty vehicles and engines in high-altitude areas, as
described earlier.
In preparing this report, the Agency has found that while
the nature of proportional requirement is clear, the exact
nature of the all-altitude requirement is not. Section
206(f)(1) and the accompanying legislative history can
plausibly be interpreted in various ways. For example, EPA
must determine whether vehicles must meet the standards of
section 202 at the highest altitude where they are sold, even
though there are no air quality problems in such areas, or at a
lower altitude, which can be justified on the basis of air
quality concerns. Another issue is whether exemptions from the
high-altitude standards (i.e., the all-altitude requirement)
are allowable.
Because the Agency has not yet taken a formal position on
these matters, the exact emission control requirements of
section 206(f)(1) are not clear at this time. Nevertheless,
this document responds to the Congressional mandate for such a
study by analyzing various control scenarios that encompass all
the possible interpretations of the all-altitude provision.
The report also explores some alternatives not currently
allowed in the statute, but does not evaluate all of the many
possible options. High-altitude emission control alternatives
are directly influenced by several factors, and probably more
importantly by the low-altitude standards. There is also
significant uncertainty in the prediction of in-use emission
factors from new technologies only recently introduced, and
this uncertainty bears directly on the Agency's ability to make
firm conclusions regarding the relative merits of some
alternatives. Certain conclusions can be confidently made and
have been. other conclusions are highly uncertain and have
been qualified as such. Hopefully, however, the options
analyzed provide an indication of the complexity of the various
interactions and a frame work for further analysis.
Furthermore, in an effort to identify the most effective
high-altitude regulatory strategy, we have examined other areas
of interest. The additional areas of investigation include:
-------
1-6
1) controlling light-duty trucks in addition to
light-duty vehicles; and
2) evaluating the consequences of possible revisions in
the current statutory (low-altitude) emission standards on the
control of emissions in high-altitude areas.
IV. ORGANIZATION AND SCOPE OF THE REPORT
Because light-duty gasoline-fueled vehicles (LDGVs)
dominate the motor vehicle fleet nationwide, any control
scenario that is unacceptable for them should be unacceptable
for the entire national fleet. Using LDGVs to screen the
various control scenarios reduces the complexity of the report,
but does not compromise identifying the most desirable strategy
for controlling emissions at high altitudes.
Thus, after the potential control scenarios are identified
in Chapter II, LDGVs are used to screen various strategies in
Chapters III through IV with regard to the requisite control
technology, the environmental and economic impacts, and their
cost effectiveness. using this information, the most
reasonable scenario of those analyzed is selected in Chapter
VII. Chapters VIII and IX then determine how the selected
scenario would effect controlling gaseous emissions from
light-duty gasoline-fueled trucks and light-duty diesel-powered
vehicles and trucks. Chapter X specifically evaluates the
selected scenario for its effect on controlling particulate
emissions from light-duty diesel-powered vehicles and trucks.
Recently, the debate concerning amending the clean Air Act
has included the possibility of revising the statutory
low-altitude emission standards for LDVs upward from the
current levels. Although such a revision remains speculative
at this time, chapter XI evaluates the effect of less stringent
low-altitude standards on potential high-altitude standards.
The final chapter (Chapter XII) presents EPA's conclusions
and recommendations.
-------
1-7
References
1. "Final Regulatory Analysis - Environmental and Economic
Impact Statement for the 1982 and 1983 Model Year High-Altitude
Motor Vehicle Emission standards," U.S. EPA, OANR, OMS, ECTD,
SDSB, October, 1980.
2. "Control of Air Pollution from New Motor Vehicles and New
Motor Vehicle Engines* Submission of Altitude Performance
Adjustments," U.S. EPA, 45 FR 66952, October 8, 1980.
3. "Control of Air Pollution from New Motor Vehicles and New
Motor Vehicle Engines; High-Altitude Emission Standards,
Voluntary Compliance Program for 1981 Model Year Light-Duty
Motor Vehicles," U.S. EPA, 45 FR 20402, March 27, 1980.
4. "Control of Air Pollution from New Motor Vehicles and New
Motor Vehicle Engines; Final High-Altitude Emission standards
for 1982 and 1983 Model Year Light-Duty Motor Vehicles," U.S.
EPA, 45 FR 66984, October 8, 1980.
-------
Chapter II
Identification of the High-Altitude
Control Scenarios
I. INTRODUCTION
The high-altitude control scenarios analyzed in this
document will be chosen from a variety of possibilities. This
screening is necessary to limit the analysis to manageable
proportions and to avoid needless elaboration on less likely
control options. The selection process in this chapter will be
conducted in two steps. First, the relevant variables that
form the basis for the alternative scenarios will be discussed
to provide a thorough understanding of the control strategies
and to introduce information that will be used to select the
specific scenarios for the study. Second, these variables will
be combined into the various potential scenarios, and those not
justifying further consideration at this time will be
eliminated. At this point in the selection process, the
scenarios that represent what currently appears to be the range
of possible interpretations of the section 206(f)(1)
"all-altitude" requirement will also be identified.
II. THE BASIS FOR THE CONTROL SCENARIOS
There are many possible strategies for controlling motor
vehicle emissions in high-altitude areas of the country. These
strategies are a combination of variables which induce: 1)
the philosophy of high-altitude emission control, 2) the
maximum altitude at which the standards will apply, 3) the
allowable emission levels for each standard, and 4) the
availability of exemptions for certain low-power vehicles.
Each of these variables is discussed below.
A. The Philosophy of High-Altitude Control
This report considers two basic certification options.
The first option requires that motor vehicles be certified to
meet the applicable standards continuously at all elevations
without modification. Scenarios that require demonstrating
compliance in this manner are referred to as "continuous"
standards in this document.
The second option is primarily patterned after the
certification program that was promulgated in the 1982 and 1983
high-altitude regulations.[1] This rule specifies that all
vehicles subject to the regulations must be capable of meeting
high-altitude standards either automatically or by
modification. Furthermore, demonstrating compliance with these
standards is limited to a single fixed-point of 5,300 feet. If
modifications are necessary to meet the emission standards at
5,300 feet, those modifications must be made to all such
-------
II-2
vehicles sold above 4,000 feet. Scenarios which incorporate
this type of certification program are referred to as
"fixed-point" standards.
B. The Maximum Elevation Required for Control
This variable applies only to the continuous standards,
since the fixed-point standards analyzed in this report only
require compliance with high-altitude regulations at a single
elevation of 5,300 feet. Under fixed-point standards
"high-altitude" vehicles are sold at any elevation above 4,000
feet.
Continuous standards require that a vehicle be designed to
compensate automatically for the effects of reduced air density
as altitude increases. Therefore, choosing the maximum
altitude at which emission control must be demonstrated is
important, since the wrong decision could either: 1) cause
manufacturers to equip their cars with costly control hardware
to reduce emissions at altitudes where such reductions are not
warranted by air quality needs, or 2) lead to the absence of
additional control in a significant number of regions that do
indeed need it. Because this issue affects the entire
analysis, it is discussed at length below.
A literal reading of the Act implies that vehicles must be
certified up to the highest altitude at which sales occur,
approximately 10,200 feet. However, the intent underlying the
provision may be satisfied with an alternative interpretation.
That interpretation, which is supported by the applicable
legislative history, requires compliance only up to a certain
elevation. A key purpose of section 206(f) is to improve the
air quality in high-altitude areas of the country that violate
the National Ambient Air Quality standards (NAAQS). The air
quality monitoring data currently available indicate that
future violations will be limited to high-altitude areas
substantially below 10,200 feet.
Colorado Springs, Colorado, at 6,012 feet and the Tahoe
Air Basin in California and Nevada at 6,225 feet are the two
highest areas designated as nonattainment areas for carbon
monoxide (CO) or ozone (O3) as specified by section 107 of
the Clean Air Act. That is, they are violating one or more of
the NAAQSs referred to in Table Il-l. Thus, an elevation of
approximately 6,000 feet forms a logical upper boundary above
which all regions are likely to be attaining the NAAQSs.
This is especially true with regard to ozone since it
normally is only a problem in areas that are more densely
populated (and that have higher emission densities of
hydrocarbons (HC) and nitrogen oxides (NOx)) than those found
above 6,000 feet. Another factor that is critical to the
formation of ozone but that is not prevalent in these high
elevations is the presence of stagnated high-pressure cells,
-------
II-3
Table II-l
National
Ambient Air quality Standard
s
Pollutant
Averaging Time
Standards
Ozone
1 Hour
235
uy/m^[aJ
Carbon Monoxide
8 Hour
1 Hour
10
40
ing/m-* [ b]
mg/m^
Nitrogen Dioxide
Annual Average
100
ug/m-*
Sulfur Dioxide
Annual Average
24 Hour
80
365
ug/m^
ug/m3
Suspended Particulate
Matter
Annual Geometric Mean
24 hour
75
260
uy/iu^
ug/m^
Lead
Quarterly
1.5
ug/m^
I a J Micrograms per cubic meter,
[b] Milligrams per cubic meter.
-------
II-4
such as those common to the Los Angeles area. These
meteorological conditions are characterized by long periods of
cloudless skies and essentially no winds, which combine to
provide ideal conditions for producing and retaining ozone.
CO, however, is a more localized problem. CO violations
can occur whenever localized traffic is heavy enough for a
sufficient time period. To determine the extent of CO
violations between the altitudes of approximately 6,000 feet
and 10,000 feet, EPA's, Region VIII monitored ambient CO levels
at these elevations during the 1978-7S winter season.[2]
Two CO monitors were available for this study. Region
VIII used the following criteria to select the best locations:
1) the level of CO emissions in the area, 2) the frequency and
severity of unfavorable meterological dispersion conditions, 3)
the altitude, 4) the availability of suitable monitoring sites,
and 5) the availability of local support for the monitoring.
One site was chosen at 10,500 feet near the Loveland Basin
ski area. Being near the east entrance of the Eisenhower
Tunnel, this site was exposed to one of the largest sources of
CO emissions in the Rocky Mountains. It was also near two
other significant sources of CO emissions: the highway to
Loveland Pass and the Loveland ski area parking lot. The ski
area complex also provided a significant source of population
exposure. The monitor was located in a trailer approximately
20 meters south of U.S. Highway 6, 50 meters south of 1-70, 300
meters east of the ski area parking lot, and 1,000 meters east
of the tunnel's entrance. This site, operated during the high
traffic ski season, was expected to measure the highest ambient
CO levels at an altitude of 10,000 feet or above.
At the elevation 8,150 feet, Vail, Colorado was selected
for the other CO monitoring site for three primary reasons.
One, Vail is a major ski area located near a major traffic
artery (1-70) and has a high level of CO from automobile
emissions. A significant additional source of CO is wood
burning in residential fireplaces. Two, Vail is located in a
deep, narrow valley and experiences some of the poorest
atmospheric dispersion conditions in the Rockies. Three, Vail
has an ongoing program that monitors air quality. It is
headquartered in City Kail approximately 50 meters south of
1-70 and 50 meters west of one of Vail's major intersections.
Established quality control procedures were followed.
No violation of the CO standard was observed at the
Loveland Basin site. Thus, controlling CO to altitudes of
10,000 feet or above should be unnecessary. However, there
were 27 violations of the CO 8-hour running average NAAQS of 10
mg/m^ during December 1978 and January 1979 at the Vail
site. The highest measured 8-hour concentration was 12.6
mg/m , 26 percent above the NAAQS. (The term 8-hour running
-------
II-5
average means that a new 8-hour average is taken every hour;
thus, as many as 24 violations could occur in a given day.)
The first five violations occurred immediately after the
instrument became operational on December 11, 1S78. Therefore,
more violations might have been recorded if the analyzer had
been put into operation sooner. Also, many of the 27
violations occurred on the same day.
This sampling was done when the low-altitude CO emission
standard for light-duty vehicles (LDVs), the largest source of
CO emissions, was 15 g/mi. In high-altitude areas, CO
emissions from in-use 1978-79 LDVs averaged 53.69 g/mi halfway
through their expected 100,000-mile lifetime.[3] In 1980, the
low-altitude CO standard for LDVs dropped to 7 g/mi and was
further lowered to 3.4 g/mi beginning with the 1981 model year.
These two reductions have lowered the average CO emission level
from in-use 1980 and 1981 LDVs to an average of 32.52 g/mi and
26.13 g/mi in high-altitude regions, respectively (Appendix
II). [3] Implementing the 1982 and 1983 (interim) high-altitude
standards will further reduce the in-use LDV CO emission level
to an average of 23.54 g/mi. This will be a net reduction of
approximately 56 percent from the 1979 model year level
(Appendix II). Thus, as newer, cleaner vehicles replace the
pre-1980 models (those affecting the results from the Loveland
and Vail studies), CO concentrations at these sites will
significantly decrease. Therefore, in all likelihood, Vail
will be in compliance with the CO NAAfcS in the future without
more stringent high-altitude control.
It should be pointed out that the Motor Vehicle
Manufacturers Association (MVMA) monitored CO in Leadville,
Colorado, from January to March 1980.14] However, because the
quality control procedures it used were not documented, the
study's findings are questionable. The results seem to
indicate that the CO levels in Leadville are less than those of
Vail. Leadville was considered as a test site for this study
since it is the highest U.S. city with an automobile dealership
and, hence, is the altitude up to which manufacturers may have
to comply with the low-altitude standards. It was rejected for
the EPA study, however, for two reasons: 1) low traffic
densities (i.e., low CO emission production), and 2) generally
good atmospheric dispersion characteristics.
In conclusion, no region above approximately 6,000 feet is
expected to have an oz.one problem mainly because the emission
density is not high enough. The only identified
automotive-related air quality problem at these elevations
concerns CO and is in Vail, Colorado, at an elevation of 8,150
feet. This problem, however, should definitely disappear when
cleaner 1980 and later model year vehicles replace their older,
higher-emitting counterparts, particularly given the presence
of interim high-altitude standards. Thus, there do not appear
to be significant air quality problems above approximately
-------
II-6
6/000 feet which would warrant control in these areas. The
maximum control altitudes for scenarios with continuous
standards are, therefore, limited to 10,200 feet (the highest
elevation where control may be required by the Act), and 6,000
feet (the highest elevation where emission controls appear to
be justifiable based on air quality concerns).
C. The Levels of the Standards
This report considers two types of high-altitude emission
standards. The first type are called "statutory" standards.
Section 206(f)(1) of the Act requires that, beginning in the
1984 model year, all LDVs must comply with the statutory
emission standards as authorized in section 202 regardless of
altitude. Thus, statutory high-altitude standards would be the
same numerical value as the existing low-altitude standards.
For LDTs, the Agency has the option of promulgating
emission standards under the general provisions of section
202(a) of the Act. This section authorizes the Administrator
to establish regulations that are necessary to protect the
public health and welfare. In the 1984 model year, statutory
HC and CO standards for LDTs become more stringent than in
previous years. These statutory standards could also be
implemented at high altitude. In this analysis, statutory
high-altitude standards for LDTs are considered for
implementation beginning in the 1984 model year.
The current gaseous and particulate emission standards are
shown in Table II-2. Under the current Congressional mandate,
it is clear that the numerical values for the LDV standards
would remain unchanged regardless of the altitude at which
compliance is required.
The second type of emission standards are referred to as
¦proportional" standards. These standards generally represent
the same reduction in vehicle emissions at high altitude as the
statutory standards require at low altitude. The levels of the
proportional gaseous emission standards for 1984 and later
model year LDVs were determined in the recent interim (1982 and
1983) high-altitude rulemaking action for an elevation of 5,300
feet.[1]
With regard to proportional gaseous emission standards for
LDTs, it was previously stated that the statutory
(low-altitude) light truck standards for HC and CO become more
stringent in 1984. These new standards will cause
manufacturers to redesign and, in turn, recertify every LDT in
their product lines at low altitude. Although at the time this
document is being prepared no high-altitude standards have been
promulgated for 1984 and later LDTs, the Agency intends to have
some type of requirement in force beginning in that year.
(This is consistent with president Reagan's recent announcement
-------
Table II-2
Current Statutory and Corresponding Proportional Emission
Standards for Light-Duty Vehicles (LDVs) and Light-Duty Trucks (LDTs)
Gaseous
Standards
Diesel
Evap.
Particulate
Type of
Year of
HC
CO
NOx
HClaJ
Stanuarus
Standard
Vehicle
Implementation
(g/mi)
(g/mi)
(g/mi)
(y/test)
( y/l.li ) I D J
Statutory
LDVLc]
1984
0.41
3.4
1.0
2.0
0.0
(Low-
1985
0.41
3.4
1.0
2.0
0.2
Altitude)
LDT[d]
1984
0.8
10
2.3
2.0
0.6
1985
0.8
10
2.3
2.0
0.26
Propor-
LDV
1984
0.57
7.8
1.0
2.6
LeJ
tional
1985
0.57
7.8
i.O
2.6
[ej
(High-
LDT
1984
1.0
14
2.3
2.6
LeJ
(Altitude)
1985
1.0
14
2.3
2.0
[ej
iH
M
[a] Evaporative emission standards do not apply (N/A) to diesel-powered
vehicles. The low volatility of diesel fuel produces few evaporative
emissions.
[b] Particulate standards apply only to diesel-powered vehicles anu trucks.
[c] Light-duty vehicle.
[dj Light-duty truck.
[e] No particulate standards have been set for high-altitude venicies or trucks
(see Chapter X).
-------
II-8
concerning regulatory relief for the automobile industry.)
Regardless of the stringency of such standards, manufacturers
will have to develop and certify their newly designed LDTs at
high altitude also.
EPA believes that the incremental cost of complying with
proportional standards, which are based on the more stringent
1984 low-altitude LDT standards, should be relatively
inexpensive. This should be true because compliance with the
interim standards for LDTs is relatively inexpensive and those
standards reflect approximately the same level of high-altitude
control technology as would be required to meet the more
stringent proportional standards beginning in 1984 for
LDTs.[3,5} Therefore, proportional control relative to the new
statutory standards should be cost effective. EPA determined
the proportional gaseous emission standards for LDTs that are
equivalent to the 1984 statutory requirements in the proposal
for interim high-altitude regulations.[5] Thus, proportional
high-altitude standards for gaseous emissions have already been
established for LDVs and LDTs.
The interim high-altitude rulemaking action, however, did
not consider particulate emissions from diesel-powered
vehicles. Such standards now exist for both LDVs and LDTs
under the authority of section 202(a) of the Act. Although the
Act does not specify a procedure for setting proportional
particulate standards, if the guidelines of section 202(f) for
determining proportional gaseous emission standards were
followed, particulate standards would be based on high-altitude
emissions from diesel-powered vehicles manufactured during the
1970 model year, such a study has never been performeo. Even
if it were, it would be of limited usefulness since the great
majority of diesel-powered vehicles sold today were not
produced in 1970. (Only a few of diesel models were available
in 1970 and these were sold in relatively small quantities.)
Unfortunately, no comprehensive study of the effect of high
altitude on later model year diesels is available, either.
Thus, proportional particulate standards will have to be
estimated from the available data. This will be done in
Chapter X.
Table II-2 summarizes the proportional gaseous emission
standards which are used in analyzing fixed-point scenarios.
For continuous scenarios, the numerical value of the
proportional standard is different at each elevation. since
emission standards have only been determined for two elevations
at this time (i.e., low altitude up to 1,800 feet and high
altitude at 5,300 feet), the proportional standard at any other
altitude can be found by assuming that a linear relationship
exists between elevation and vehicle emissions. In other words,
the proportional standard for altitudes between 1,800 feet and
5,300 feet lies along a straight line between the known
emission standards at these two altitudes. This is graphically
depicted for CO in Figure Il-l, which also shows that the
-------
11—9
FIGURE II-1
Graphic Example of All-Altitude and Proportional
Reduction Standards Based on Current Automobile Requirements
Key:
10
-7.8
a4
Proportional CO Reduction Standard
— — — — All-Altitude CO Statutory Standard
DENVER TEST
SITE
(5,300 FT.)
STATUTORY
(1,800 FT.)
I >
10
ALTITUDE (thousands of feet)
-------
11-10
proportional standard for altitudes above 5,300 feet is found
by extending the line to the desired elevation.
The statutory CO standard is included in Figure Il-l for
comparison. As shown, this type of standard represents an
increasingly more stringent requirement at higher elevations
when compared to the proportional standard.
D. High-Altitude Exemptions
It may be technically difficult to modify some vehicles to
meet the various high-altitude standards shown in Table II-2 in
a cost-effective manner or# as might occur in some instances,
in a safe manner. These vehicles generally should be
low-power, high fuel economy cars and trucks that perform
acceptably at low altitude but poorly at higher elevations.
Their poor performance arises from using smaller engines and
low numerical axle ratios for improved fuel economy.
The less dense air at high altitude provides less oxygen
(per unit of volume) to combust the fuel/air mixture in the
engine's cylinders. This reduces the engine's ability to
produce power. Some of this lost power, however, can be
recovered by adding extra fuel to the engine through the
power-enrichment system of the fuel-metering device. This
results in a richer fuel/air ratio which, in turn, produces
more power when it is burned, unfortunately, richer fuel/air
mixtures also produce excessive HC and CO emissions.
At low altitude, the deleterious effects on emissions from
using power-enrichment rarely affects compliance with emission
standards because the system is seldom engaged during the
Federal Test Procedure (FTP), the test used to measure
compliance. Power-enrichment operation is increasingly more
frequent as altitude increases and drivers attempt to
compensate for lost vehicle performance. Therefore, it can
become much more difficult to control emissions and maintain
acceptable vehicle operation at successively higher elevations.
There are two principal reasons for exempting such
low-power vehicles from high-altitude standards, as briefly
alluded to above. The first reason is to reduce the economic
burden of the standards by saving manufacturers the needless
expense of developing and certifying these vehicles for high
altitude when they are either not normally sold there or are
sold there in only small numbers because of their poor or
unacceptable performance. For this reason, exemption criteria
were included in the 1982 and 1983 high-altitude regulations.
The second and more compelling reason primarily affects
control strategies that require vehicles to meet the stringent
statutory standards at higher elevations without modification.
For many low-power vehicles, compliance with these standards at
-------
11-11
high altitude may degrade performance to such low levels that
the vehicles may actually become unsafe to use at higher
elevations. Rather than market such potentially unsafe
vehicles, manufacturers would probably remove them from the
national market. This would adversely affect model
availability and fuel economy at low altitude, which accounts
for about 97 percent of the total market.
Regardless of the reason for exemptions, such vehicles
would not be allowed to be sold in high-atitude areas with air
quality problems in order to maximize the environmental
benefits of the regulations. In most cases, however, these
exemptions would not seriously affect model availability at
high altitude because such low-power vehicles would not
normally be sold in these areas, as noted above.
For theses reasons, this analysis evaluates exemptions
based on performance. As more information becomes available,
however, it may actually be preferable to implement other
types of exemption schemes. For example, waivers could be
granted to certain vehicles that could then be sold at high
altitude. Such an important determination, however, is not
within the scope of this study. It is more appropriately made
as part of the rulemaking process used to establish any new
regulations, depending on available statutory authority.
At present, estimating the number of exemptions that may
be required in the 1984 high-altitude regulations is
speculative. Not enough data are available from the 1982 and
1983 high-altitude program to estimate the number of vehicles
needing exemptions accurately. In addition, fuel economy
pressures are forecast to significantly change to the motor
vehicle fleet in the future. such changes may include
continued downsizing (weight reduction) of the fleet, which
could manifest itself in the need for more exemptions than may
currently be expected. Conversely, as described in Chapter
III, EPA estimates that to comply with the current statutory
(low-altitude) emission standards, vehicle manufacturers will
increasingly rely on more sophisticated electronic control
systems having the inherent capability to significantly control
emissions at high altitude significantly with little or no
modification. Such systems could reduce the need for
exemptions in the future. in fact, the beneficial aspects of
these new emission control devices may more than offset the
negative effects that vehicle downsizing has on exemptions.
The following estimates for the various control scenarios
are, therefore, based primarily on: 1) experience in
developing the 1982 and 1983 high-altitude standards, 2) the
knowledge that at successively higher elevations the technical
difficulty of achieving the standards is greater, and 3) the
-------
11-12
fact that compliance with more strident standards . is more
difficult to achieve (i.e., statutory versus proportional
standards, or continuous versus fixed-point requirements).
EPA estimates the following maximum (i.e., worst case)
volume of exemptions for each scenario:
1. Five (5) percent of the fleet for scenarios with
fixed-point proportional standards;
2. Ten (10) percent of the fleet for scenarios with
continuous proportional standards up to 6,000 feet;
3. Fifteen (15) percent of the fleet for scenarios with
fixed-point statutory standards;
4. Twenty-five (25) percent of the fleet for scenarios
with continuous statutory standards up to 6,000 feet;
5. Forty (40) percent of the fleet for scenarios with
continuous proportional standards up to 10,200 feet; and
6. Sixty (60) percent of the fleet for scenarios with
continuous statutory standards up to 10,200 feet.
III. IDENTIFYING AND SELECTING THE CONTROL SCENARIOS
This report coulo analyze 12 possible scenarios, as
presented in Table n-3. They are combinations of the four
variables just discussed: 1) continuous or fixed-point
certification requirements, 2) the maximum elevation of
control, 3) the levels of the standards, and 4) the possibility
for exemptions.
A. Eliminating Five Scenarios
Of the 12 scenarios that have been identified, several can
be discarded without compromising the analysis. All of the
scenarios requiring emission control up to 10,200 feet are
likely candidates for elimination. The air quality information
previously presented in this chapter showed that no future
NAA^S violations are likely in areas above approximately 6,000
feet. In addition, only one of the four possible scenarios
with a ceiling of 10,200 feet need be retained to represent
what is widely believed to be the most stringent of the
possible interpretations of the Congressionally mandated
program. This scenario is shown as Number 1 in Table II-3 and
requires continuous statutory standards without exemptions and
a ceiling of 10,200 feet. Finally, the continuous statutory
and continuous proportional standards that may require
exempting 60 and 40 percent, respectively, of the motor vehicle
fleet are
-------
11-13
Table II-3
Potential High-Altitude Control scenarios
Number Description
1-4 Continuous statutory standards:
1. Without exemptions and a ceiling of 10,200
feet.
2. With exemptions (60 percent) and a ceiling of
10,200 feet.
3. Without exemptions and a ceiling of 6,000 feet.
4. With exemptions (25 percent) and a ceiling of
6,000 feet.
5-8 Continuous Proportional Standards:
5. Without exemptions and a ceiling of 10,200
feet.
6. With exemptions (40 percent) and a ceiliny of
10,200 feet.
7. Without exemptions and a ceiling of 6,000 feet.
8. With exemptions (10 percent) and a ceiling of
6,000 feet.
9-10 Fixed-Point Statutory standards (certification at
5,300 feet):
9. Without exemptions.
10. With exemptions (15 percent).
•
11-12 Fixed-Point Proportional Standards (certification
at 5,300 feet):
11. Without exemptions.
12. With exemptions (5 percent).
-------
11-14
not viable options because they could severely restrict model
availability at higher elevations (above 1,800 feet). This
would negate the reason Congress vacated the 1977 high-altitude
standards. Therefore, three control scenarios are eliminated
from further study:
1. Continuous statutory standards with exemptions and a
ceiling of 10,200 feet (Number 2 in Table II-3);
2. Continuous proportional standards without exemptions
and a ceiling of 10,200 feet (Number 5 in Table II—3); and
3. Continuous proportional standards with exemptions
and a ceiling of 10,200 feet (Number 6 in Table 11-3).
Two additional scenarios can also be eliminated. The
fixed-point proportional standards are evaluated as a
continuation of the 1982 and 1983 high-altitude regulations.
Since exemptions already have been found to be necessary in
these regulations, it is likely that exemptions will remain
necessary in 1984 and later model years. Also, the fixed-point
statutory standards are considered in this report primarily to
represent a variation of the 1982 and 1983 high-altitude
regulations. As such, it is relevant to consider implementing
statutory standards at 5,300 feet with exemptions, while
retaining the provision of the 1982 and 1983 program that
allows vehicles to be specifically modified for sale at high
altitude (this latter provision would likely require statutory
changes). This scenario is shown as Number 10 in Table II-3.
Thus, the two scenarios which can be eliminated are:
1. Fixed-point proportional standards without
exemptions (Number 9 in Table I1-3); and
2. Fixed-point statutory standards without exemptions
(Number 11 in Table II-3).
B. Categorizing the Remaining Scenarios
With this elimination, seven scenarios remain to be
analyzed. For clarity, the scenarios can be grouped into four
broad categories with specific variations listed under each
category. This hierarchy is presented in Table II-4 and will
be referred to throughout this document.
One further remark concerning the analytical methodology
of the report is necessary before proceeding with the
analysis. All alternative strategies are evaluated by
comparing them with a continuation of the 1982 and 1983
(interim) high-altitude standards. EPA believes that the value
of these interim regulations was proved during the recent final
rulemaking process, and that little benefit would result from
-------
11-15
Table II-4
Control scenarios Selected for Evaluation
1. Continuous Statutory Standards:
a. Without exemptions and a ceiling of 10/200 feet.[a]
b. With exemptions and a ceiling of 6,000 feet.
c. Without exemptions and a ceiling of 6,000 feet.
2. Fixed-Point (5,300 feet) Statutory standards, with
exemptions.
3. Continuous Proportional Standards:
a. With exemptions and a ceiling of 6,000 feet.
b. Without exemptions and a ceiling of 6,000 feet.
4. Fixed-Point (5,300 feet) Proportional standards, with
exemptions (referred to as the "base" scenario).
[a] Scenario la appears to be consistent with the most
stringent interpretation of section 206(f)(1) of the Act.
-------
11-16
again presenting the detailed analyses that support those
standards.[3,7,8] The technical requirements of fixed-point
proportional standards were found to be readily feasible. In
fact, some vehicles could already comply with the emission
levels with no changes to their original low-altitude hardware
designs or control settings. A significant air quality
improvement was also forecast to occur by reducing the
pollution from 1982 and 1983 high-altitude vehicles. For
example, in Denver, Colorado, the total HC emissions would be
reduced by up to 1.0 percent and CO emissions would be reduced
up to 3.4 percent. The proportional standards were also found
to be cost effective, at $393 per metric ton for HC and $12 per
metric ton for CO. Therefore, through the remainder of the
analysis the fixed-point proportional standards are referred to
as the "base" scenario (Table II-4), and are specifically
analyzed in this report as is required to complete the analysis
of alternative control scenarios. Appendix I contains more
information on the costs ana benefits of the base scenario.
-------
11-17
References
1. "Control of Air Pollution from New Motor Vehicles
and New Motor Vehicle Engines Final High-Altitude Emission
Standards for 1982 and 1983 Model Year Light-Duty Motor
Vehicles," U.S. EPA, 45 FR 66984, October 8, 1980.
2. "High Altitude and Street Canyon Carbon Monoxide
Monitoring in Region VIII During the Winter of 1978-79," U.S.
EPA, Region VIII, Denver, CO.
3. "Final Regulatory Analysis - Environmental and
Economic Impact Statement for the 1982 and 1983 Model Year
High-Altitude Motor Vehicle Emission Standards," U.S. EPA,
OANR, OMS, ECTD, SDSB, October 1980.
4. "Carbon Monoxide Data - High Altitude, January-March
1980," Motor Vehicle Manufacturer's Association, April 1980.
5. "Proposed High-Altitude Emission Standards for 1982
and 1983 Model Year Light-Duty Motor Vehicles," U.S. EPA, 45 FR
5988, January 24, 1980.
6. Control of Air Pollution from New Motor Vehicles and
New Motor Vehicle Engines: Gaseous Emission Regulations for
1985 and Later Model Year Light-Duty Trucks and 1986 and Later
Model Year Heavy-Duty Engines," U.S. EPA, OANR, OMS, ECTD,
SDSB, 45 FR 5838, January 19, 1981.
7. "Summary and Analysis of Comments on the Notice of
Proposed Rulemaking for High-Altitude Emission Standards for
1982 and 1983 Model Year Light-Duty Motor Vehicles," U.S. EPA,
OANR, OMS, ECTD, SDSB, October 1980.
8. "Technical Feasibility of the Proposed 1982-83
High-Altitude Standards for Light-Duty Vehicles and Light-Duty
Trucks," U.S. EPA, OANR, OMS, ECTD CTAB, August 1980.
-------
Chapter III
Technology Assessment
I. INTRODUCTION
In this chapter, the control technology expected to be
required for light-duty gasoline-fueled vehicles (passenger
cars) to comply with the various high-altitude control
scenarios under consideration will be discussed. In
particular, the control technology required by each alternative
scenario over and above that required by the base scenario will
be identified. The potential control scenarios were determined
in the previous chapter and are summarized below for
convenience.
A. Base Scenario: Fixed-Point Proportional standards with
Exemptions
This scenario will require vehicles to comply with
high-altitude standards of 0.57 g/mi HC, 7.8 g/mi CO, 1.0 g/mi
NOx, and 2.6 g/test evaporative HC at only one elevation (i.e.,
5,300 feet). It is essentially a continuation of the current
high-altitude requirements for 1982 and 1983 model year
vehicles.[1] Exemptions may be granted for certain low-power
vehicles that would perform unacceptably at high altitude and
that may have technical difficulty in meeting the standards
cost effectively.
B. Scenario 1; Continuous statutory Standards
This alternative scenario will require vehicles to comply
with standards of 0.41 g/mi HC, 3.4 g/mi CO, 1.0 g/mi NOx, and
2.0 g/test evaporative HC (the current low-altitude standards)
is required at all elevations up to a maximum altitude. This
scenario is subdivided further, depending on the maximum
altitude to which compliance must be demonstrated and on
whether performance-based exemptions are provided.
1. la - Compliance required up to 10,200 feet and no
exemptions allowed;
2. lb - Compliance required up to 6,000 feet with
exemptions allowed; and
3. lc - Compliance required up to 6,000 feet and no
exemptions allowed.
C. Scenario 2: Fixed-Point Statutory Standards with
Exemptions
This strategy is similar to the base scenario, except that
at 5,300 feet vehicles must meet the low-altitude statutory
emission standards (0.41 g/mi HC, 3.4 g/mi CO, 1.0 g/mi NOx and
-------
III-2
2.0 g/test evaporative HC) instead of the proportional
standards presented in the base scenario.
D. Scenario 3: Continuous Proportional standards
Under this control strategy, vehicles must meet standards
that increase proportionally with altitude up to 6,000 feet.
At 1,800 feet, the emission standards are the low-altitude
standards; at 5,300 feet, they are the high-altitude standards
outlined in the base scenario. The standards vary linearly in
between these two altitudes and up to 6,000 feet. Like
scenario 1, this scenario has two variations:
1. 3a - Exemptions are allowed.
2. 3b - No exemptions are allowed.
The above scenarios differ with respect to their basic
approach to solving the high-altitude emissions problem. The
base scenario and scenario 2 are termed "two-car" strategies
since they allow vehicle modifications to be performed on
vehicles sold for principal use above 4,000 feet. This is not
the case for scenarios la, lb, lc, 3a, and 3b. Any
modifications which are necessary to satisfy the high-altitude
requirements in these scenarios must be performed on all
vehicles regardless of the altitude at which they are sold.
These scenarios are termed "one-car" strategies.
The technical analysis of the various control scenarios is
presented in three separate sections. First, the effect of the
reduced air density found at higher elevations on regulated
gaseous emissions from current automotive systems will be
discussed. In addition, the ability of current low-altitude
control systems to meet the alternative standards and the
techniques available to reduce high-altitude emissions will be
described. Second, the requisite control technology for each
scenario will be estimated. Finally, any potential adverse
effects of high-altitude standards on low-altitude control
technology will be assessed.
II. THE EFFECTS OF INCREASING ALTITUDE ON EXHAUST EMISSIONS
As altitude increases, the density of air decreases. in a
conventional (e.g., carbureted) fuel-metering system (the type
found on most cars today), the amount of fuel metered is a
function of the velocity of air passing through a venturi
tube. Since the density of air is lower at high altitude, the
mass of air (and oxygen) corresponding to a given mass of fuel
is less than that occurring at low altitude. Therefore, as
altitude increases, the fuel/air ratio that enters the
combustion chamber will increase, or become richer.
-------
III-3
The production of emissions from a gasoline engine is very
sensitive to this fuel/air ratio. As the ratio increases, HC
and CO emissions increase markedly, because not enough oxygen
is available to burn the fuel completely. At the same time,
NOx emissions decrease, because the peak combustion temperature
is lower with rich fuel/air mixtures. Thus, in order to meet
emissions standards for HC and CO at high altitude, one of the
primary considerations is to prevent excessive enrichment of
the fuel/air ratio.
A. Basic Types of Exhaust Emission Control Systems
The degree to which reduced air density at high altitude
affects emissions depends on the type of emission control
system already on the vehicle. By 1S84, nearly all light-duty
gasoline-fueled vehicles will be equipped with three-way
catalysts to reduce HC, CO, and NOx emissions to the current
statutory levels. The term three-way comes from the fact that
all three of the regulated pollutants are controlled by this
catalyst.
While it is fairly easy for the catalyst to oxidize the HC
and CO in the exhaust to carbon dioxide and water, it is more
difficult to remove the NOx in the exhaust. Effective NOx
control depends on keeping the level of oxygen in the exhaust
to a fairly low level. Otherwise, any excess level of oxygen
will react with the HC and CO, preventing the CO from reducing
the NOx to elemental nitrogen.
One basic method for controlling the level of oxygen in
the exhaust is called "feedback" (or closed-loop) control,
where the oxygen level in the exhaust is measured by an oxygen
sensor. The electrical signal produced by this sensor is sent
to a minicomputer (microprocessor) which makes the appropriate
adjustment in the fuel/air mixture setting. While requiring
fairly sophisticated electronics, this type of system is fairly
easy to set up to work efficiently. Also, changes in engine
operating conditions due to temperature, engine wear, and, to
some extent, altitude are automatically compensated for since
the exhaust oxygen level is measured directly.
Due to the predominance of feedback control systems and
their unique way of controlling the engine's fuel metering
device, all other types of control systems can be grouped
together in a single category termed "nonfeedback" (or
open-loop) control systems. These systems are used on vehicles
equipped with three-way catalysts and on smaller vehicles using
only oxidation catalysts to control only HC and CO.
Unlike feedback systems, nonfeedback systems have no
inherent ability to control the fuel/air mixture automatically
(and oxygen concentration for three-way catalysts) in response
to changing engine-related parameters. Instead, "fixed"
settings meter the fuel into the engine.
-------
III-4
The effect of reduced air density (i.e., high altitude) on
the effectiveness of these two types of exhaust emission
control systems will now be examined. The nonfeedback control
systems will be examined first and the feedback control systems
second. Brief discussions of the high-altitude control
techniques available for each of these two types of systems
will also be included.
B. Nonfeedback Control Systems
The effects of increasing altitude on exhaust emissions
from typical nonfeedback systems are shown in Table III-l. The
high-altitude data were obtained from emission tests conducted
near Denver, Colorado at 5,300 feet. The fact that nonfeedback
systems have no inherent ability to compensate for the
increasingly rich fuel/air mixture at higher elevations is
reflected by the significantly greater levels of HC and CO
emissions at high altitude compared with their low-altitude
performance. The increase for HC emissions is 31 to 625
percent, and for CO emissions, 195 to 924 percent. Emissions
of NOx generally decrease by 4 to 75 percent, although for one
vehicle they actually increased.
There are several conventional methods to compensate for
the increasingly rich fuel/air ratio at high altitude. Some of
these methods were used to comply with the initial
high-altitude emission standards, which were only in effect for
the 1977 model year. Manufacturers might again use such
techniques, in varying degrees of sophistication, to comply
with the requirements of the various control scenarios being
analyzed here. In general, for the 1977 program, manufacturers
used different carburetors on their high-altitude vehicles that
were designed for the air densities found at higher
elevations.
These carburetor changes were generally accomplished by
either a different jet size to reduce the amount of fuel
metered, or by providing an air bleed separate from the main
fuel metering system. The air bleed allowed fresh air to enter
the intake manifold without introducing additional fuel. It
could be introduced manually at high altitude, for example, by
opening a separate air bypass with a screwdriver. However, the
same air bleed (or bypass) could be automatically openea or
closed by using an aneroid control device. An aneroid is a
pressure-sensing device usually consisting of a diaphragm that
expands with decreasing pressure and contracts with increasing
pressure. The diaphragm can be constructed so that when it
expands at high altitude it forces the air bleed to open,
allowing additional air to flow into the engine. At low-
altitudes, the diaphragm keeps the air bleed closed.
Aneroids also can be employed to control the operation of
essentially any other parameter on the vehicle. For example,
-------
III-5
Table III-l
Low- and Hiyh-Altitude Emissions From various
Nonfeedback Vehicles for Past and Present Model Years
Low
Altitude
Hiyh
, Altitude
(g/mi)
(g/mi)
Manufacturer
Car
HC
CO
NOX
HC
CO
NOX
Comment
Nissan[2]
510
0.34
4.1
U.51
0.94
31.5
0.13
[aj
210
0.34
2.4
0.98
0.51
12.9
0.94
[a]
Ford [2]
4.2L
0.39
1.9
0.87
0.51
5.6
0.57
La,bJ
Subaru[3]
97 CID
0.39
4.2
3.09
1.54
8.0
2.63
i a, c]
Volkswayen[3]
97 CID
0.16
2.9
0.35
1.16
29.7
1.14
[ c,d]
[a] Carburetor system
i.
[b] Values include assigned deterioration factors of 1.3 for HCf
1.2 for CO, and 1.1 for NOx (Reference 4).
[c] In-use vehicle tests.
[d] Fuel injection system.
-------
III-6
aneroids can be adapted to adjust continuously (or in steps)
the operation of the exhaust gas recirculation (EGR) system,
the spark advance, the transmission shift points, the flow from
the air injection system into the exhaust, the deceleration
valve calibration, and many other engine parameters as well.
c. Feedback Control Systems
As discussed earlier, the feedback control system uses an
oxygen sensor to measure the concentration of oxygen in the
engine exhaust which then sends an electrical signal to an
electronic control unit indicating whether the system is
operating too rich (too much fuel) or too lean (not enough
fuel). The control unit adjusts the amount of fuel being
metered accordingly. Thus, the sensor will automatically
compensate for the natural enrichment of the fuel/air mixture
when a vehicle is driven at high altitude.
However, this system has two basic limitations in its
ability to maintain good air/fuel ratio control at all
altitudes. The first occurs during certain portions of vehicle
operation in which the feedback system operates in what is
termed "open-loop." This means that the feedback
characteristic of the system is not functioning, and the oxygen
sensor is not controlling the amount of fuel being metered into
the system. Open-loop operation commonly occurs during two
types of vehicle operation. The first type occurs when a
vehicle is cold started and requires a richer than normal
fuel/air ratio in order to operate. This continues until the
engine is warmed-up. The second type is wide-open throttle
(WOT) operation (or nearly wide-open throttle operation), where
a very rich fuel/air ratio is required to increase engine
power. This technique is called power enrichment. In both
cases, the enrichment at high altitude will be greater than at
low altitude, and CO and HC emissions will increase.
Therefore, additional altitude compensation must be provided
for these particular modes of operation of the vehicle to
assure maximum emission control.
Two basic types of electronic fuel metering devices that
may be used in the future are carburetor systems or fuel
injection systems. Beth of these feedback systems may function
much like the system on a nonfeedback controlled vehicle during
open-loop operation where the fuel metering setting is fixed.
Under these conditions the fuel/air mixture is significantly
richer at high altitude than at low altitude. Unless some
altitude compensation is added during these open-loop periods,
HC and CO emissions will increase substantially.
Some electronic fuel injection systems (e.g., General
Motors' throttle body injection (TBI)) are currently more
sophisticated than their carbureted counterparts during
open-loop operation. Such fuel injection systems may continue
-------
III-7
to monitor a variety of engine sensors (other than the oxygen
sensor) to maintain the correct fuel/air mixture entering the
engine.[5] More specifically, the electronic microprocessor of
these systems in effect "senses" the atmospheric pressure to
determine the engine's fuel requirement. This automatically
compensates for the effects of high altitude, since the fuel
metering system will account for the lower atmospheric
pressures at higher elevations.
It is possible, however, that even TBI systems may not be
fully compensating during open-loop operation. For example,
the microprocessor may be incapable of fully using the range of
sensor outputs that would accompany not only the normal engine
operating regime, but also the added pressure variation because
of changes in elevation. Also, the pressure sensors themselves
may lack an adequate response range to account for the pressure
variations. Finally, it is possible that all future TBI
systems may not include such sophisticated fuel management
because of cost considerations. Any of these circumstances
would make these systems behave more like the less
sophisticated carbureted feedback systems during open-loop
operations. The fuel/air mixture would likely become richer
with increasing altitude. As a result, hydrocarbons and CO
will increase during these open-loop periods without some type
of additional compensation.
The second limitation of these feedback control systems to
compensate for changes in altitude pertains predominately to
closed-loop operation and is related to the basic design of the
fuel-metering system itself (i.e., the carburetor or fuel
injection). Feedback carburetors typically incorporate two
circuits to meter fuel. One circuit, the lean authority limit,
meters fuel at a set air/fuel ratio through a fixed orifice,
while the other circuit, the controllable portion, meters an
increased level of fuel. The amount of additional fuel metered
through this controllable circuit is varied to maintain the
air/fuel ratio at the proper level as dictated by the oxygen
sensor. As the density of air decreases at higher elevation,
the controllable portion of the fuel flow is cut back to
maintain the proper air/fuel ratio.
This compensation is limited by the absolute amount of
fuel metered through the lean authority limit, which is fixed.
Once this limit is reached, the ability to compensate for
altitude disappears, and HC and CO emissions will increase with
further increases in altitude. The degree to which the lean
authority remains rich at high altitude will determine the
degree to which emissions increase. Therefore, to compensate
for higher elevations, the lean authority limit must not be
reached before the maximum altitude of control.
For some fuel injection systems, an analogous situation
exists. The fuel injectors used in these systems also have
-------
III-8
mechanical limitations regarding the minimum amount of fuel
that can be metered into the engine. [5] As previously
discussed for carburetors, this could affect the ability of the
system to adequately lean the fuel/air mixture and control HC
and CO emissions at high altitude.
The effects of increasing elevation on the emissions from
vehicles equipped with various feedback control systems are
shown in Table III-2. The inherent ability of these systems to
compensate for the effects of altitude at least partially is
readily apparent when the data in Table III-2 are compared to
the data listed in Table III-l for nonfeedback systems. The
low- and high-altitude results for feedback systems show
changes of about -11 to 110 percent for HC, -5 to 376" percent
for CO, and -8 to 31 percent for NOx. These low-to-high
altitude differences are significantly less than those
previously cited for nonfeedback systems. In addition, the
throttle body fuel injection systems shown in this table
clearly demonstrate the ability to control emissions at higher
elevations, although some increase still occurs.
It is unclear, however, if these systems, which are
currently produced in limited numbers, characterize those which
will be used on many future vehicles. If future throttle body
systems are less sophisticated than the few for which data are
available (Table III-2), the emissions increase with altituae
may be significantly greater and could even be as poor as that
for the Nissan electronic fuel injected vehicle.
The absolute emission levels shown in Table III-2 at high
altitude are also important to note. Many feedback control
systems already meet the proportional standards of 0.57 g/mi
KC, 7.8 g/mi CO, and 1.0 g/mi NOx. Some systems also meet the
statutory standards of 0.41 g/mi HC, 3.4 g/mi CO, and 1.0 g/mi
NOx at high altitude.
As previously mentioned, the greater emissions from
feedback systems at high altitude result from inadequate fuel
metering compensation when the fuel-metering system is
operating in the "closed-loop" mode, or from an increasingly
rich fuel/air mixture when the system is operating in
"open-loop" modes (e.g., wide-open throttle and cold-start
operation) . For the feedback systems that will be used in the
1984 and later model years, the rich fuel/air mixtures during
"open-loop" operation will be the biggest roadblock to
compliance with the various scenarios. As with nonfeedback
systems, there are various ways to reduce emissions from
feedback systems. These procedures and the requisite hardware
for each scenario will be outlined in the next section.
Generally, however, the open-loop fuel settings will have to be
recalibrated either by changing the electronic module of the
microprocessor of high-altitude vehicles, by expanding the
capability of the existing electronics to adjust the open-loop
-------
III-9
Table III-2
Summary of Unmodified Feedback systems
for the 1981 Model Year
Manufacturer
Nissan[2]
Ford[ 6 ]
GM[ 6 ]
Low Altitude
Hiyh Altitude
(y/mi)
(y/rai)
Car
1IC CO - NOX
HC
CO
NOX
Comment
280ZX
0.31 2.1 0.47
0.65
10.0
0.46
[a]
2.3 L
0.53
5.7
U. 8
[b]
5.8 L
0.30
2.3
1.6
[b]
0.24
2.4
1.6
[b]
5.0 L
0.35
3.6
0.5
[bl
0.37
4.4
0.5
Lb]
2.5 L
0.57
6.6
0.8
[b]
0 .60
7.5
0.9
[b]
3.8 L
0.41
4.4
0.8
[b]
0.46
4.6
0.8
[b]
4.3 L
0.52
4.7
0.7
[b]
0.53
4.5
0.8
[b]
4.9 L
0.35
3.5
0.8
[b]
0.46
4.2
0.7
t b J
4.4 L
0.52
9.4
0.6
[b]
0.55
8.8
0.6
[b]
5.5 L
0.29
2.7
0.7
[b]
0.41)
3.8
0.8
[b]
5.7 L
0.24
1.9
0.5
£o]
0.34
2.8
0.6
[b]
Chrysler[2]
GM[7]
Volvo[0]
1.7 L 0.17
2.2 L 0.13
225CID 0.29
2.5 L 0.36
2.5 L 0 .36
2.5 L 0.17
4.9 L 0.48
4.9 L 0.34
0.16
1.62 0.88
0.78 1.38
1.94 0.85
0.44 0.74
2.00 0.72
1.58 0.52
2.14 0.85
2.38 0.60
2.15 0.39
0.26 4.10
0.73 6.65
0.74 4.34
0.32 0.80
0.42 3.19
0.2b 2.15
0.49 4.78
0.38 2.26
0.31 2.86
0.88 [b,c]
1.58 [b,c]
0.90 [b,c]
0.97 [c,d]
0.92 [c,dJ
0.48 [c,dj
U.80 [c,d]
0.77 [c,d]
0.32 [dj
[a] Electronic fuel injection system.
[b] Carburetor system.
[c] Values include assigned deterioration factors of 1.3 for HC/
1.2 for CO, and 1.1 for NOx (Reference 4).
[d] Throttle body injection system.
-------
111-10
fuel/air mixture, or by adding a pressure sensing device which
will automatically allow the microprocessor to adjust the
open-loop calibration for changes in altitude.
III. EFFECTS OF INCREASING ALTITUDE ON EVAPORATIVE EMISSIONS
Evaporative emissions consist of hydrocarbon (hC) vapors
that escape (evaporate) primarily from the fuel tank and the
carburetor bowl of a gasoline-fueled vehicle. They are,
therefore, relatively independent of the type of exhaust
emission control system (feedback or nonfeedback) used on the
vehicle. Evaporative emissions are affected by the
distillation temperature curve of the fuel. At higher
elevations where atmospheric pressure is less, the curve is
lowered so that the fuel becomes relatively more volatile and
evaporation increases.
Evaporative emissions are currently controlled by routing
the fuel vapors from the carburetor and fuel tank into the
intake manifold of the engine when the vehicle is in
operation. In this way, the vapors are burned in the
combustion chamber along with the main fuel/air mixture. When
the vehicle is not operating, the fuel vapors are routed to a
charcoal-filled canister where they are stored until the engine
is restarted. At that time, the vapors are transferred to the
intake manifold and are burned in the combustion chamber.
Unless the capacity and purge rate of the canister are properly
designed, excess evaporative emissions may result at high
altitude.
IV. TECHNOLOGIES NECESSARY FOR COMPLYING WITH ThE SCENARIOS
The emission control devices and techniques necessary to
comply with the various scenarios will be outlined in this
section. The first step in this task will be to explain the
methodology used to determine the techniques required for
compliance. The second step, will be to actually estimate the
requisite emission control hardware which may be necessary to
comply with the control scenarios that provide exemptions for
low-power-to-weight vehicles (the base scenario and scenarios
lb, 2, and 3). The third step will be to estimate the
requisite hardware which may be necessary for vehicles to meet
the control scenarios not providing such exemptions (scenarios
la, lc, and 3b).
A. Methodology
The feasibility of meeting high-altitude standards
involves several significant issues. One of the most important
is that whatever the solution, it must be acceptable in terms
of its social, environmental and economic impacts. Some ways
of achieving the high-altitude standards may be environmentally
and technically sound, but unacceptable because of their social
-------
III-ll
or economic impacts. Therefore, some judgment has been applied
at an early stage in an effort to restrict the alternative
technologies to only those with the potential to be
environmentally, technically, socially, and economically
acceptable.
The technical analysis was limited to the use of
conventional power-plant systems. No exotic technology was
considered, since such applications would be unavailable in the
immediate future. Some conventional means of meeting
high-altitude standards also were excluded because of their
potential for significant adverse energy and economic impacts
on the national automotive fleet. While no guarantees can be
made that some vehicles sold at low altitude will not be
adversely affected by the high-altitude requirements, there are
ways to mitigate this prospect, and the control hardware for
each scenario has been chosen with this in mind. For instance,
with respect to the standards that are continuous with
altitude, any significant increase in the noble metal loading
of catalysts beyond that required to meet the standards at low
altitude was rejected. Among other things, the high cost of
such metals (platinum, palladium, and rhodium) would easily
preclude high-altitude standards from being cost effective.
(This is not to imply that the high-altitude requirements could
be met in every case simply by increased catalyst loading.)
Other more conventional means of complying with
high-altitude standards were also rejected if they would need
to be implemented on all vehicles (continuous strategies).
Increases in engine size or drive axle ratio that could be used
to effectively increase the power-to-weight ratio of low-power
vehicles are among these techniques. Although these techniques
would reduce the time spent at or near wide-open throttle,
thereby reducing the overly-rich mixtures associated with power
enrichment operation, they could also increase fuel consumption
throughout the nation by eliminating the more fuel economic,
smaller displacement engines or lower numerical axle ratios at
low altitudes. Of course, manufacturers might be able to
recover that lost fuel efficiency through such other, unrelated
means as weight reduction, but such programs would be very
costly and their cost would have to be considered a consequence
of high-altitude control. By assuming that these options would
be unacceptable, this analysis focused on only the more likely
and, therefore, reasonable emission control technologies.
The preferred approach in conducting a technological
assessment of any emission requirement is to use actual test
data. However, since Congress enacted the Clean Air Act
Amendments of 1977, the automotive industry has reported very
little development data that demonstrate the capability of
current or future emission control systems to abate gaseous
pollutants significantly at higher elevations. This limited
amount of testing information prevents identifying the
-------
111-12
requisite exhaust emission control hardware for each scenario
based only on empirical data. In addition, no development data
have been submitted regarding the statutory evaporative
emission standards. When such data are lacking, the types of
systems required must be chosen based on engineering judgment.
In addition, all of the information that is available was
obtained from tests conducted at Denver, Colorado's altitude of
approximately 5,300 feet. While tests at this altitude are
useful in characterizing emission performance at 6,000 feet
because of the relatively small difference in elevation, the
data cannot be extrapolated to characterize emission
performance at 10,200 feet. In these cases, technical judgment
must be used to estimate emission control requirements. Also,
no emission tests have been conducted to demonstrate a control
technology's ability to compensate continuously for changes in
altitude and still meet the appropriate standards. This should
not be a serious fault, however, since the data collected at
5,300 feet can easily be extrapolated to characterize the
continuously applicable standards. The same parameters that
must be recalibrated to comply with a fixed-point standard can
be made to vary continuously with altitude b^ changing the
electronics, adding an aneroid, and using a servo motor.
Rather than attempt to specify the requisite control
hardware for each individual manufacturer, a task that would be
impossible given the limitations of the data base, vehicle
types will be grouped according to the design of their control
system and general control techniques will be selected which
have a high probability of achieving the desired emission
levels. Of course, implicit in this approach is that some
individual systems will cost more or less than the generic
system used in the analysis. Also, the hardware estimates are
based on EPA's projections of the 1984 and later fleet mix of
fuel-metering devices.[8] These estimates are, in turn, based
on statements from various manufacturers regarding their future
product plans. Because of the current state of the automotive
industry, however, considerable uncertainties exist with these
projections.
B. Technology Required for the Scenarios with Exemptions
The scenarios providing low-performance exemptions are lb,
2, 3a, and the base scenario. The existing exhaust emission
control systems fall into three basic categories: 1) those
that will not require modification to meet high-altitude
requirements, 2) feedback systems that will require
modification, and 3) nonfeedback systems that will require
modification. Each of these three basic types will be
discussed separately in relation to their ability to meet the
increasingly stringent proportional and statutory exhaust
emission standards. This discussion will be followed by a
brief analysis of the control hardware required to meet the
-------
I11-13
proportional and statutory evaporative emission standards.
Evaporative systems can be estimated independently from the
required exhaust emission systems.
1. Systems Requiring No Modification
As previously stated, some emission control systems that
were originally designed for compliance with low-altitude
standards also have the ability to meet the proportional and
statutory high-altitude standards. For example, with respect
to current (1981) nonfeedback systems, the Ford vehicle in
Table III-3 met the proportional standards with no known
modifications. However, since nonfeedback systems do not have
the inherent ability to compensate for changes in altituae, the
resulting excessively rich mixture at high altitude should
prevent all but a very few vehicles from meeting the
high-altitude standards. Thus, the Ford vehicle's compliance
is considered the exception rather than the rule. The opposite
is true for unmodified feedback vehicles. Table III-2 shows
that many of these systems met the proportional standards and
that a few of the GM and Volvo systems even complied with the
statutory standards at high altitude.
Feedback systems have the inherent ability to compensate
for changes in altitude by automatically adjusting the fuel/air
mixture. In the 1982 and 1983 high-altitude standards
rulemaking action,[1] EPA determined that about 78 percent of
the feedback systems should be capable of meeting the
proportional standards without modification, since they already
appear to possess an adequate range of adjustability. No
better estimate regarding the number of future feedback systems
which may comply with proportional standards without
modification is available. Therefore, this percentage is
assumed in this study to characterize both fuel injected (i.e.,
throttle body injection) feedback systems and carbureted
feedback systems. These systems are expected to account for 59
percent and 10 percent, respectively, of the total motor
vehicle fleet. As shown in Table III-4, the resulting market
shares are usea in the base scenario and also in scenario 3a
since these feedback systems compensate continuously for
altitude as well as at a fixed point of 5,300 feet.
Even though a few of the vehicles in Table III-2 with
unmodified feedback systems were able to meet the statutory
standards at high altitude, EPA has estimated that no
unmodified feedback systems will be certified to these emission
levels in scenarios lb and 2. The rationale for this estimate
is included in the following discussion of modified feedback
systems.
-------
111-14
Table in-3
Summary of Unmodified Nonfeedback Systems
for the 1981 Model Year
LOW
Altitude
(y/mi)
High
Altitude
(g/mi)
Manufacturer
Car
HC
CO
NOX
HC
CO
NOX
Comment
Nissan[2]
510
210
0.34
0.34
4.1
2.4
0.51
0.98
0.94
0.51
31.5
12.9
0.13
0.94
[a]
[a]
Ford[2]
4.2L
0.39
1.9
0.87
0.51
5.6
0.57
[ a,b]
[a] Carburetor system.
[b] Values include assigned deterioration factors of 1.3 for HC,
1.2 for CO, and 1.1 for NOx (Reference 4).
-------
Table III-4
Estimated Exhaust Emission Control Requirements
for Scenarios with Exemptions
Control
Hardware
Nonfeedback Vehicles
Fixed-Step Aneroid:
A. Carburetor
B. Power Enrichment
C. EGR or air
injection rate
Continuous Aneroid:
A. Carburetor
B. Spark
Feedback Vehicles[b]
Feedback Control w/Map
(Three-Way Ford and
Foreign Market Share)
Expand Function of
Existing Electronic
A. Expand Capability
B. Add MAP Sensor
Scenario lb:
Continuous
Statutory,
6,000 Feet
W/Exemptions
N/ A [ a ]
N/A
Scenario 2:
Fixed-Point
Statutory,
5,200 Feet
W/Exemptions
31% A,B,C
N/A
31%
59% (TBI) A
10% (FBC) A,B
N/A
59% (TBI)
N/A
Scenario 3a:
Continuous
Proportional,
6,000 Feet
W/Exemptions
N/A
31% A, B
N/A
13%
2%
Base Scenario:
Fixed-Point
Proportional,
5,200 Feet
W/Exemptions
31% A
N/A
(TBI) A
(FBC) A
N/A
J.3% (TBI) A
-------
Table III-4 (cont'd)
Estimated Exhaust Emission Control Requirements
for Scenarios with Exemptions
Control
Hardware
Feedback Vehicles
Chanye Electronic
Modules (FBC Only)
A. Recalibrate fuel
metering
B. Recalibrate fuel
metering, spark,
and EGR plus add
MAP for fuel
No Change FBC
No Change TBI
Scenario lb:
Continuous
Statutory,
6,000 Feet
W/Exemptions
N/A
N/A
N/A
Scenario 2:
Fixed-Point
Statutory,
5,200 Feet
W/Exemptions
10% (FBC) B
N/A
N/A
Scenario 3a:
Continuous
Proportional,
6,000 Feet
W/Exemptions
N/A
8%
46%
Base scenario:
Fixeu-Point
Proportional,
5,200 Feet
W/Exemptions
2% (FBC) A
8%
46%
Type of Strategy
One-Car
Two-Car
One-Car
Two-Car
[a] Not Applicable.
[bj FBC means nonfeedback carburetor systems.
TBI means throttle body injection feedback systems.
-------
111-17
2. Feedback Systems Requiring Modification
Some feedback systems will require modification to meet
either the proportional or statutory standards. Table III-5
and Figure III-l present data on systems with varing degrees of
modification. The emission levels in Table III-5 are
representative of recalibrating (leaning) the fuel/air mixture
during the cold-start portion of the test cycle when the
systems are operating in the open-loop mode (i.e., the oxygen
sensor is not functioning to control the fuel/air ratio
entering the combustion chamber).
Figure 111-1 presents the available data on vehicles that
have had more significant modifications to their feedback
emission control systems than the vehicles listed in Table
III-5. These tests were conducted in an attempt to achieve sea
level emission performance at high altitude. Before discussing
the results shown in Figure III-l further, an explanation of
the GM test program which generated the results is necessary.
The test results were obtained using three-way catalyst
vehicles and, therefore, are not representative of the
three-way-plus-oxidation catalyst vehicles that were certified
by GM for the 1981 model year. For this reason, it is not
surprising that test vehicles in Figure III-l failed to meet
even the statutory low-altitude standards. Also, modifications
were apparently made to the vehicles after the low-altitude
tests had been conducted. This prevents a direct comparison
between emissions performance at sea level and high altitude.
The high-altitude modifications included adding EGR at
wide-open throttle for all vehicles, disconnection of the power
enrichment circuit for all vehicles and recalibration of the
open-loop, cold-start operating mode for some vehicles.[7]
Generally, comparing the low- and high-altitude results
presented in Table III-5 shows that the vehicles with only
recalibrated cold start fuel/air mixtures experience rather
small emission increases of up to 0.2 g/mi HC, 3 g/mi CO, and
0.4 g/mi NOx. These increases should be able to be reduced or
eliminated by compensating other engine operating parameters
for the effects of altitude (e.g., fuel/air ratio wide-open
throttle (WOT), spark timing, and exhaust gas recirculation
(EGR)).
The vehicles depicted in Figure V-l show mixed results,
but some vehicles are definitely below the corresponding
low-altitude levels. This shows that such extremes as
disconnecting the power enrichment circuit can be very
effective. Both Table III-5 and Figure III-l indicate that
controlling HC and CO emissions at high altitude may increase
NOx to unacceptable levels, even when EGR is used at WOT.
Looking specifically at only the high-altitude data, the
effectiveness of controlling emissions to the proportional
-------
111-18
Table III-5
Modified Feedback Systems
Low
Altitude
(g/mi)
High Altitude
(g/mi)
Manufacturer
Car
UC
CO
NOX
HC
CO
NOx
Comments
J RT[6]
122 CID
0 .40
0.55
7 .34
7.34
0.45
0.45
[a,b]
[a,b]
Chrysler[2]
1.7 L
2.2 L
225 CID
0.17
0.13
0.29
1.62
0.78
1.94
0.88
1.38
0.85
0.21
0.25
0.52
2.65
3.76
2.74
1.25
1.41
0.98
L a,b,c]
[ a#b,c]
La,b,c]
Toyotat 6]
168 CID
0.47
6.94
0.31
[a,b]
Nissan
2.0 L[ 5 ]
280ZX[2]
0.31
2.1
0.47
0.42
0.52
4.3
5.4
0.5
0.35
[ a,b]
[b,d]
TaT Carburetor system.
[bj Recalibrated open-loop mode.
[cj Values include assigned deterioration rates of 1.3 for HC,
1.2 for CO, and 1.1 for NOx (reference 4).
[d] Electronic fuel injection system.
-------
Ill—19
FIGURE III-1
Emissions Performance
Of Modified C-4 Systems
LXJ
OC
LU
a.
CO
S
<
QC
a
m
>
z
o
8
UJ
a.
u.
HC
CO
NOx
0.8
0.6
0.4
0.24
0
12
10<
PROPORTIONAL CONTROL
SEA - LEVEL STANDARD
PROPORTIONAL CONTROL
SEA-LEVEL STANDARD
1.5
1.0
0.5
ELEVATION, THOUSANDS OF FEET
NOTE: ALL TESTS AT LOW MILEAGE.
ADJUSTED TO 50,000-MILE LEVELS
USING DETERIORATION FACTORS FROM
1981 C-4 CERTIFICATION DATA.
-------
111-20
standards with a relatively easy recalibration of the
cola-start fuel/air mixture is apparent (Table III-5). The HC
and CO emissions are below the proportional standards for every
vehicle. NOx is exceeded by two vehicles but one of these also
violated the standard at low altitude. The other had such low
HC and CO levels that these emissions could be allowed to
increase by increasing the fuel/air ratio somewhat so that NOx
would be reduced. (Richer mixtures increase HC and CO, but
reduce NOx.) Therefore, a recalibration of the cold start,
fuel/air mixture should bring even these problem vehicles into
compliance with the proportional standards.
An estimate can now be made regarding the vehicles that
will require modification to their feedback systems in order to
meet the proportional standards in the base scenario and
scenario 3a. In the interim (1982-83) high-altitude
rulemaking[9] (the base scenario), EPA estimated that all
manufacturers using carbureted feedback systems that could not
comply with the regulations in their unmodified configuration
would use differently calibrated electronic modules on the
vehicles sold above 4,000 feet than on vehicles sold below that
elevation. As previously shown in Table I1I-4, carbureted
feedback systems are expected to be used on about 10 percent of
the total fleet ano that 78 percent of these feedback vehicles
will not require changes to their systems. This leaves 2
percent that will require a module change in the base scenario.
The above emission control technology estimates from the
interim rulemaking pertained predominately to carbureted
systems which are expected to dominate the motor vehicle fleet
during the early lS80's. In later model years, TBI systems are
expected to dominate with 59 percent of the fleet utilizing
such systems. In this analysis it is assumed that, although
TBI may compensate fuel/air mixtures during open-loop
operations, some of these vehicles may not be adequately
designed to account for the effects of high altitude in the
absence of high-altitude standards. Since TBI systems are
still being refined, there is a serious dearth of information
regarding the possible modifications which may be needed to
ensure compliance in every case. To estimate the potential
modifications, an analogy using carbureted systems is useful.
Previously it was stated that carbureted systems would be
modified by providing new open-loop calibrations on only
high-altitude vehicles. Since TBI systems are already assumed
to recalibrate open-loop calibrations to some degree, the
modifications to these systems should be less rigorous. It may
only be necessary to ensure that the system's microprocessor
(or perhaps sensors) has sufficient capability to account for
the changes in various engine operating parameters as a result
of higher elevations. For TBI systems it would be far less
expensive to build this capability into every vehicle sold in
the nation rather than change electronic hardware only on
-------
111-21
high-altitude vehicles. (The cost implications of such a
change are discussed further in Chapter V.) Basically, the
carbureted systems requiring modification have no existing
capability to recalibrate their open-loop fuel/air mixtures
automatically and would require additional hardware to be able
to do so. Such automatic systems would be expensive to add to
every vehicle when the modules on only high-altitude vehicles
would be changed instead. TBI systems are assumed to already
possess the inherent ability to compensate, hence, changes to
the existing electronics would be much easier than for
carbureted systems.
As previously described, TBI feedback systems are expected
to comprise about 59 percent of the total fleet and 78 percent
of these vehicles will not require modifications to comply with
proportional standards. This leaves 13 percent that will
require expansion of existing electronics in the base
scenario. The requisite control hardware for this scenario is
summarized in Table III-4.
Scenario 3a has the same standards at high altitude as the
base scenario, but, in addition, a vehicle must automatically
meet a proportional standard at all altitudes up to 6,000
feet. This one-car strategy will prevent the affected vehicles
from being modified for high altitude use by requiring that all
such vehicles incorporate appropriate modifications regardless
of where they are sold. The continuous requirements of this
scenario present no problems for either the feedback systems
which already comply with the proportional standards in the
base scenario or the TBI systems that would be modified in the
base scenario since these systems compensate automatically with
changes in elevation. This scenario does affect the carbureted
systems that would have had new electronic control modules
installed on only high-altitude vehicles in the base scenario,
however.
The necessary changes for this scenario can readily be
accomplished by expanding the memory capacity of the electronic
control unit in either of two ways. First, control of the
fuel/air mixtures can be enhanced by enlarging the
microprocessor's memory to include a greater number of engine
operating values (i.e., combinations of engine load and
speed). Increasing the number of values with which to set the
carburetor will result in more precise control of fuel metering
and, ultimately, better emissions performances.[10]
Second, a "continuously-powered memory" may be used to
compensate the fuel-metering system during open-loop
operations. This is accomplished by utilizing the computer's
existing capability to memorize the carburetor settings which
are required to maintain the proper fuel/air ratio during
"closed-loop" operation before the ignition is switched off.
When the engine is restarted, the last closed-loop operating
-------
111-22
conditions can be used to modify the open-loop carburetor
settings. Open-loop compensation for the effects of altitude
is, therefore, automatic since as previously stated, feedback
systems inherently compensate for variations in the fuel/air
ratio during closed-loop operation. Such systems are readily-
available with current technology. In scenario 3a, 2 percent
of the fleet will require an expanded memory in the electronic
control unit rather than simply changing the module on
high-altitude vehicles (Table III-4).
The statutory levels at high altitude represent a more
stringent standard than at low altitude because of the effect
of decreasing air density at higher elevations on engine
performance. This fact is reflected in the data. Table III-2
shows that even systems with the best ability to control
emissions at higher elevations (i.e., TBI) did not meet
statutory standards at high altitude in every case when
compliance at low-altitude was also demonstrated. Furthermore,
Table III-5 indicates that the relatively simple readjustment
of the fuel/air mixture that is sufficient to meet the
proportional levels will not be enough to bring "problem"
vehicles into compliance with the statutory standards at high
altitude. No vehicle simultaneously met these more stringent
standards for all three pollutants. Furthermore, while HC and
CO are usually considered the "problem" emissions at high
altitude, the data in Table III-5 and Figure III-l suggest that
as these pollutants are controlled to lower levels, NOx may
increase to marginal or unacceptable levels. Therefore,
additional emission reductions beyond those required by the
proportional standards appear be relatively easy for most
vehicles, but as the statutory levels are approached many
engines will require more significant modifications to their
feedback systems, including some that are represented in Figure
III-l. The available techniques include further leaning of the
cold start fuel/air mixture, disconnecting the power enrichment
circuit of the carburetor, modifying the choke setting,
changing spark timing, changing the automatic transmission
shift points, modifying the air injection rates, recalibrating
EGR at part-throttle operation, and adding EGR at WOT.
Specifically, to meet the statutory standards at 5,300
feet in scenario 2, the vast majority of feedback systems would
require more significant changes than were needed to comply
with the proportional levels. Of course, Table III-2 shows
that some unmodified systems currently comply with the
statutory standards of 0.41 g/mi HC, 3.4 g/mi CO, and 1.0 g/mi
NOx. Conversely, some vehicles may require such extreme fixes
as disconnecting the power enrichment circuit and adoing EGR
at WOT. It is expected that such vehicles would be relatively
few in number because exemptions would be available.
Therefore, the generic emission control system would be
somewhere in between these two extremes and would affect 100
percent of the feedback systems or 69 percent of the total
fleet.
-------
II1-23
Since scenario 2 is a two-car strategy# the necessary
modifications need only be made to high-altitude vehicles where
it is practical by changing the electronic module as in the
base scenario. In scenario 2, however, the modifications would
be more extensive. The cold start fuel/air mixtures of
carbureted systems would be recalibrated as in the base
scenario but would be more refined to provide the precise fuel
metering that would be required. This recalibration would
probably include a revised choke setting to prevent excessively
rich mixtures when the engine is first started. The
high-altitude electronic control module would probably include
special calibrations for spark timing and EGR to provide
further emission reductions. Another major change would likely
include the addition of a manifold absolute pressure (MAP)
sensor to the high-altitude electronic module. This sensor
would be used to ensure acceptable driveability if the vehicle
is driven at sea level. During the open-loop portion of the
feedback carburetor system's operating regime, the fuel/air
ratio is not properly compensated for changes in altitude. As
a result, the mixture becomes leaner at lower elevations since
the air is more dense and has more oxygen per unit volume, but
the amount of fuel metered remains the same. A very lean
mixture will cause hard starting under certain conditions and
lean misfiring. The MAP sensor will prevent unacceptable
driving characteristics by signaling the microprocessor to
provide a richer fuel/air mixture at a predetermined altitude
below 4,000 feet. This device was not needed in the base
scenario, because the special high-altitude calibration is not
significant enough to cause unacceptable performance if the car
was used at sea level.
In scenario 2, TBI systems probably would have the
capability of their existing electronics expanded as previously
described for the base scenario. Such expansion may be more
significant than for proportional standards, as it would be for
carbureted systems. Throttle body systems would not require
the addition of a MAP sensor because this device is already
present.[5] The emission control technology for scenario 2 is
shown in Table III-4.
To meet the statutory standards at all altitudes up to
6,000 feet (scenario lb) would require essentially the same
emission control hardware as in scenario 2. However, the
one-car strategy of scenario lb would mean installing the MAP
sensor on all feedback-carbureted vehicles regardless of the
altitude at which they are sold. The memory of the electronic
control unit would have to be expanded beyond that required in
scenario 3a to include additional open-loop calibrations for
various altitudes. As before, TBI systems would not require
further modifications to meet the statutory requirements of
scenario lb beyond that needed for scenario 2. Table III-4
shows that in this scenario (lb) all feedback carbureted
systems (10 percent of the fleet) are expected to require
-------
111-24
expansion of their electronic capability and the addition of
MAP sensor, while all TBI systems (59 percent of the fleet) are
expected to require only the expansion of their existing
electronics.
In addition to modifying current feedback systems, meeting
the statutory standards at all altitudes is expected to be
difficult for nonfeedback systems even with significant
modifications. While it is possible to design continuously
compensating nonfeedback systems, as will be discussed in the
next section, controlling these systems to the precise levels
needed to attain the statutory standards may be of such
complexity as to preclude that approach in most instances.
This difficulty could precipitate a change from nonfeedback to
feedback carburetor systems which are capable of much more
precise control and, therefore, can meet the requirements.
This is not to say that nonfeedback systems cannot be made to
do an adequate job for some vehicles, but even in these
instances the amount of development effort may be so costly, in
terms of time and money, that feedback systems will probably be
used in the vast majority of cases. For scenario lb, then, EPA
estimates that the 31 percent market share held primarily by
Ford and the foreign manufacturers would be converted to
feedback carburetor systems using the same modifications as
previously described. (Table II1-4).
3. Nonfeedback Systems Requiring Modification
As previously stated, it is expected that all nonfeedback
systems will require some type of modification in order to
comply with the high-altitude standards. Table III-6 presents
data on modified nonfeedback systems. These data were compiled
in response to the rulemaking that implemented the 1982 and
1983 high altitude standards.[1] The goal of the test programs
which generated these values was to demonstrate compliance with
proportional standards, and was not to develop a system that
would achieve the statutory standards at high altitude. For
this reason, the vehicle modifications aepicted in the table
are relatively simple. The two Nissan vehicles utilized an
aneroid to lean the fuel/air mixture of the carburetor during
all operating modes other than wide-open throttle (WOT) (i.e.,
no compensation for power enrichment compensation). The
remaining vehicles had aneroids installed to control spark
timing and automatic transmission shift points.
The effectiveness of installing aneroids to contol certain
engine parameters is best shown by comparing the high-altitude
emission levels in Tables III-3 and III-6 for unmodified and
modified nonfeedback vehicles, respectively. The Nissan and
Ford vehicles in both tables are identical except for the
installation of aneroids. The aneroid compensated carburetor
on the Nissan vehicles in Table III-6 reduced HC up to 64
percent, CO up to 82 percent, and increased NOx from 70 to S00
-------
I11-25
Table III-6
Modified Nonfeedback Systems
Manufacturer
Car
Low Altitude
(q/mi)
HC
CO
NOX
High Altitude
(g/mi)
HC
CO
NOX
Comments
Nissan[2] 510 0.34 4.1 0.51
(Datsun) 210 0.30 2.4 0.98
Toyota[6] 73 CID
Chrysler[6] 2.6 L
Ford[2] 4.2 L 0.30 1.3 0.96
0.34
5.7
1.3
[ a,b]
0 .36
5.1
1.6
[a,b]
0.51
4.73
1.0
[a,c]
0.44
5.6
1.0
[ a, c]
0.37
6.8
1.0
[ a, c]
0.49
2.3
0.99
[a,c]
[a] Carburetor system.
[b] Aneroid on carburetor.
[c] Aneroids on transmission and ignition timing.
-------
111—2 6
percent. The ignition and transmission aneroids on the Ford
vehicle in Table III-6 reduced HC by 4 percent, CO by 59
percent and increased NOx by 74 percent. As expected/
compensating the carburetor (i.e., controlling the fuel/air
ratio) is the most effective way to control HC and CO emissions
at high altitude, although compensating the ignition and
transmission are also quite effective. The data also shows
that NOx emissions may be greater at high altitude for vehicles
with some forms of altitude compensation. Generally, however,
the information in Table 1II-6 shows the above modifications
can be effective in reducing high-altitude emissions.
It is clear that the HC and CO proportional standards can
readily be met with relatively simple recalibrations of key
engine operating parameters (Table III-6). The results also
show that leaning the fuel/air mixture to reduce HC and CO
emissions may increase NOx emissions beyond allowable levels
(1.0 g/mi). However, the NOx emissions in Table III-6 should
not be a significant problem with regard to the proportional
standards, because the values were achieved during early
development testing. The Nissan vehicles which exceed the NOx
standard are sufficiently below the HC and CO requirements of
0.57 g/mi HC and 7.8 g/mi CO so that enriching the fuel/air
mixture should reduce NOx to acceptable levels while still
meeting the KC and CO standards. Also, other NOx
counter-measures may be employed, such as recalibrating the EGR
system.
The specific emission control requirements for the base
scenario are shown in Table III-4. As explained earlier, the
estimates of the generic systems for the base scenario are
taken from the interim high-altitude rulemaking. In that
action, EPA estimated that all nonfeedback vehicles, 31 percent
of the future fleet sold above 4,000 feet, would be equipped
with aneroid-controlled carburetors even though special
high-altitude hardware with fixed calibrations could be used at
a lower price. Aneroids are the preferred solution for
compensating fuel/air mixtures so that if a vehicle is driven
at sea level, the otherwise accompanying lean mixture would be
avoided. Therefore, using an aneroid to slightly richen the
mixture at a predetermined altitude below 4,000 feet will
ensure acceptable driveability at lower elevations. This
approach is followed in this study, as shown in Table III-4.
To meet the proportional standards at all altitudes up to
6,000 feet in scenario 3a, nonfeedback vehicles will have to
switch from aneroids that engage at a predetermined altitude to
aneroids that operate in a continuous manner. The perfection
of a continuous aneroid that will properly meter the precise
degree of fuel needed over a wide range altitudes (i.e., 1,800
to 6,000 feet is a more difficult task than developing a
fixed-step aneroid). While one fixed-step aneroid was
estimated for the base scenario, the complexity of adopting a
-------
111-27
single aneroid which must precisely meter an air bleed to lean
the fuel/air mixture may result in a less than perfect device.
Therefore, an additional aneroid may be needed to control a
second engine operating parameter, such as ignition timing, to
assure compliance with the standards at all elevations up to
6,000 feet. Thus, two aneroids of this type have been
estimated for the generic emission control system required
under scenario 3a.
The choice of two aneroids for scenario 3a is a
compromise. It is possible for one aneroid to operate more than
one engine parameter. However, since it is also reasonable to
expect that more than one aneroid may be required to meet the
emission requirements for some vehicles, it has been
conservatively estimated that two aneroids are needed in the
average case.
In order to comply with the statutory standards at high
altitude (scenarios 2 and lb), nonfeedback systems will require
additional compensation. Table III-6 shows that no vehicle
simultaneously met the statutory levels of 0.41 g/mi HC, 3.4
g/mi CO, and 1.0 g/mi NOx. To reduce HC and CO emissions
further, these vehicles may employ a variety of techniques.
The Nissan vehicles may require a greater degree of aneroid
control of the fuel-metering system. This could be
accomplished by further refinement of the existing aneroid
control mechanism or by adding additional aneroid control to
the power enrichment or accelerator pump circuits of the
carburetor. Also, aneroids could be added to control spark
timing or the shift points of automatic transmissions. The
addition of an air pump or a change from a pulse air pump to a
continuous air pump could also be used to reduce the CO
emissions to the required levels. The Toyota, Chrysler, and
Ford vehicles already have the equivalent of aneroid-controlled
spark timing and transmission shift points but the remaining
emission control techniques mentioned above could also be
used. This includes the adoption of the very effective
aneroid-controlled carburetor which the Nissan vehicles already
have.
Regarding NOx emissions, the Nissan vehicles already
exceed the allowable level and further leaning of the fuel/air
mixture entering the combustion chamber may increase the amount
of this pollutant to unacceptable levels for the Toyota,
Chrysler, and Ford vehicles also. Added EGR control should
help remedy this problem. Generally, this requirement is most
likely for low-power vehicles which spend a significant amount
of time during the Federal Test Procedure (FTP) at or near WOT
when tested under high-altitude conditions. Many conventional
EGR systems are inoperative under these circumstances in order
to maximize engine power. Alternatively, additional air
injection rates from an existing system might be sufficient for
some vehicles to reduce the HC and CO levels without
-------
111-28
significantly increasing NOx. These aneroids may be used to
control various engine parameters. The first aneroid can
provide the most effective control by compensating the fuel/air
ratio of the fuel-metering device. The second and third
aneroids would probably be most effective if used to regulate
spark timing and air injection rates. One of these aneroids
may be needed to control the rate of exhaust gas recirculation
(EGR) for reduced NOx emissions.
From this, EPA estimates that the typical emission control
system used under scenario 2 will include three fixed-step
aneroids (Table III-4). Again, some vehicles may require more
significant emission control countermeasures and some less, but
three aneroids should adequately characterize the average
vehicle. As an example, one of the more expensive options for
reducing emissions is the addition of an air injection system.
Air injection systems are used to oxidize HC and CO emissions
to water and carbon dioxide by introducing additional air into
the exhaust manifold after combustion has occurred. NOx
emissions are relatively unaffected since they are produced
during the high temperature and pressure of the combustion
process.
The addition of an air injection system, or a shift from a
less expensive pulse air system to a more costly air pump
system, was not specifically included in the generic emission
control system for two reasons. First, this course of action
is primarily restricted to the nonfeedback systems of smaller
displacement engines since the vast majority of other engines
currently employ air pump systems. Such smaller displacement,
nonfeedback engines are estimated to be less than about 10
percent of the LDV market. Also, many of these low-power
vehicles may qualify for exemption under this scenario.
Second, the significant increase in cost will limit its
application to those vehicles that simply cannot comply with
the standards by using aneroids or other less costly techniques.
The exhaust emission control requirements for scenario lb
already have been described in the section regarding modified
feedback systems but will be summarized here for convenience.
Meeting the statutory standards at all altitudes up to 6,000
feet may be possible by using continuous aneroids such as those
estimated for scenario 3a. However, the precise control of
several engine operating parameters which would be necessary to
attain the statutory standards is expected to be so difficult
as to preclude that approach in most instances. Therefore, EPA
estimates that the 31 percent market share v»hich is currently
held by nonfeedback systems will be converted to feedback
carbureted systems (Table III-4).
-------
111-29
4. Evaporative Emission Systems
Evaporative HC emissions are greater from uncontrolled
vehicles at higher elevations because of the reduced barometric
pressure at these locations. The increase in evaporative
emissions is proportional to the change in barometric pressure
and, therefore, altitude.
In the interim (1982-83) nigh-altitude rulemaking (the
base scenario in this analysis), EPA found that existing
evaporative emission control systems should have adequate
capacity to comply with the proportional standard of 2.6 g/test
HC at 5,300 feet.[l] Any system capable of meeting the
standard at this altitude should be able to comply with a
proportional standard at any elevation, since both the level of
the standard and evaporative emissions increase linearly with
the change in barometric pressure. Therefore, no additional
evaporative emission control hardware should be necessary for
scenario 3a and the base scenario (Table III-7).
The statutory evaporative emission standard (2.0 g/test)
is about 25 percent more stringent at 6,000 feet than the
proportional standard. EPA estimates that compliance with this
standard will generally require an increase of approximately 25
percent in the capacity of the carbon storage canister.
Additional capacity can be acquired in two ways. First, the
quantity of activated charcoal, upon which the fuel vapors are
adsorbed, can be increased. Second, a more efficient adsorber
can be used, usually a better grade of activated charcoal. A
properly designed system that complies with the statutory
standard at the highest elevation where control is required,
should comply with the standard at any altitude. Therefore,
the required evaporative emission control hardware for
scenarios lb and 2 are essentially the same (Table III-7). The
difference is, of course, that in scenario lb higher capacity
canisters would be required on all vehicles sold in the nation,
where scenario 2 would require these canisters on only vehicles
sold above 4,000 feet.
C. Required Technology for the Scenarios Without Exemptions
Scenarios la, lc, and 3a have no provision for exempting
certain vehicles from the regulatory requirements. Without
granting exemptions, these scenarios will certainly limit model
availability at both high and low altitudes to varying
degrees. Such limitations are of concern at both altitudes, of
course, but they become more onerous at low altitude where 95
to 97 percent of the sales occur. This is because many of the
vehicles which may require exemptions are low-power, high fuel
economy vehicles principally designed for low-altitude
operation. Many of these vehicles would not be sold at high
altitude, or would be sold in very small numbers, since the
reduced air density at higher elevations reduces their
-------
111-30
Table III-7
Estimated Evaporative Emission Control
Technology Requirements for All Scenarios
No Change to 25% Increase 50% Increase
Scenario Carbon Canister Carbon Canister Carbon Canister
la N/A[a] N/A 100%
lb N/A N/A N/A
lc N/A N/A N/A
2 N/A 100% N/A
3a 100% N/A N/A
3b 100% N/A N/A
Base 100% N/A N/A
[a] Not applicable.
-------
111-31
performance below acceptable levels. In the future, this
situation can be expected to become more common as engines
become even smaller to further improve fuel economy.
The lack of exemptions in scenarios la, lc, and 3b would
probably result in one of two possible courses by the
manufacturer. Either some low-power, high fuel economy
vehicles would be discontinued with an attendant increase in
national energy consumption, or more costly emission control
hardware would be added to the vehicles so that they may be
certified and sold. This analysis assumes that the obvious
disadvantages of dropping these vehicles from the national
sales offering would be avoided by using the more costly
emission control hardware. Nevertheless, at this time it is
impossible to state that simply adding more costly emission
control systems would, in every instance, allow these low-power
vehicles to meet the applicable standards, while retaining
acceptable performance.
1. Feedback and Nonfeedback Systems
The national vehicular fleet can be separated into two
categories for the purposes of estimating the emission control
hardware required in scenarios la, lc, and 3b: vehicles that
would not qualify for exemptions and vehicles that would
qualify for exemptions. The requisite hardware for the
vehicles that would not qualify for exemption is the same as
was estimated in the previous section for scenarios with the
same standards and control ceiling but differing in that they
allow exemptions. For these vehicles, scenario lc corresponds
to lb and scenario 3b corresponds to 3a. In scenario lc and 3b
the technology mix which has been assumed in the analysis up to
this point (10 percent feedback carbureted, 59 percent TBI, and
31 percent nonfeedback systems), would change somewhat if the
previously exempted vehicles were included in the fleet, since
such a change would be well within the uncertainty of the
analysis, the original fleet composition is assumed to remain
the same as previously described for consistency. Scenario la
has no corresponding scenario that allows for exemption.
Therefore, it is necessary to discuss the requisite control
technology for the nonexempt vehicles in this scenario before
describing the hardware that may be needed to bring exempt
vehicles into compliance.
In scenario la, all vehicles must meet statutory standards
without modifications up to 10,200 feet. As previously
described for scenario lb, the continuous statutory
requirements will likely force all nonfeedback-equipped
vehicles to use feedback systems of some kind. The added
requirement to meet these stringent standards to an elevation
of 10,200 feet will require the manufacturers of many if not
all feedback carburetor equipped vehicles to seek exemptions.
Since this scenario does not include the possibility of
exemptions, these vehicles would be forced to use the control
-------
111-32
technology discussed in the succeeding paragraphs. This same
predicament would be faced by manufacturers of TBI systems, in
this case, however, the sophistication of these systems with
their ability to control fuel/air mixtures more precisely
during both open- and closed-loop operation makes it more
likely that many of these systems could comply with the
requirements of scenario la. These systems may require
additional refinement of the open-loop fuel/air mixtures than
was estimated to occur in scenario lb. It is also possible
that additional changes to the power enrichment and EGR
algorithms contained in the microprocessor unit will be needed.
The requisite exhaust emission control hardware for
vehicles that would qualify for exemptions is, in reality,
likely to be quite varied and would be difficult to establish
at this time. To simplify the analysis, therefore, two
emission control systems have been chosen as representative.
These two systems will be applied to various percentages of the
fleet under the scenarios of interest. The number of vehicles
that would otherwise qualify for exemption increases with the
stringency of the standards (proportional or statutory), and
the altitude at which the standards must be met (6,000 feet or
10,200 feet). As shown in Table III-8, the fraction of the
fleet that is expected to have significant difficulty meeting
the standards is approximately 60 percent for scenario la, 25
percent for scenario lc, and 10 percent for scenario 3b. The
derivation of these percentages was previously presented in
Chapter II, Identification of the High-Altitude Control
Scenarios, and represents a "worst case" assumption.
The first system for vehicles which might otherwise be
eligible for exemption is the electronic load control system
(ELCS). GM devised this system as a way to achieve sea level
emissions at higher elevations. Basically, the ELCS is an
existing GM feedback carburetor system which includes the
addition of a manifold absolute pressure (MAP) sensor. This
device was used in conjunction with engine speed to calculate
the engine load. The load experienced by the vehicle's engine
determines the amount of fuel or quantity of fuel/air mixture
which is required. Since road load is also nearly independent
of altitude, the MAP sensor input can be used to meter the
correct amount of fuel, EGR rate, and spark advance for a given
operating condition regardless of the altitude. GM has stated
that this engine-load control system (ELCS) should result in
emissions performance that is independent of any change in
elevation. Indeed, Figure III-2 shows that this is generally
the case. HC and CO emissions are below the statutory
standards and are basically equivalent at both low and high
altitude. However, NOx emissions are greater at high altitude
than at low altitude. This problem may require further
refinements in the EGR calibration at high altitude. The
ability of the ELCS to control emissions, albeit at a higher
cost, makes it suitable for some "problem" vehicles which are
equipped with carburetors. This system is estimated to be
-------
111-33
Table III-8
Estimated Exhaust Emission Control Requirements
for Vehicles Requiring Exemption
(based on the total fleet)
Control
Hardware
Electronic Load
Control System
(ELCS)
Scenario la:
Continuous
Statutory,
10,200 Feet
W/0 Exemptions
20%
Scenario lc:
Continuous
Statutory,
6,000 Feet
W/Q Exemptions
20%
Scenario 3b:
Continuous
Proportional,
6,000 Feet
W/O Exemptions
10%
Turbocharger
40%
5%
0%
-------
111-34
FIGURE III-2
Emissions Performance
Of Engine Load Control Systems
0.8<
HC
0.6 <
0.4 i
0 2
_ PROPORTIONALCONTROL_
SEA-LEVEL STANDARD
'25L
HI
-I
s
cc
LU
Q.
CO
S
<
cc.
o
co-
CO
o
12 <
10'
PROPORTIONAL CONTROL
6<
CO
CO
LU
Q.
H
4 <
SEA-LEVEL STANDARD
15
1j0« -
NOx
0.5 <
ELEVATION, THOUSANDS OF FEET
NOTE: ALL TESTS AT LOW MILEAGE.
ADJUSTED TO 50,000-MILE LEVELS
USING DETERIORATION FACTORS FROM
1981 C-4CERTIFICATION DATA.
-------
111-35
representative of the emission control hardware that may be
used by 10 percent of the vehicles in scenario 3b, 20 percent
in scenario lc, and 20 percent in scenario la (Table III-8).
It should be noted that vehicle performance may be very
bad if the high-altitude emission control system, regardless of
whether it is carbureted or TBI, includes eliminating power
enrichment at WOT and very heavy EGR. In these cases,
maintaining vehicle performance may force the use of the second
control system which is more expensive. Turbocharging can be
used to overcome the difficult task of meeting the emission
standards at altitude when such situations as increased use of
the power enrichment circuit of the fuel metering device may-
preclude such compliance or when total disconnection of this
circuit as well as heavy EGR lead to unacceptable vehicle
performance. Such drastically degraded performance could
possibly force some vehicles off the market because of safety
considerations or adverse consumer reaction.
A turbocharger is essentially an intake air compressor
which is propelled by an exhaust gas turbine. The compressor
is used to increase the charge of intake air by increasing its
pressure (and density). In most light-duty vehicle
applications, the reason for the use of a turbocharger has been
to improve performance. With increased inlet charge density,
more fuel and air can be processed through a given engine
displacement and, therefore, more power can be generated. In
the future, however, this increased performance may be traded
for higher fuel economy since the same amount of power can be
generated by a lighter, smaller displacement turbocharged
engine as can be generated by a larger, naturally aspirated
engine.
Use of a turbocharger at high altitude would definitely
prevent the loss of power and the attendant performance and
emission control problems that plague naturally aspirated
engines. The turbocharger simply maintains the density of
intake air independent of altitude and the engine cannot tell
the difference. The use of a turbocharger as part of a system
that maintains low-altitude performance at high altitude is not
new. Turbochargers were first widely used for heavy-duty truck
application in the western states in order to maintain
performance when crossing the mountains. The concept of
turbocharging light-duty vehicles in order to maintain
low-altitude performance at high altitude, therefore, has some
precedence. While some performance gains at low altitude still
may be possible, the prime function of the turbocharger in the
high-altitude case is to maintain low-altitude performance.
The application of turbocharging as a high-altitude concept
would enable manufacturers to retain low-performance packages
which may be important for meeting the corporate average fuel
economy (CAFE) standards or advertising considerations.
-------
111-36
While turbochargers should theoretically solve the
high-altitude exhaust emission problem for some spark-ignition
engine vehicles# there appear to be some practical problems
which must be resolved before their use becon.es widespread.
Ford, Chrysler, and GM have all reported some durability and
emission control difficulties with current turbocharged
spark-ignition engines. Also, the current production of these
units is very low. significant expansion of manufacturing
capacity would have to take place if turbochargers were used to
the degree required by scenario la. Although these problems
can be overcome, it would take time. The required leadtime
would very likely extend beyond the 1984 implementation date
for high-altitude standards. Nevertheless, this is the type of
exhaust emission control hardware which may be required to keep
some models in production. Turbocharging is estimated to be
used by all vehicles unable to meet the standards with
modifications to their TBI systems or with ELCS. This accounts
for approximately 5 percent of the vehicles in scenario lc and
40 percent of the vehicles in scenario la (Tables III-8 and
III-9).
The exhaust emission control technology requirements for
the entire fleet under each scenario (la, lc, and 3b) are
summarized in Table III-9.
2. Evaporative Emission Systems
As previously stated, the evaporative emission control
systems are generally independent of the type of exhaust
emission control system used on the vehicle. Therefore, the
evaporative systems for vehicles that would qualify for
exemption are the same as previously estimated nonexempted
vehicles in scenarios with similar control ceilings. For
scenario 3b, complying with the proportional standard up to
6,000 feet would be accomplished with existing hardware. For
scenario lc, complying with the statutory standard at every
elevation up to 6,000 feet would generally require that
affected vehicles be equipped with charcoal storage canisters
with 25 percent larger capacity.
EPA expects that meeting the statutory evaporative
emission standard up to 10,200 feet in scenario la would be
difficult. Additional countermeasures other than the
relatively easy task of increasing the canister capacity ma^ be
required. A lack of data, however, prevents making any
estimate of these added requirements at this time. if
increasing the working capacity of the canister is sufficient
to meet the requirements of scenario la, EPA estimates that
such an increase would be about 50 percent greater than the
capacity of current systems (Table lll-7). This estimate is
based on the fact that the statutory standard at 10,200 feet is
approximately twice as stringent as the proportional standard
at 6,000 feet.
-------
Table III-9
Estimated Exhaust Emission Control Requirements
for Scenarios Vfithout Exemptions
Contol
Control
Scenario la:
Scenario ic:
Scenario 3b:
Hardware
Hardware
Continuous
Continuous
Continuous
Before
After
Statutory,
Statutory,
Proportional,
Hodifi-
Modifi-
10,200 Feet
6,000 Feet
6,000 Feet
cation[a]
cation[a]
W/O Exemptions
W/O Exemptions
W/O Exemptions
OL
OL w/aneroid
0%
0%
23%
OL
Turbocharged
18%
5%
0%
OL
FBC
13%
26%
8%
FBC
ELCS [b]
21%
20%
10%
FBC
Turbocharged
2%
0%
0%
FBC
No Change
0%
0%
8%
TBI
Expansion
39%
59%
13%
TBI
Turbocharged
20%
0%
0%
TBI
No Change
0%
0%
46%
La] OL means nonfeedback or open-loop system.
FBC means feedback carburetor syscem.
ELCS means electronic load control system.
TBI means throttle body injection system.
Expansion means the capability of the existing electronic components
is upgraded.
[b] The feedback carbureted systems that change to electronic load control
systems include a portion of open-loop systems from tue previous
category that have switched to feedback systems. Therefore the
percentages listed for each scenario do not total 100.
-------
111-38
V. EFFECTS OF HIGH-ALTITUDE STANDARDS ON LOW-ALTITUDE CONTROL
TECHNOLOGY
The generic systems presented in Tables 111-4/ III-7 and
III-8 were chosen to reduce the economic impact on the
low-altitude fleet. However, the regulatory strategies in some
scenarios will have, by definition, some unavoidable effects.
These effects may be positive or negative.
High-altitude standards will significantly affect
low-altitude vehicles in scenarios 1 and 3 that require
compliance with proportional or statutory standards at all
altitudes by all vehicles. since all vehicles must be capable
of meeting the applicable standards without modification, these
one-car strategies will require high-altitude emission controls
on any low-altitude vehicles that could not otherwise comply
with the applicable high-altitude standards. Many of these
vehicles will never be operated at high altitudes, hence, it is
logical to expect that the cost of these one-car strategies
generally will be significantly higher than the two-car
strategies without a corresponding reduction in high-altitude
emissions.
The fuel efficiency of low-altitude vehicles may also
change. For instance, if turbochargers are used as a result of
these standards, the fuel economy of the low-altitude fleet
will increase. (This topic is discussed further in the next
section.)
High-altitude regulations could also have a significant
environmental impact throughout the nation (i.e., even at low
altitude). Tables III-4, III-8 and III-9 show the dramatic
change from nonfeedback systems to feedback in scenarios la,
lb, and lc where all vehicles must be certified in compliance
with the statutory standards regardless of altitude. Since
individual control technologies exhibit different in-use
emission characteristics, fleet composite emissions will change
under these scenarios.
The available evidence suggests that the electronic
components of current feedback systems may have a significant
failure rate, and that such failures could lead to emission
increases as a result of excessively rich operating
conditions.[12] Nonfeedback systems, therefore, may have lower
in-use emissions because they do not exhibit such catastrophic
failures. Any projection that the catastrophic failures of
feedback systems will continue is very speculative, however,
because of the preliminary nature of the evidence, and the
uncertain ameliorative affect of inspection/maintenance
programs. Therefore, increasing the number of feedback systems
at all altitudes may have a significant adverse impact on air
quality throughout the nation, but the effect cannot be
confirmed at this time. This will be discussed further in the
Chapter IV.
-------
111-39
VI. EFFECT OF HIGH-ALTITUDE CONTROL TECHNOLOGY ON FUEL ECONOMY
The fuel economy of vehicles affected by a 1984
high-altitude standard could increase or decrease due to the
requisit emission control hardware. Because these fuel economy
changes could significantly affect the overall net cost of 1984
high-altitude standards and they need to be determined as
accurately as possible. Although limited data are available
concerning the fuel economy of vehicles equipped with altitude
compensating control hardware, enough information was available
to estimate the effect of a turbocharger/ a closed-loop
feedback control system, and one or more aneroids on fuel
economy.
A. Turbochargers
When comparing a turbocharged vehicle with a naturally
aspirated vehicle of similar driving performance, the engine of
the turbocharged vehicle will most likely have fewer cylinders,
less total engine displacement, and therefore less weight than
the naturally aspirated engine. This will usually result in a
fuel economy difference. Data examined by EPA showed both an
increase and decrease in fuel economy. In particular, GM and
Ford turbocharged gasoline vehicles were analyzed from data
presented in technical reports or from emission certification
test results.[13,14,15]
First, GM's turbocharged 6-cylinder Buicks were examined:
a 1978 Lesabre and a 1978 Regal. GM studies showed that a
3.8-liter turbocharged 6-cylinder engine gave very similar
driving performance to that of a 5.7-liter, 8-cylinder
engine.[13] The urban fuel economy of the turbocharged vehicle
was 19 miles per gallon (mpg), while the urban fuel economy of
the naturally aspirated engine was 18* mpg.[13] Thus,
turbocharging enhanced fuel economy by roughly 5 percent in
this case. The turbocharged Regal had an urban fuel economy of
21 mpg. The Regal with the naturally aspirated, 8-cylinder
engine exhibited an urban fuel economy of 20 mpg. Again,
roughly a 5 percent increase in fuel economy occurred.
Second, Ford's turbocharged 4-cyclinder, 2.3-liter engine
was examined. According to Ford, the turbocharged 4-cylinder,
2.3-liter engine had slightly better driving performance than
the naturally aspirated 5.0-liter, 8-cylinder engine.[14] The
fuel economy for the turbocharged 2.3-liter engine was 22.0
mpg, while that for the naturally aspirated engine was 19.0
mpg, a 15 percent improvement.
Recent certification data on the Ford 2.3-liter
turbocharged engine, however, conflict somewhat with Ford's
data.[15] The certification data reveal a fuel economy of
18-19 mpg for the 2.3-liter turbocharged Ford engine and lists
a fuel economy of 19, 20, and 21 mpg for Ford's 6-cylinder,
-------
111-40
2.3-liter engines and 18 mpg for an 8-cylinder, 4.2-liter
engine. The 2.2-liter and the 4.2-liter engines were naturally
aspirated. These engines were all installed in the same type
of vehicle as the turbocharged 2.3-liter engine and it can be
assumed that these vehicles had similar driving performance.
Therefore, based on Ford's data, turbocharging the engine of a
vehicle may either reduce fuel economy by about 15 percent or
increase fuel economy by about 5 percent.
The above certification data may not be a true comparison
between vehicles with similar driving performance. Ford has
also indicated in its SAE paper that a 2.3-liter turbocharged
engine has a driving performance rating similar to that of a
much larger 5.0-liter engine, while the certification data
compared the 2.3-liter turbocharged engine with either
2.3-liter engines or a 4.2-liter engines.[15] Thus, the
previous comparisons of fuel economy and performance by GM and
Ford may be more reliable than that constructed from
certification data. It is likely, then, that turbocharged
vehicles will show some fuel economy improvement when compared
with a vehicle of similar driving performance.
In this analysis, a fuel economy increase of 5 percent
will be used as a single best estimate. This estimate is very
conversative considering that both Ford and GM show
improvements of up to 15 percent. In the sensitivity analysis
of Chapter V, a larger fuel economy range of 0-5 percent will
be examined to evaluate the effect on the overall net cost of
these regulations of an even more conservative fuel economy
estimate.
B. Closed-Loop Feedback Control
The fuel ecfonomy change associated with converting an
engine from nonfeedback to feedback control will be examined
for the Ford Granada, Ford Thunderbird, and other Ford vehicles
equipped with the 2.3-liter engine. EPA records show that Ford
certified each of these vehicles with both nonfeedback and
feedback emission control systems within the same model
years.[16,17 J
First, for the Ford Granada, which had an engine
displacement of 4.2 liters and a weight of 3,625 pounds, the
nonfeedback version had a fuel economy of 20 mpg while the
feedback version had a fuel economy of 21 mpg 5 percent
improvement.[16] Second, the Ford Thunderbird with a weight of
3,750 pounds and an engine size of 5.0 liters had a fuel
economy of 20 mpg is shown for both nonfeedback and feedback
systems. Thus no fuel economy improvement is observed.
Finally, the certification results for the Ford 2.3-liter
engine showed the feedback version obtaining 22 mpg, while the
nonfeedback version had fuel economies of 20.0, 20.2, and 20.7
mpg.[17] The fuel economy improvement in this case is 6 to 20
-------
111-41
percent. Thus, based on fuel economy data from the Ford
Granada, Ford Thunderbird, and other Ford vehicles, converting
an engine from a nonfeedback system to a feedback system may
improve fuel economy from 0 to 20 percent.
Admittedly, the data presented here are very limited and
show a very wide range of fuel economy improvement. Thus, it
is difficult to make this range smaller, but EPA believes a
conservative best estimate would range from 0 to 5 percent.
For purpose of estimating fuel savings in this report, a 3
percent improvement will be used as a best estimate. The 0 to
5 percent range and its effect on the net cost of a 1984
high-altitude standard will be analyzed further in the
sensitivity of Chapter IV, Economic Impact.
C. Aneroids
Vehicles equipped with aneroids generally show better fuel
economy than the same vehicle without an aneroid. The fuel
economy improvements of adding one, two, or three aneroicis on
Ford vehicles are shown in Table Ill-10.[18] A straight
average of the numbers shown in this table yields an urban fuel
economy benefit of 3 percent and a highway fuel economy benefit
of 7 percent for vehicles with one or two aneroids, ana an
urban fuel economy of 9 percent and a highway fuel economy of
10 percent for vehicles with three aneroids. The composite
fuel economies (based on a 55/45 urban/highway weighting)
yields the following improvements: 1 aneroid, 5 percent; 2
aneroids, 5 percent} and 3 aneroids, 9 percent.
To be conservative, it will be assumed that the average
fuel economy gains that were estimated above will be the
maximum fuel economy improvements that would occur by utilizing
aneroids. Thus, 1, 2, and 3 aneroids give improvements of 0-5,
0-5, and 0-9 percent, respectively, over vehicles with no
aneroids. Of course, the fuel economy benefit will take place
at high-altitude areas, as this is where an aneroid allows a
leaner air/fuel mixture.
At the time 1984 high-altitude standards become effective,
vehicles will have already had to comply with the 1982-83
interim standards. As a result, most high-altitude vehicles
with nonfeedback emission control systems will already be
equipped with an aneroid on the carburetor. The incremental
benefits of fuel economy from one aneroid in addition to the
one already in place should not bring about any fuel economy
improvement (or an improvement of 0 percent). However, two
aneroids in addition to the existing one could provide a fuel
economy improvement of 0-4 percent. Based on this very limited
amount of data, a single best estimate would be the midpoint of
this range (0-4 percent), or 2 percent. This incremental
savings will be carried through as EPA's best estimate of fuel
economy savings for the addition of two aneroids. However, a
-------
111-42
Table 111-10
Percent Fuel Economy Improvement Due to Aneroids
ine Size
liter)
Fuel Economy
Improvement
of 1 Aneroid
Urban HW
Fuel Economy
Improvement
of 2 Aneroids
Urban HW
Fuel Economy
Improvement
of 3 Aneroids
Urban UW
5.0
+4
NA
+ 5
NA
+8
NA
5.8
+1
+2
0
+2
+9
+7
5.8
+6
+ 10
+7
+ 13
+11
+14
6.6
-1
+13
+1
+11
+5
+ 13
6.6
4
+4
+4
0
+11
+5
-------
111-43
range of fuel economy increments will be evaluated in the
sensitivity analysis of Chapter VI, Economic Impact, because of
the large variability in the Ford data and because of the
limited amount of information that was available with which to
formulate a "best estimate." This range will be 0-2 percent
for the addition of one aneroid and U-4 percent for the
addition of two aneroids.
D. Expansion of Adaptive Memory
Expanding the range of authority for feedback control
systems could lead to fuel economy improvements at hiyh
altitude. Discussions with a few manufacturers indicate that
this improvement could be 1 percent. However, no data has been
encountered concerning fuel economy on this modification, and
thus to be conservative, the benefit will not be considered in
this report.
E. Summary
The following best estimates for fuel economy improvement
were determined: turbocharger, 5 percent? closed-loop feedback
control (compared to open-loop of same vehicle), 3 percent; one
additional aneroid (to a vehicle with an existing aneroiu), 0
percent; and 2 additional aneroids (to a vehicle with an
existing aneroid), 2 percent. These fuel economy benefits are
summarized in Table III-ll. The effect of these fuel economy
improvements on the overall net cost of potential high-altitude
regulations will be analyzed in Chapter V of this document.
-------
III-44
Table III-ll
Percent Fuel Economy Improvement Due to Altitude
Compensating Emission Control Hardware
"Best Estimate"
Fuel Economy
Control Hardware Improvement
Turbocharger 5
Feedback control 3
Aneroids:
1 additional 0
2 additional 2
-------
II1-45
References
1. "Control of Air Pollution from New Motor Vehicles
and New Motor Vehicle Engines: Final High-Altitude Emission
Standards for 1982 and 1983 Model Year Light-Duty Motor
Vehicles," U.S. EPA, 45 FR 66984, October 8, 1980.
2. Manufacturers Submittals to Rulemaking Docket No.
A-79-14, High-Altitude Emission Standards for 1982 and 1983
Model Year Light-Duty Vehicles, Central Docket section, U.S.
EPA, Washington, D.C.
3. "Final Report - Contract No. 68-03-2891, Task Order
No. 2," Automotive Testing Laboratories, Inc., Prepared for
U.S. EPA, OANR, OMS, ECTD, SDSB, October 23, 1980.
4. "OMS Advisory Circular, Assigned Deterioration
Factors for Small-Volume Manufacturers," U.S. EPA, OANR, OMS,
CDS, February 8, 1979.
5. "Throttle Body Fuel Injection (TBI) — An Integrated
Engine Control System," General Motors Corporation, Emission
Control System Project Center, Bowler, Lauren L., SAE Paper No.
800164.
6. "Technical Feasibility of the Proposed 1982-83
High-Altitude Standards for Light-Duty Vehicles and Li^ht-Duty
Trucks," CTAB/TA/80-3, U.S. EPA, OANR, OMS, ECTD, SDSB, August
1980.
7. Hearing Transcripts for the 1984 High-Altitude
Requirements, U.S. EPA, Washington, D.C., Central Docket
Section, Docket No. A-80-01.
8. "Motor Vehicle Emission Standards for Carbon
Monoxide and Nitrogen Oxides," Draft, U.S. EPA, OANR, OMS,
ECTD, April 1, 1981.
9. "Final Regulatory Analysis - Environmental and
Economic Impact Statement for the 1932 and 1983 Model Year
High-Altitude Motor Vehicle Emission Standards," U.S. EPA,
OANR, OMS, ECTD, SDSB, Ann Arbor, MI, October 198U.
10. Results of Meeting Between Chrysler and EPA
Representatives, Memo to the Record from R. s. Wilcox, U.S.
EPA, OANR, OMS, ECTD, SDSB, January 19, 1980.
11. Manufacturers Submittals, 1984 High-Altitude
Requirements, U.S. EPA, Central Docket Section, Docket No.
A-80-01, Washington, D.C.
12. "Derivation of 1981 and Later Light-Duty Vehicle
Emission Factors for Low-Altitude, Non-California Areas," U.S.
EPA, OANR, OMS, ECTD, SDSB, EPA-AA-IMS/80-8, November 1980.
-------
111-46
References (cont'd)
13. "Buick's Turbocharged V-6 Powertrain for 1973,"
General Motors Corporation, Buick Motor Division, Wallace, T.
F., SAE Paper No. 780413.
14. "Turbocharging Ford's 2.3-Liter Spark Ignition
Engine," Ford Motor Company, Dertian, H. H., G. W. Holiday, and
G. W. sandburn, SAE Paper No. 790312.
15. 1980 Gas Mileage Guide, First Edition, U.S. EPA,
OANR, OMS, Certification Division, September, 1979.
16. 1981 Test Car List, First Edition, U.S. EPA, OANR,
OMS, Certification Division, Ann Arbor, MI.
17. Certification Weekly Report to Michael Walsh, U.S.
EPA, OANR, OMS, Certification Division, December 12, 1980.
18. EPA/Ford Meeting on Ford Motor Company High-Altitude
Emission Program Plans, Ford Motor Company, June 20, 1978.
-------
Chapter IV
Environmental Impact for Light-Duty Vehicles
I. INTRODUCTION
A. Scenarios Being Studied
The four basic control scenarios being evaluated were
described earlier. However, to facilitate the analysis in this
chapter, they are restated briefly below:
1. Base Scenario: Fixed-Point Proportional
This scenario is essentially a continuation of the 1982-83
interim high-altitude emission standards.[1] These standards
apply to areas above 4,000 feet and were determined by
increasing the low-altitude emission standards in proportion to
the effect of altitude on uncontrolled emissions.
Certification in high-altitude areas would be at a specific
altitude (e.g., 5,300 feet).
2. Scenario 1: Continuous Statutory
All light-duty vehicles (LDVs) sold in this country must
certify to tne same emission standards, independent of where
they are sold. These standards would be the same as those in
low-altitude areas (below 1,800 feet) and would apply to every
vehicle sold up to a maximum elevation of 6,000 feet.
3. Senario 2: Fixed-Point Statutory
The same emission standards would apply to vehicles sold
in low-altitude areas (below 1,800 feet) and in high-altitude
areas (above 4,000 feet) areas between 1,800 and 4,000 feet
would not be subject to special control.
4. Scenario 3: Continuous Proportional
All li^ht-duty vehicles would be subject to control but
the decree of control would vary in direct proportion to
altitude for areas between 1,800 and 6,000 feet. The standards
at 5,300 feet would be the same as in the base scenario.
B. General Approach
As in the previous chapter, the fixed-point proportional
standards of the base scenarios are assumed to be in effect for
1984 and later. Therefore, all alternative control strategies
(scenarios 1 through 3) will be analyzed to determine their
incremental environmental impacts.
-------
IV-2
Two basic measures will be used to evaluate the impact of
the various alternative scenarios. The first is simply the
overall emission reduction of hydrocarbons (HC), carbon
monoxide (CO), and oxides of nitroyen (NOx). These reductions
will be used primarily to determine the cost effectiveness of
each scenario.
The second method goes further and evaluates the relative
impacts of the three scenarios on air quality. Here, computer
models will be employed to examine the consequences of each
scenario on the abilities of four specific high-altitude cities
to comply with the National Ambient Air Quality Standards
(NAAQS) for CO, NOx, and ozone. These models also project
yearly emission reductions relative to pre-control base years.
This chapter concludes with a sensitivity analysis of
certain critical assumptions.
II. TOTAL EMISSIONS
Although this report focuses on high-altitude areas, as
described earlier, scenarios 1 and 3 have standards that apply
to intermediate altitudes as well. Therefore, any
determination of the emission reductions under these scenarios
should include the impact in these regions in addition to
high-altitude areas.
Lifetime emissions were determined from LDVs sold during
1984 through 1988 in the areas affected by the different
scenarios. This particular 5-year period was chosen to
correspond with the time increment used to determine tne
aggregate economic costs of the four scenarios. The cost
effectiveness of the various options being studied can then be
determined since both cost and emissions data cover the same
periods of time.
A. Basic Methodology
In order to determine total lifetime emissions from tne
vehicles being studied, three basic factors were multiplied
together: 1) the number of miles the average vehicle is
expected to cover in its lifetime, 2) emission factors (the
amount of pollutant emitted per vehicle per mile of travel),
and 3) the number of vehicles sold in the affected areas.
1. Average Miles Traveled
For light-duty vehicles, the average lifetime mileage used
in this study is 100,000 miles.[1] For light-duty trucks
(LDTs) (addressed in Chapter VIII), the lifetime is 120,000
miles. [2] In both cases it is assumed that the lifetimes are
independent of altitude, that is, a light-duty truck or vehicle
is expected to travel the same number of miles during its
lifetime whether it operates at low- or high-altitude regions.
-------
IV-3
2. Emission Factors
EPA has done considerable work in an attempt to determine
accurate emission factors for mobile sources. To determine how
well vehicles perform in actual use, EPA administered a series
of exhaust emission surveillance programs. Test fleets of
consumer-owned vehicles in various major cities 1 were selected
by model year, make, engine size, transmission, and carburetor
in such proportion as to be representative of both the normal
production of each model year ana the contribution of that
model year to total vehicle miles traveled. These programs
have focused principally on light-duty vehicles and light-duty
trucks. As a result of EPA's test programs, emission factors
for past model years are known with the greatest degree of
accuracy, while projected emission factors for vehicles with
recently introduced control technologies are subject to
considerable uncertainty.
The surveillance program data were used to determine mean
emissions by model year in each calander year, the change in
emissions with the accumulation of mileage and age, the
percentage of vehicles complying with standards, and the effect
of vehicle parameters on emissions (engine displacement,
vehicle weight, etc.). These surveillance data, along with
prototype vehicle test data, assembly line test data, and
technical judgment, formed the basis for the M0BILE2 emission
factor model which was then used to determine fleetwide
composite emission factors for the specified study years (used
as inputs to EKNA and Rollback models discussed later in this
chapter).[3]
Table IV-1 lists the low-altitude (1,800 feet) emission
factors in grams per mile for the gasoline-fueled and
diesel-powered light-duty vehicles that were used as the basis
for determining emission factors at other altitudes in this
study. These emission rates were determined from tests
conducted using the Federal Test Procedure (FTP) and are
further discussed in Appendix II.[4] The control technology
expected to be required for each scenario has already been
discussed in Chapter III.
When determining the high-altitude emission rates for
areas above 4,000 feet, two pertinent factors must be
considered which are unique to this study. The first is the
exact altitude for which emission rates are desired and the
second is vehicle age in terms of miles traveled. The
high-altitude emission data available were based on tests
conducted at Denver (elevation 5,300 feet). Thus, the emission
factors developed to study the high-altitude impact of the
various scenarios, listed in Table IV-2, were based on this
elevation. implicit in this procedure is the assumption that
the average high-altitude place of residence is 5,300 feet.
Since Denver, the largest high-altitude cit^, lies at this
approximate elevation and since 5,300 feet is roughly the
average altitude above 4,000 feet that these scenarios apply,
this assumption does not significantly affect the results of
-------
IV-4
Table IV-1
Emission Rates for 1984-88 Light-Duty Vehicles at 1,800 Feet[a]
LDGVCb]
Pollutant
HC Zero-Mile Emission
Level (g/mi)
Deterioration Rate
(g/mi/10,000 miles)
Average Lifetime Emis-
sion level (g/mi)tc]
CO Zero-Kile Emission
Level (g/mi)
Deterioration Rate
(g/mi/10,000 miles)
Average Lifetime Emis-
sion Level (g/mi)-
NOx Zero-Mile Emission
Level (g/mi)
Deterioration Rate
(g/mi/10,000 miles)
Average Lifetime Emis-
sion Level (g/mi)
Evap. Zero-Mile Emission
HC Level (g/mi)[d]
Deterioration Rate
(g/mi/10,000 miles)
Scenario 1
0.37
0.17
1.22
4.27
2.13
14.52
0.56
0.10
1.06
0.10
0.0
Average Lifetime Emission 0.10
Level (g/mi)
Scenarios 1,
2, 3, & Base
0.33
0.18
1.23
3.21
1.88
12.61
0.63
0.11
1.18
0.10
0.0
0.10
LDDV[b]
All Scenarios
0.39
0.03
0.54
1.27
0.05
1.52
0.75
0.05
1.0
[a] Emission Rate = Zero Mile Level + (Cummulative
Mileage/10,000)(Deterioration Rate)
Values on this table are further discussed in the Appendix.
[b] Light-duty gasoline-powered vehicle
Light-duty diesel-fueled vehicle
[c] Based on average lifetimes of 100,000 miles for light-duty vehicles,
[dj See Table IV-7 for derivation of these figures.
-------
IV-5
Table IV-2
Emission Rates for 1984-88 Light-Duty Vehicles at 5/300 Feet[a]
Scenario
Base and #3 #1 #2
Pollutant LDGV LDDV LDGV LDDV LDGV LDDV
HC Zero Mile Emission 0.44 0.54 . 0.38 0.39 0.34 0.39
Level (g/mi)
Deterioration Rate 0.19 0.03 0.18 0.03 0.19 0.03
(g/mi/10,000
miles)
Average Lifetime Emis- 1.39 0.69 1.28 0.54 1.29 0.54
sion level (g/mi)[b]
CO Zero-Mile Emission 6.25 2.22 5.22 1.27 3.75 1.27
Level (g/mi)
Deterioration Rate 2.22 0.05 2.76 0.05 2.24 0.05
(g/mi/10,000
miles)
Average Lifetime Emis- 17.35 2.47 19.02 1.52 14.95 1.52
sion Level (g/mi)
NOx Zero-Mile Emission 0.58 0.75 0.54 0.75 0.58 0.75
Level (g/mi)
Deterioration Rate 0.10 0.05 0.10 0.05 0.10 0.05
(g/mi/10,000
miles)
Average Lifetime Emis- 1.08 1.0 1.04 1.0 1.08 1.0
sion Level (g/mi)
Evap. Zero-Mile Emission 0.13 - 0.10 - 0.10
HC Level (g/mi)[c]
Deterioration Rate 0 - 0 - 0 -
(g/mi/10,000
miles)
Average Lifetime Emis- 0.13 - 0.10 - 0.10
sion Level (g/mi)
[a] Emission Rate = ZML + (M)(DR)
Where: ZML = Zero-Mile Level
M = Cumulative Mileage/10,000 (cumulative mileage = 50,000
miles, half the lifetime mileage)
DR = Deterioration Rate
Values in this table are further discussed in Appendix II.
[b] Based on average lifetimes of 100,000 miles for light-duty vehicles.
[cj See Table IV-7 for the derivation of these numbers.
-------
IV-6
this analysis. The procedure used to determine emissions in
intermediate altitudes (1,800-4,000 feet) is discussed below.
Light-duty vehicles sold at intermediate altitudes are
subject to special standards under scenarios 1 and 3 (as
opposed to scenario 2 and the base scenario where only those
vehicles sold above 4,000 feet are subject to control). Thus,
to determine the emissions from vehicles in these intermediate
altitudes another set of emission factors needs to be
developed. Ideally, a different set of emission factors would
be derived for each altitude above 1,800 feet at which vehicles
are driven. These values would then be used to examine
scenarios 1 and 3. Such a methodology would involve literally
thousands of emission factors, however, and is thus beyond the
scope of this report. Instead, EPA has determined the average
altitude where the population above 1,800 feet (the altitude at
which these scenarios first apply) resides and calculated an
average set of emission factors based on that altitude to
approximate the emission levels from vehicles above 1,800
feet. By examining the population and altitude distributions
of the 20 largest U.S. cities, (excluding those in California,
which has its own motor vehicle emission program) this
elevation was determined to be 4,100 feet. For the purpose of
this study it is assumed that the per capita ownership of
light-duty vehicles does not significantly vary with altitude.
Thus, population distribution is an appropriate surrogate for
light-duty vehicle usage.
The 5,300 feet emission factors can be used to estimate
emissions in areas above 4,000 feet. By using the 4,100 feet
emission factors, emissions in areas above 1,800 feet can be
determined. The difference between the total emissions
calculated by these two sets of emission factors is equal to
the emissions in the intermediate altitudes of 1,800 to 4,000
feet.
The second important factor used in determining
high-altitude emission rates is vehicle age. This is important
since emissions tend to increase with vehicle usage. To
estimate lifetime emissions, emission factors were determined
for vehicles that were halfway through their average lifetimes
(i.e., 50,000 miles). The actual equations used in this
procedure can be found in Tables IV-1 and IV-2.
Table IV-3 contains emission factors for the 4,100 feet
elevation; these factors were obtained by interpolating values
from Tables IV-1 and IV-2. The emission factors developed to
determine total emissions do not include the potential benefits
of inspection/maintenance programs. This was done because most
areas affected by the various scenarios are not expected to
have such programs. A further discussion of the derivation of
the emission factors in Tables IV-1 ana IV-2 can be found in
Appendix II.
-------
IV-7
Table IV-3
Average Lifetime Emission Rates for 1984-88
Light-Duty Vehicles at 4,100 Feet (
-------
IV-8
Frora Table IV-1, it is apparent that under scenario 1 the
LDGV lifetime emission levels for CO are significantly greater
than those for the other control scenarios at 1,800 feet. This
is the result of one of the most uncertain assumptions used in
developing the emission factors for this analysis (i.e., the
expected failure rate for feedback emission control systems).
In Chapter III, it was estimated that manufacturers would
need to replace existing nonfeedback emission control systems
with feedback systems to meet the stringent all-altitude
requirements of scenario 1. Current information indicates that
some in-use feedback systems are subject to catastrophic
failures which cause CO emissions to increase well beyond
allowable levels. Since nonfeedback systems do not exhibit
such gross failures, emissions in both low- and high-altitude
areas of the country could increase with a switch to feedback
systems. However, the data base with which these emission
factors were developed is very limited and extrapolating these
results to characterize the behavior of feedback systems in the
future is speculative. indeed, it is possible that additional
information will show markedly lower failure rates than are
assumed in this analysis. If this occurs, the potential
adverse impact could be moderated or eliminated at low altitude
and emission reductions like those found under scenario 2 may-
be possible at high altitude. The sensitivity of the analysis
to such a reduction in the expected catastrophic failure rates
for feedback vehicles in scenario 1 is discussed further at the
end of this chapter.
3. Number of Vehicles
Nationwide sales projections of new light-duty vehicles
for each year are presented in Table IV-4. These projections,
published by Data Resources Corporation, [5] take into account
such factors as the driving-age distribution of U.S. citizens,
changes in real disposable income, and unemployment estimates.
The percentage of diesel-powered vehicles in the total fleet
was taken from previous EPA estimates made to support the
recently promulgated particulate emission standards from LDDVs
and LDDTs.[6]
As discussed earlier, the scenarios being evaluated do not
always affect the emissions performance of each car sold in
this country. Scenario 2 and the base scenario only apply to
vehicles sold above 4,000 feet and scenarios 1 and 3 have
intermediate altitude control in addition to that at high
altitude. Thus, sales projections need to be determined for
areas affected by each of the scenarios. Based on cities whose
population is over 10,000 (excluding those in California), EPA
found that approximately 5.25 percent of the U.S. population
resides above 1,800 feet and 3.1 percent above 4,000 feet.
For the purposes of this study, it will be assumed that
all cars sold above 4,000 feet will comply with high-altitude
standards as depicted bi the various scenarios. (The fraction
-------
IV-9
Table IV-4
Nationwide Sales Projections of
Light-Duty Vehicles!5,6] (millions of vehicles)
Year Gasoline-Fueled Diesel-Powered
1984 10.5 1.1
1985 10.3 1.3
1986 9.9 1.6
1987 9.5 1.9
1S88 9.4 2.0
5-Year Total 49.6 7.9
-------
IV-10
of cars above the ceilings for which the standards apply is too
small to significantly affect the outcome of this analysis.)
Scenarios 1 and 3, therefore, affect the emissions of 5.25
percent of the non-California, light-duty motor vehicle fleets,
while scenario 2 and the base scenario will apply to 3.1
percent of these vehicles. As previously stated, it is assumed
that population distribution is an appropriate surrogate for
light-duty vehicle sales.
Since the scenarios being analyzed do not apply to
California for reasons discussed earlier, the above-mentioned
intermediate- and high-altitude sales fractions should apply to
the nationwide sales estimates in Table IV-4 minus those in
California. Population estimates indicate that roughly 10
percent of the nation resides in California.[7] Therefore,
assuming this percentage remains fairly constant for the years
being studied and that light-duty vehicle sales are
proportional to the population distribution, the numbers of
vehicles subject to emission control under the various
scenarios can be readily calculated and are shown in Table IV-5.
4. Evaporative Hydrocarbon Emissions
The 1982-83 interim high-altitude emission regulations
call for an evaporative hydrocarbon emission standard of 2.6
grams per test for light-duty gasoline-fueled vehicles and
trucks; the low-altitude standard is 2.0 grams per test.tl]
Each test consists of a diurnal and a hot-start portion in
order to simulate actual evaporative conditions.[8] To
determine the amount of evaporative hydrocarbons emitted per
mile, results from the hot-start test are first multiplied by
the expected number of trips per day. Next, results from the
diurnal test are added to this product and the sum is divided
by the number of miles traveled per day. Based on in-use data
and the above-mentioned standards, evaporative hydrocarbon
emissions up to 1,800 feet are 0.10 grams per mile (Table
IV-6). similarly, 0.13 grams of evaporative hydrocarbons are
emitted per mile at 5,300 feet. E>i interpolating the inputs
used to derive these emission rates, it was determined that
0.12 grams of evaporative hydrocarbons are emitted per mile at
4,100 feet. Evaporative emissions under the various scenarios
were then determined by using these emission factors together
with the same vehicle miles traveled used to determine total
exhaust emissions.
B. Presentation and Discussion of Results
Following the procedures outlined in the previous section,
total hydrocarbon (HC), carbon monoxide (CO) and oxides of
nitrogen (NOx) emissions from 1984-88 LDVs were determined for
each scenario. These are given in Tables IV-7 and IV-8. In
any evaluation of pollution control options, a comparison of
the relative amounts of emission reduction is beneficial. Such
a presentation can be found in Tables IV-9 and IV-10.
-------
IV-11
Table IV-5
Number of New Vehicle Sales in Selected Areas
for Indicated 5-Year Periods (millions of vehicles)
LDGV[a] LDDV
Scenarios 1 and 3 2.34 0.37
(above 1,800 feet)
Scenarios 2 and the 1.38 0.22
Baseline (above 4,000
feet)
1 The 5-year period for LDVs is 1984-88.
-------
Table IV-6
Determination of Evaporative Hydrocarbon Emission Levels[aJ
Altitude Hot Start Diurnal Emission Rate
(feet) (grams/test) Tr ips/pay (^rams/test) Miles/pay (^raus/mile) [l>J
1,800 0.63 3.05 1.07 31.1 0.10
4,100 0.75 3.05 1.28 31.1 0.12
5,300 0.82 3.05 1.39 31.1 0.13
[a] Values reported for 1,800 and 5,300 feet were based on in-use uata.
The 4,100 values were obtained by interpolation.
[b] Evaporative Emission Rate = (HS)(TPD) + P
MPD
Where:
HS = Hot Start
TPP = Trips Per Pay
D = piurnal
MPP = Miles Per Pay
-------
IV-13
Table IV-7
Total Lifetime Emissions of 1984-88
LDVs Above 4#00Q Feet (1,000 metric tons)
Scenario
Base
and #3
Pollutant
HC
CO
NOx
Evap. HC
Vehicle Class
LDGV
191.8
2394
149.0
17.94
LDDV
15.18
54.34
22.0
HC
CO
NOX
Evap. HC
176.7
2625
143.5
13.8
11.88
33.44
22.0
HC
CO
NOX
Evap. HC
178.0
2063
149.0
13.8
1.88
33.44
22.0
-------
IV-14
Table IV-8
Total Lifetime Emissions of 1984-88 LDVs from
1,800-4,000 Feet for Affected Scenarios
(1,000 metric tons)
Vehicle Class
Scenario
Pollutant
LDGV
LDDV
Base [a]
HC
122.9
9.5
CO
1302
24.86
NOx
109.5
15.0
Evap. HC
10.2
—
1
HC
118.2
8.12
CO
1496
22.76
NOx
102.2
15.0
Evap. HC
9.6
—
3
HC
121.8
8.52
CO
1284
24.86
NOx
106.1
15.0
Evap. HC
10.2
-
[a] Note that the base scenario calls for no special standards
for these altitudes. The emission factors used here come
from Table IV-3 and Appendix II, adjusted to the desired
altitude by the same procedures discussed previously.
Evaporative emissions are the same as those in scenario 3.
These base scenario emission levels are used to determine
emission reductions from scenarios 1 ana 3 in intermediate
altitudes.
-------
IV-15
Table IV-9
Lifetime Emission Reductions for 1984-88
High-Altitude Vehicles Relative to Base Case
for Regions Above 4/000 Feet (metric tons)
Vehicle Class
Scenario
Pollutant
LDGV
LDDV
1
HC
15,200
3,300
CO
-231,000
20,900
NOx
-5,500
0
Evap. KC
4,140
-
Total HC
19,340
3,300
2
HC
13,800
3,300
CO
331,000
20,900
NOx
0
0
Evap. HC
4,140
-
Total HC
17,940
3,300
3
HC
0
0
CO
0
0
NOX
0
0
Evap. HC
0
0
Total HC
0
0
-------
IV-16
Table IV-10
Lifetime Emission Reductions for 1984-88
High-Altitude Vehicle Relative to Base Case
for Regions Between 1/800 and 4#000 Feet (metric tons)
Vehicle Class
Scenario
Pollutant
LDGV
LDDV
1
HC
4,700
1,380
CO
-194 f 000
2,100
NOX
-7,300
0
Evap. HC
600
-
Total HC
5,300
1,380
3
HC
1,100
980
CO
18,000
0
NOX
3,400
0
Lvap. HC
0
-
Total HC
1,100
980
-------
IV-17
As shown in Tables IV-9 and IV-10, implementing scenario 1
would reduce exhaust HC emissions from 1984-88 LDGVs by 15,2013
metric tons in areas which lie above 4,000 feet. In areas
between 1,800 and 4,000 feet, an additional reduction of 4,700
metric tons beyond those provided by the base case could be
realized.
Looking at the results for CO emissions in these areas, it
is obvious that scenario 1 may have significant detrimental
effects. Above 4,000 feet, 231,000 more metric tons of CO could
be emitted under scenario 1 than under tne base scenario. From
1,800 to 4,000 feet, an additional iy4,000 metric tons could be
emitted. These emission penalties, as discussed earlier, are
due to the projected higher emission rates of feedback vehicles
when they undergo catastrophic failure. (Scenario 1 would
require that essentially all light-duty gasoline-fueled
vehicles be equipped with feedback systems, as opposed to only
69 percent under the other scenarios.) However, the predicted
increase in CO is particularly sensitive to the failure rate
that has been assumed in developing the emission rates used in
this analysis. The failure rate may, in reality, be far less.
The effect of such a decrease in the failure rate is further
discussed at the end of this chapter.
NOx emissions would decrease under scenario 1 by 5,500
metric tons in regions above 4,000 feet and by 7,300 metric
tons between 1,800 and 4,000 feet. Similarly, evaporative HC
emissions would decrease under scenario 1 by approximately
4,740 metric tons above 1,800 feet. This is due to the
application of larger canisters (as described in the Technology
section). These evaporative HC reductions plus reductions in
exhaust HC emissions yield a total HC reduction of
approximately 26,640 metric tons for scenario 1.
Again, if the assumed catastrophic failure rates for
feedback vehicles in this analysis are valid, scenario i will
also affect emissions in areas below 1,800 feet since vehicles
sold at lower elevations would also adopt feedback systems.
The effects this might have on emissions in tnese areas is
indicated by Table IV-1. HC emissions from 1984 and later
model year LDGVs would decrease by 0.8 percent but CO emissions
would rise by 18.3 percent. NOx emissions would decrease by
8.3 percent. While the increases in CO emissions do not appear
to be significant, it is important to remember that these
increases would occur over approximately 95 percent of tne
nationwide non-California fleet. Table IV-4 shows that roughly
47 million vehicles would be so affected in the years being
studied. Thus, the implementation of scenario 1 would cause CO
emissions from 1984-88 light-duty, gasoline-fueled vehicles to
increase by approximately 10,900,000 metric tons in areas below
1,800 feet. However, as mentioned before, the estimated rate
of catastrophic failure is tenuous at best and could easily
change in the future. The effect of lowering the failure rate
will be examined at the end of this chapter.
-------
IV-18
Comparisons of scenario 2 to the base scenario in terms of
total emissions can be found in Table IV-9. This option, if
implemented/ would decrease exhaust HC emissions from 1984-88
LDGVs by 13,800 metric tons and lower CO emissions by 331,000
metric tons in regions above 4,000 feet. No further NOx
reductions would result under scenario 2 but an additional
4,140 metric tons of evaporative HC emissions would be removed
from the atmosphere in these high-altitude areas. This would
yield a total HC reduction of roughly 17,900 metric tons. In
other areas of the country, scenario 2 and the base scenario
have the same emission characteristics.
To examine the benefits of requiring LDGVs sold in
intermediate regions to comply with proportional standards
similar to those for high-altitude areas under the 1982 and
1983 interim standards, scenario 3 was devised. As mentioned
earlier, scenario 3 and the base scenario would offer the same
emission reduction potential in areas above 4,000 feet. From
Table IV-10, it can be seen that 1,100 metric tons of HC and
18,000 metric tons of CO would be removed per year under this
scenario in intermediate altitudes relative to the base
scenario, NOx emissions would not change in this scenario over
the base.
In conclusion, even though scenario 1 would reduce HC
emissions nationwide, the potential negative effect of
catastrophic failures in feedback-equipped vehicles could cause
CO emissions to increase throughout the country. If the
assumed failure rate for feedback systems is valid, scenario 1
does not appear to be a viable alternative to the base
scenario. Scenario 2 woulo apparently reduce emissions from
light-duty gasoline-powered vehicles in high-altitude areas
more than any other option. While scenario 3 would not further
reduce emissions over the base scenario above 4,000 feet, it
would reduce HC, CO, and NOx emissions in regions of the
country lying between 1,800 and 4,000 feet, scenario 2 would
not. However, high-altitude areas are in greater need of
improved emission control than the intermediate areas, and the
CO emission reduction offered by scenario 2 is much greater
than that of the other scenarios. Also, relatively little
difference lies in the reduction potential tor the other
pollutants among the scenarios. Thus, scenario 2 appears to
offer the greatest emission reductions of all the scenarios
which were analyzed as alternatives in this report to
continuing the current proportional standards.
When considering the impact of the various options on
emissions from diesel-powered light-duty vehicles (LDDVs),
comparisons are simplified. Returning to Tables IV-9 and
IV-10, scenarios 1 and 2 offer the same reduction potentials in
regions above 4,000 feet. (This is because both scenarios
require essentially the same type of control technology.) The
emissions reductions for these scenarios are approximately
3,300 metric tons of HC ana 20,900 metric tons of CO from
-------
IV-19
1984-88 vehicles. As was the case with LDGV emissions,
scenario 3 and the base scenario produce the same benefits in
areas above 4,000 feet. No NOx reductions are expected from
the control technology which is applied to these vehicles.
Since diesels are not a significant source of evaporative HC
emissions (due to inherent diesel fuel characteristics), diesel
evaporative emissions are riot evaluated in this report.
In areas between 1,800 and 4,000 feet, scenario 1 would
reduce LDDV HC emissions by rouyhly 1,380 metric tons and CO
emissions by 2,100 metric tons. Scenario 3, the only other
control option which controls emissions in this range, would
reduce lie emissions by 980 metric tons. No benefit is seen
with reyard to CO reductions since diesels emit relatively
small amounts of CO compared to yasoline-powered vehicles and
can comply with either the statutory or proportional standards
without significant modifications.
In order to gain further insight into the environmental
consequences of the four scenarios being examined, an analysis
of the air quality impact was performed. This is discussed in
the following section. After the results of that analysis are
presented and discussed, the effect of critical assumptions
made in this environmental impact study will be assessed.
III. AIR QUALITY
A. Basic Methodology
In order to evaluate the relative air quality impacts of
the control scenarios, two computer models were used whicn
attempt to represent the relationship between pollutant
emissions and the resulting ambient pollutant concentration.
These are the Rollback and Emperical Kinetic Modeliny Approach
(EKI-IA) models, rollback was used to study CO and NOx while the
more complex EKMA model was utilized for oi.one. Detailed
discussions of these models can be found in the
literature.[9,10]
In preparing the air quality projections, future control
strategies and growth rates were applied to baseline emission
rates for various non-motor vehicle source categories taken
from the National Emissions Data System (NEDS).[11] These data,
in combination with similar projections made for motor vehicle
sources of air pollution (discussed below), allow an evaluation
of the effects on air quality of the various scenarios. Due to
the large number of composite emission factors used in this
analysis, they are not included in this chapter but are listed
in Reference 11.
With both the Rollback and EKMA models, the relative
changes from scenario-to-scenario are more reliable than the
absolute predictions of pollutant concentration. Following
-------
IV-20
this guide, results are compared to the baseline emission
inventories in two ways: 1) changes in the percent reductions
from the base year, 1979, and 2) changes in the number of NAAyS
violations expected. This was done for 1986 through 199U and
in 1993 and 1995.
The air quality analysis in this chapter focuses on tiiose
high-altitude cities which are projected to maintain or develop
air quality problems in the future. For CO, these are:
Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque,
and Salt Lake City. For ozone, Denver and Salt Lake City were
modeled. For NOx, only Denver is expected to have difficulties
meeting the NAAy)S.
B. Growth Rates and stationary source Control Assumptions
To project base year inventories and air quality
concentration levels, it is necessary to estimate the future
activity levels of both mobile and stationary pollution
sources. In general, two sets of growth rates were used in
this analysis to provide a range of air quality estimates for
each pollutant. A special zero-growth rate case was also run
for CO and NOx to indicate the air quality impact of a
significant error in estimated growth rates. However, since
high-altitude areas currently experience relatively high-growth
rates, the results of this analysis are not discussed here and
are instead presented in Appendix II.
As shown in Table IV-11, LDV and LDT vehicle miles
traveled (VMT) were projected to grow by 0.4 to 2.4 percent per
year, compounded. This range was arbitrarily established as
being +1 percentage point of the historical growth rate for
these vehicle classes. [9] Similarly, HDG and iIDD VMT were
projected to change at the rate of -3 to -1 and 4 to 6 percent,
respectively. These heavy-duty growth rates are based on sales
figures indicating that diesel trucks are replacing
gasoline-powered trucks in the heavy-duty fleet.[12]
The stationary and off-highway mobile source growth rates
which were used in this analysis are consistent with EPA's most
recent guidelines for conducting air quality modeling
analyses. The various growth rates for each pollutant are
presented in Appendix II. The additional assumptions regarding
stationary source emission controls are described in References
9 and 13.
C. Presentation and Discussion of Results
As mentioned earlier, Rollback and EKMA results from
scenarios 1, 2, and 3 will be compared to those from the base
scenario in two ways: changes in projected numbers of NAAyS
violations and relative percentage reductions in pollutant
concentrations. The results in the first half of Tables IV-12
-------
IV-21
Table IV-11
Growth Rates for Light-Duty Vehicles,
Light-Duty lrucks#and Heavy-Duty Vehicles
Annual Compound Growth Rate (percent)
Low High
Light-Duty Gasoline Vehicles +0.4 +2.4
Light-Duty Gasoline Trucks +0.4 +2.4
Heavy-Duty Gasoline Vehicles -3.0 -1.0
Light-Duty Diesel Vehicles +0.4 +2.4
Heavy-Duty Diesel Vehicles +4.0 +6.0
-------
Table IV-12
Average Percent Reduction in Expected Maximum 1-Hour ozone
Concentrations from 1979 Base year in Denver and Salt Lake City (low and his
-------
IV-23
through IV-15 are based on controls placed on light-duty
vehicles and assume the presence of inspection/maintenance
(I/M) programs in these areas. Similar results without I/M are
found in the lower half of these tables. Light-duty trucks
will be discussed separately in Chapter X.
Tables IV-12 and iv-13 contain the air quality impacts of
ozone. It should be pointed out that the hydrocarbon emission
factors used as input to these and other ozone projections in
the air quality analysis contain contributions made by
evaporative emission losses. As can be seen, there is a range
of values reported in these tables. This was done in order to
reflect two different possible ratios of nonmethane hydrocarbon
to oxides of nitrogen ambient concentrations (i.e., 7 to 1 and
9.5 to 1). (The EKMA model relies upon such a technique to
yield ozone projections.) The future ozone air quality should,
therefore, fall within the range of the estimates.
Table IV-12, shows that only scenarios 1 and 2 are
predicted to yield reductions in the expected maximum ozone
concentrations in Denver and salt Lake City below that already
provided by the base scenario. Table IV-13 shows that in 1989
and 1990 without I/M, one less violation of the ozone NAAQS
could occur under the high-growth case of both scenarios 1 and
2, compared to the base scenario. In 1993, only scenario 1
appears to provide fewer violations than the base scenario, one
less under high growth. All scenarios lead to the same number
of violations in 1995.
The results with I/M indicate that the number of
exceedances is projected to be the same for each scenario in
all years except 1989. In that year, scenario 1 could lead to
one less violation under low growth.
Tables IV-14 and IV-15 contain similar results for CO
levels as only scenarios 1 and 2 offer different projections
from the base scenario. Looking just at scenario 1, one more
violation is projected for 1993 for the high-growth, no-I/M
case compared to the base scenario (Table IV-15). In other
years, there appear to be no high-growth differences in the
number of exceedances under these two scenarios. With I/M,
Table IV-14 shows that scenario 1 should result in a slightly
greater reduction in the expected second highest 8-hour CO
concentrations in several of the years studied compared to the
base scenario. However, this benefit is quite small since
scenario 1 and the base scenario do not differ in the number of
projected violations under the I/M case (see Table IV-15).
Scenario 2 appears to be the only alternative to the base
scenario which can offer fewer CO NAAQS violations. Without
I/M, the expected benefits of scenario 2 begin in 1986, the
first year investigated, when one less violation is projected
under the low-growth case. In each year after that until 1993,
-------
Table IV-13
Number of Violations of Ozone NAAQS
in Denver and Salt Lake City (Low and High Growth)!a]
With Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Scenario
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-0
0-1
0-0
0-1
0-1
0-1
1
0-1
0-1
0-1
0-1
0-1
0-1
0-0
0-1
0-0
0-1
0-0
0-1
0-1
0-1
2
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-0
0-1
0-0
0-1
0-1
0-1
3
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-0
0-1
0-0
0-1
0-1
0-1
Without
Inspection/Maintenance
1986
1987
1988
1989
1990
. 1993
1995
Scenario
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base
0-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
0-2
0-1
1-2
1
0-2
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
0-1
0-1
0-1
0-1
1-2
2
T)-2
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
0-1
0-1
0-2
0-1
1-2
3
0-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
0-2
0-1
1-2
<
I
N>
[a] Note that a range of values are reported. These reflect two different ratios of HC/NOx ambient ^
concentrations, as discussed in the text. Values from the low HC/NOx ratio are listed first.
-------
Table IV-14
Average Percent Change in Expected Second Highest 8-Hour CO Concentrations
from 1979 Base Year in 6 High-Altitude Cities (low and high growth)[a]
With Inspection/Maintenance
1986
1987
19
88
1989
1990
1993
1995
Scenario
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base
55
48
60
54
64
58
68
61
71
65
76
69
78
70
1
55
48
60
54
65
58
69
62
71
65
77
70
78
71
2
55
49
61
55
66
59
69
63
72
65
77
71
80
73
3
55
48
60
55
64
58
68
61
71
65
76
69
78
70
Without
Inspection/Maintenance
1986
1987
19
88
1989
1990
1993
1995
Scenario
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base
45
37
51
42
55
47
60
51
63
54
70
61
72
62
1
45
37
51
42
55
47
60
51
63
54
68
59
71
60
2
46
38
51
43
57
48
61
53
64
56
71
62
74
65
3
45
37
51
42
55
47
60
51
63
54
70
61
72
62
<
[a] The cities modeled are Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque, and Salt Lake City. ^
m
-------
Table IV-15
Number of Violations of CO NAAQS
in 17 High-Altitude Counties (low and high growth)[a]
With inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Scenario
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base
7
18
2
8
0
3
0
1
0
0
0
0
0
0
1
7
18
2
8
0
3
0
1
0
0
0
0
0
0
2
7
16
2
7
0
3
0
1
0
0
0
0
0
0
3
7
18
2
8
0
3
0
1
0
0
0
0
0
0
Without
Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Scenario
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base
21
43
10
28
3
17
1
9
0
4
0
0
0
0
1
21
43
10
28
3
17
1
9
0
4
0
1
0
0
2
20
43
10
26
3
14
0
8
0
3
0
0
0
0
3
21
43
10
28
3
17
1
9
0
4
0
0
0
0
n
<
I
[a] The cities modeled are Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque, and Salt Lake City.
-------
IV-27
scenario 2 may yield from one to three fewer violations than
the base scenario, with benefits appearing under both low- and
high-growth rates. In 1993 and 1995, no violations are
projected under either the base case or scenario 2. Table
IV-14 indicates, however, that lower CO concentrations could be
expected in these two years under scenario 2.
With I/M, the advantages of scenario 2 over the base
scenario still appear to be present, although they are not as
pronounced. Table IV-14 indicates a general trend of lower CO
concentrations under scenario 2 as compared to the base
scenario. However, from Table IV-15, the number of CO NAAQS
violations under these two scenarios are projected to differ
only in 1986 and 1987 when scenario 2 has 1-2 less violations.
As was mentioned earlier, none of the potential
alternative control scenarios nor the base scenario have
high-altitude NOx standards that are numerically different from
those at low altitudes. This is because NOx emissions
generally decrease with altitude. The rollback analysis was
nonetheless performed for Denver, the only high-altitude city
which is projected to have a NOx problem, to examine the
possible situation whereby some control strategies for HC or CO
could detrimentally affect NOx emissions. The results of this
analysis indicate that there are no significant air quality
differences among the scenarios. For this reason, the results
of the NOx air quality analysis are not presented in this
chapter. (The actual NOx air quality data are included in
Chapter XI.)
In conclusion, without I/M, scenarios 1 and 2 could reduce
ozone concentrations slightly more than the base scenario in
Denver and salt Lake City. With I/M, only scenario 1 appears
to reduce ozone concentrations. In terms of violation of the
ozone NAAQS standard, scenarios 1 and 2 could yield one less
violation in 1989 and 1990 under the high-growth, no-I/M case.
Scenario 1 alone could yield one less violation in 1993 under
the high-growth, no-I/M case. With I/M, scenario 1 is
projected to provide one less violation in 1989 under low
growth. Thus, scenario 1 appears to have a slight overall
advantage with regard to ozone NAAQS violations. Scenario 2,
however, seems to offer by far the most benefit with regard to
CO air quality. It is projected to reduce the second highest
8-hour concentrations by as much as 3 percent and provide the
fewest NAAQS exceedances. No significant NOx air quality
differences are expected among the scenarios.
These air quality projections must be used with a
considerable degree of caution, however. The errors associated
with air quality models can be considerable. One source of
potential error involves the emission factors projected for
future model years which are based on limited data. Also, due
to time and resource constraints many of the input parameters
used in the model, such as growth rates, were national averages
-------
IV-28
and not site-specific values. While ranges of values were used
in many cases to ensure the inclusion of varying local values,
it is still possible that the use of site-specific input data
would result in more optimistic or pessimistic projections.
This is especially true when these improved input data are
coupled with more sophisticated air quality models optimized
for a specific locale. Thus, the air quality projections made
here can serve as a general indicator of when most
high-altitude areas will comply with the NAAQS for CO and
ozone. However, they cannot be used to predict with confidence
when any one, or all, of the areas will comply with the NAAQS.
Unfortunately, the absolute air quality need for further
control beyond that provided by the base scenario depends on
the number of violations of the NAAQS occurring under the base
scenario. For ozone, only one or two violations are projected
in 1986 or beyond under any scenario. At the same time, these
violations persist to beyond 1S95. While additional control
would appear to be needed to bring these one or two areas into
compliance, more detailed analyses are necessary to ensure that
this is the case. At the same time, those areas just complying
with the NAAQS should also be examined in greater detail to
ensure that they will indeed comply in the future. The same
qualifications hold for any conclusions concerning attainment
of the NAAQS for CO. Thus, while further control appears to be
merited at this time, a firm decision cannot be made. Further
air quality analysis will be necessary before this decision can
be made with confidence.
IV. SENSITIVITY ANALYSIS
Of the assumptions made in this analysis, two appear to
have the possibility of changing the conclusions drawn as to
which of the alternative scenarios is most beneficial: 1) the
catastrophic failure rate of feedback systems, and 2) the
impact of low-altitude LDVs at high altitude.
A. Catastrophic Failure Rate
The environmental impact of scenario 1 is greatly
influenced by the assumed catastrophic failure rate for
feedback systems. This failure rate is important in scenario 1
because EPA expects the stringent requirements of meeting the
statutory standards at all altitudes to force manufacturers to
replace nonfeedback systems, currently 31 percent of the
market, with feedback systems. Nonfeedback systems do not
require electronic controls; therefore, they do not exhibit
catastrophic failures that can lead to excessive in-use vehicle
emissions. Because of this, any increase in the use of
feedback systems throughout the nation has the potential to
affect air quality adversely in both low- and high-altitude
areas.
-------
IV-29
The effect of a substantial reduction in the catastrophic
failure rate assumed in this analysis needs to be considered
because the currently projected failure rate was derived from a
very liiaited data base. Only a small number of feedback
systems have currently been tested in the Agency's in-use
surveillance programs and many of the test vehicles in these
programs were equipped with "early" feedback systems which may
be less reliable than future designs. Thus, EPA expects that
as more information becomes available the observed rate of
catastrophic failures will change. The direction of this
change is, of course, unknown at this time, but it is most
likely that the rate would decrease. If this does occur, the
effect of scenario 1 on emissions in areas above 4,00u feet may
be much like the benefits of scenario 2, since both strategies
control emissions at high-altitude to statutory levels.
Benefits may also occur at intermediate altitudes (1,8UU feet
to 4,GU0 feet) that are greater than scenario 3 which controls
emissions to only proportional levels. Therefore, depending on
the assumed catastrophic failure rate of feedback systems,
scenario 1 could have a positive environmental impact instead
of the negative effect that was projected in the air quality
analysis. For example, the CO benefit may be a total reduction
of approximately 350,000 metric tons as opposed to an increase
of approximately 425,000 metric tons.
B. Low-Altitude Vehicles at High Altitude
The recreation popularity of high-altitude areas and the
relatively high population growth rate of high-altitude states
could significantly affect the outcome of this analysis.
People who visit and migrate to these areas from low-altitude
sections of the country often bring low-altitude vehicles with
them. Under fixed-point high-altitude emission standards,
these low-altitude vehicles would pollute more than their
counterparts sold in high-altitude areas since many of them
would not be able to compensate for the lower air density at
higher elevations. (As mentioned in Chapter III, approximately
69 percent of the future low-altitude LDGV fleet under the
fixed-point scenarios, number 2 and the base scenario, are
expected to be equipped with closed-loop emission control
systems which will be able to compensate to some extent for
altitude effects.) This implies that the estimates in this
chapter of future pollution levels under the fixed-point
scenarios are somewhat lower than will actually occur.
Estimates made of future pollution levels under tne
continuous scenarios, numbers 1 and 3, would not be affected by
bringing low-altitude vehicles to high-altitude regions since
they require all vehicles to be able to compensate for chanyes
in altitude automatically. Thus, if enough cars in
high-altitude areas (above 4,000 feet) were in fact from lower
altitudes, the benefits of scenarios 1 and 3 could increase
relative to the base scenario to such an extent that scenario 2
would no longer offer the most emissions reduction potential.
-------
IV-30
One way to examine the impact of low-altitude venicles at
high-altitude in the context of this analysis is to replace a
certain fraction of the high-altitude vehicles expected to be
sold there with low-altitude vehicles. Unfortunately, there
are no studies available which indicate what this percentage
should be. A relatively high estimate would be to assume that
one out of every ten vehicles operated at high altitude is a
low-altitude vehicle. This conservative approach was followed
below.
To compare the effectiveness of scenarios 1, 2, and 3 to
the base scenario in this condition, emission factors must be
determined for low-altitude vehicles in high-altitude areas.
Table IV-16 contains these emission factors which were derived
based on the in-use surveillance proyram mentioned earlier. As
can be seen by comparison to Table IV-1, vehicles not clipped
for high-altitude conditions would be expected to emit
approximately 22 percent more HC and 47 percent more CO than
they would at low-altitude. Since NOx emission rates naturally
decrease with increasing altitude, they are not included in
this sensitivity analysis.
By weighting the emission factors in Tables IV-2 and IV-16
for a fleet comprised of nine high-altitude vehicles to every
one low-altitude vehicle, the average lifetime HC and CO
emission factors for 1964-88 model year LDGVs oecome 1.40 g/mi
and 17.46 y/mi, respectively, for the base scenario. For
scenario 2, this same procedure yields adjusted emission
factors of 1.31 g/mi for HC and lb.30 g/mi for CO. It should
be pointed out that evaporative HC emissions would also rise
uiider scenario 2 from low-altitude cars brought to
high-altitude regions since the canisters used at low-altitude
would not be able to control evaporative emissions to the level
specified by scenario 2. However, this emission increase is
negligible and will not be considered further.
The two sets of adjusted emission factors described in the
previous paragraph, can be multiplied by the same sales and
mileage data discussed earlier in this chapter, to determine
the revised total lifetime emissions from 1984-88 model year
LDGVs above 4,000 feet. According to this uetnouology,
approximately 1,400 more metric tons of HC (113,200 versus
111,800) and 15,000 more metric tons of CO (2,409,000 versus
2,394,000) would be emitted over the lifetime of 1984-88 model
year LDGVs sold above 4,000 feet under the base scenario if one
out of every ten was a low-altitude vehicle. For scenario 2,
2,800 more metric tons of HC (180,800 versus 178,000) and
48,000 more metric tons of CO (2,111,000 versus 2,063,000)
would be emitted by factoring in the impact of low-altitude
vehicles.
The revised emission reductions, relative to the new base
scenario, for scenarios 1, 2, and 3 are listed in Table IV-17
-------
IV-31
Table IV-16
Emission Rates of 1984-88 Model Year Low-
Altitude Vehicles at 5,300 Feet Under
the Base Scenario and Scenario 2[a]
Pollutant
HC CO
Zero-Mile Emission Level (g/mi) 0.55 7.44
Deterioration Rate 0.19 2.21
(g/mi/10,000 miles)
Average Lifetime Emission Level 1.50 18.49
(g/mi)
Emission rate = ZML + (M)(DR)
Where:
ZML = Zero-Mile Level
M = Cumulative Mileage/10/000 (cumulative mileage
50 miles, half the lifetime mileage).
DR = Deterioration Rate
-------
IV-3 2
Table IV-17
Lifetime Emission Reductions for 1984-88 LDGVs Relative
to Base Scenario for Regions Above 4fQ0Q Feet (metric tons)
Scenario
Pollutant
With Low-
Altitude Vehicles
Without Low-
Altitude Vehicles
1
HC
16,600
15,200
CO
-216,000
-231,000
NOX
-1,300
-5,500
Evap HC
4,140
4,140
Total HC
20,740
19,340
2
HC
12,400
13,800
CO
298,000
331,000
NOx
0
0
Evap HC
4,140
4,140
Total HC
16,540
17,940
3
HC
1,400
0
CO
15,000
0
NOx
0
0
Evap HC
0
0
Total HC
1,400
0
-------
IV-33
for areas above 4,000 feet. For comparison, this table also
includes the emission reductions from Table IV-9 which were
generated without accounting for low-altitude vehicles at
high-altitude. As can be seen, the overall merits of scenario
2 still appear to outweigh those of scenarios 1 and 3 since the
potential CO penalty associated with scenario 1 remains
sizeable. The consequences of adding low-altitude vehicles to
high-altitude areas with regard to cost effectiveness will be
addressed in Chapter VI.
V. SUMMARY
In this chapter, the emissions and air quality
characteristics of the three basic control options were
compared to the base scenario (a continuation of the 1982 and
1983 interim high-altitude standards). The analysis found that
scenario 1 would reduce HC and NOx emissions in areas above
1,800 feet. Unfortunately, this scenario also appears to have
the potential for increasing CO emissions in all parts of the
country due to the predicted catastrophic failures of the
requisite emission control systems. Neither scenario 2 nor 3
exhibited such potential problems as both yielded reductions in
HC and CO emissions. under scenario 2, the projected HC
emission reductions at high-altitude were slightly less than
that of scenario 1 but substantially greater than under
scenario 3. The projected CO benefit of scenario 2, over both
scenarios 1 and 3, is significant however. Because of this
effect on CO, scenario 2 appears to be preferable to the other
options. The air quality analysis of selected high-altitude
cities verifies this finding as scenario 2 also appears to be
the course of action which could provide benefits to both CO
and ozone air quality beyond that provided by the base
scenario. The slight HC emissions benefit of scenario 1 over
scenario 2 mentioned above was large enough to yield only one
less violation of the ozone NAAQS for that scenario relative to
scenario 2 both with and without I/M.
Scenario 1 could still be a viable control alternative,
however, in spite of its potential adverse environmental
impact. The adverse impact is totally dependent on the single
assumption regarding the catastrophic failure rate of the
requisite technology. This assumed failure rate is based on
preliminary information and may change as additional data
becomes available. In fact, it is possible that the air
quality impact of scenario 1 may be greater than the combined
benefits of scenario 2 and 3. Therefore, the viability of
scenario 1 remains to be decided in subsequent chapters.
It would appear that additional control beyond that
provided by the base scenario (e.g., scenario 2) is merited,
since violations of the NAAQS for ozone could persist to at
least 1995. Violations of the NAAQS for CO could persist as
far in the future as 1993-95 or may disappear as early as 1988,
depending on external factors such as VMT growth and the
-------
IV-34
presence or absence of local I/M programs. However, as
discussed earlier, these projections must be used with caution
because of the many known sources of potential error. Before
any firm conclusions can be drawn concerning the absolute need
for additional high-altitude control beyond that provided by
the base scenario, more detailed air quality analysis of those
areas just in and out of compliance with the NAA^S will be
necessary.
-------
IV-35
References
1. "Environmental and Economic Impact Statement for the
1982 and 1983 Model Year High-Altitude Motor Vehicle Emission
Standards," U.S. EPA, OANR, OMS, ECTD, SDSB, October 1980.
2. "Average Lifetime Periods for Light-Duty Trucks and
Heavy-Duty Vehicles," U.S. EPA, OANR, OMS, ECTD, SDSB, Glenn W.
Passavant, November 1979.
3. "User's Guide to MOBILE 2," U.S. EPA, OANR, OMS,
ECTD, SDSB, February 1981.
4. Code of Federal Regulations, Title 40, Part 86.
5. Data Resources, Long-Term Review, Fall 1980.
6. "Standard for Emission of Particulate Regulation for
Diesel-Fueled Light-Duty Vehicles and Light-Duty Trucks," 45 FR
14496, March 5, 1980.
7. Rand McNally Road Atlas, 1980.
8. Code of Federal Regulations, Title 40, Part 86,
Subpart B.
9. "Methodology to Conduct Air quality Assessments of
National Mobile Source Emission Control Strategies - Final
Report," U.S. EPA, OANR, OMS, ECTD, EPA-450/4-80-026, October
1980.
10. Modified Rollback Computer Program, Interim User's
Manual, Draft, U.S. EPA, OAQPS, April 1979.
11. "Air Quality Analysis for the 1984 High-Altitude
Report to Congress," U.S. EPA, OANR, OMS, ECTD, SDSB, March
1982.
12. "Regulatory Analysis for Advanced Notice of Proposeu
Rulemaking on NOx Emissions from 1986 and Later Model Years
Heavy-Duty Engines and Light-Duty Trucks," U.S. EPA, OANR, OMS,
ECTD, SDSB, January 1981.
13. Modeling Assumptions for Ozone Attainment Status
Projections, Memorandum to File, Freas, W., U.S. EPA, OAwPS,
December 26, 1981.
-------
Chapter V
Economic Impact for
Gasoline-Fueled Light-Duty Vehicles
I. INTRODUCTION
This chapter will examine the incremental costs of
complying with each of the alternative high-altitude regulatory
scenarios outlined in Chapter II relative to complying with the
base scenario. This will be done by evaluating the cost to
manufacturers and the cost to consumers. Manufacturers'
primary expenses will involve the variable costs of adding
emission control hardware to their vehicles, the fixed costs of
new vehicle certification, and the development of special
high-altitude calibrations. The consumer will pay for the
expenses incurred by the manufacturer and, in addition, pay for
a profit that the manufacturer must make on his investment.
Consumers will also bear the cost or savings of any changes in
vehicle maintenance or fuel economy.
Following these two sections, the aggregate cost to the
nation for the first five years that high-altitude standards
are in effect will be determined. After this, the
socioeconomic impact on high-altitude areas and the nation of
each alternative control scenario will be discussed.
II. COSTS TO MANUFACTURERS
The costs of high-altitude standards to manufacturers can
be conveniently separated into two types: variable and fixed.
The variable costs, which are essentially the costs of emission
control hardware, will be analyzed first. This cost will be
determined on a per vehicle basis in terms of the retail price
equivalent (i.e., the change in the purchase price of the
vehicle). The fixed costs will be analyzed next. These costs
will be determined for the entire vehicle fleet and then
converted to a per vehicle basis. The fixed costs are examined
separately because they represent the capital investment
manufacturers must make prior to the actual implementation of
the standards. These fixed costs will include research and
development (R&D), equipment for manufacturing control
hardware, development of high-altitude calibrations,
certification, and test facility additions or modifications.
A. Emission Control System Costs
In this section, the retail price equivalent (RPE) of the
emission control hardware required by each control scenario
will be determined. First, the methodology used in the
analysis will be presented and discussed. Second, the cost of
each emission control component will be estimated using this
methodology. Finally, the total control hardware cost for each
-------
V-2
scenario will be summarized. The control technology upon which
these cost estimates are based was identified in Chapter III.
1. Cost Methodology
In general, the RPE of emission control hardware includes
the direct material, direct labor, fixed and variable overhead
and profit at the vendor level, tooling expense, land and
building expense, and overhead and profit at the corporate and
dealer level. In this analysis, R&D is not included in the
emission control hardware costs because it is not considered to
be a variable cost. R&D costs will be estimated separately
under "Fixed Costs." With this one exception, the RPE and
estimates used in this chapter will follow the methodology used
in recent regulatory analyses,[1,2] and will not be discussed
in detail here. These regulatory analyses consider corporate
overhead and profit, in addition to dealer overhead and profit
(29 percent of the vendor level costs) since the manufacturers
must receive a return on their investment. For the most part,
estimates of vendor-level costs will be taken from
Lindgren.[3,4]
In addition, the cost of each component depends on its
production volume. For the continuous statutory or continuous
proportional scenarios (numbers 1 and 3), all vehicles must be
certified for use at high altitude unless they are exempted
from sale at high altitude. Thus, control hardware will be
installed on nearly all vehicles nationwide unless the
standards can be met without additional altitude-compensating
equipment. For fixed-point scenarios (the base scenario and
number 2), only those vehicles sold at high altitude will
require control devices (unless the vehicle can meet
high-altitude standards with its normal low-altitude
configuration). This amounts to only about 3 percent of all
vehicles certified as was discussed in Chapter IV. Thus, the
production volumes vary markedly between scenarios and hardware
costs will have to reflect this variation.
In the estimates which follow, the range of costs
determined for scenarios 1 and 3 is large enough to account for
variations in production volumes. For scenario 2 and the base
scenario, approximately 3.1 percent of vehicles are affected by
a high-altitude standard. As discussed in Chapter IV,
approximately 1.38 million non-California light-duty
gasoline-fueled vehicles will be sold between 1984-88, and 3.1
percent of this total is approximately 280,000 vehicles
annually. This will affect the costs of fixed-step aneroids.
The cost for this device has been estimated by Lindgren based
on a production volume of 1 million units per year. In
scenario 2 and the base scenario, it is estimated that 31
percent of all high-altitude vehicles will require aneroids,
corresponding to a volume of 87,000 units per ^ear. It is
assumed that these aneroids will be produced by, at most, two
-------
V-3
outside suppliers with an annual production of 38,000 units for
each supplier. Therefore, the aneroid cost estimate from
Lindgren must be adjusted to reflect the lower production
volumes. If a 12 percent learning curve is used, as in past
regulatory analyses,[2,3] an economy-of-scale factor is
calculated to be 1.8. This economy-of-scale factor should be a
conservatively high estimate, when considering that at most two
suppliers would manufacture these aneroids. In scenario 2
only, it is estimated that 10 percent of all high-altitude
vehicles will require the addition of a MAP sensor. Unlike
aneroids, however, no adjustment to Lindgren's cost estimate is
necessary for the MAP sensor in this scenario. These sensors
already will have attained the proper economy-of-scale since
they are assumed to be used in throttle body injection (TBI)
systems on 59 percent of the fleet even in the absence of
high-altitude standards (this is further discussed in Chapter
III). Thus, only the costs of aneroids will be adjusted to the
lower production volumes associated with scenario 2 and the
base case.
In this analysis, the number of suppliers for emission
control development is also of particular concern for
estimating the total R&D cost for expanding the capability of
existing electronic control units. The total number of
suppliers is estimated to range from two to four, depending on
the number of vehicles affected. In this case, these suppliers
may be the larger vehicle manufacturers or outside suppliers,
and for purposes of this analysis, it is assumed that half of
the total suppliers are the vehicle manufacturers themselves.
This information will be used for estimating the capital costs
to manufacturers in a later section.
All costs in this analysis will be estimated in 1981
dollars. Nearly all cost estimates are taken from Lindgren[4,5]
where costs are quoted in 1977 dollars. As in past regulatory
analyses,[1,2,3] an 8 percent per annum inflation rate will be
used to convert control hardware costs from 1977 dollars to
1981 dollars. This inflation rate can be supported by the fact
that the new car price index (NCPI) for the years 1977, 1978,
1979, and 1980 was 7.2, 6.2, 7.4, and 7.5 percent,
respectively. While the NCPI is much lower than the Consumer
Price Index for the past 3-4 years, it is a better indicator of
the specific inflation rate for vehicle manufacturing. The
NCPI may reflect some lowering of profits to sell cars in the
last few years. However, the 8 percent inflation rate provides
some degree of compensation for the effect of such practices.
2. Estimated Cost for Each Component
This section analyzes the cost of each emission control
component that is expected to be used to meet high-altitude
standards under the various scenarios. The cost estimates are
presented as RPE, and in most cases a cost range is given due
-------
V-4
to variations in the production volume, material, and design
that could occur with each scenario.
The following emission control items or engine
modifications will be examined in this section: electronic
load control system (ELCS), fixed-step aneroid, continuous
aneroid, turbocharger, feedback control system, expansion of
existing electronic functions, modification of the electronic
control module, and enlargement of the charcoal canister for
the storage of evaporative emissions. A summary of these costs
are shown in Table V-l.
a. Electronic Load Control System (ELCS). The ELCS was
developed by GM and a complete description of the exact
hardware included has never been given. However, from the
preliminary description that is available, EPA estimates the
primary components should include a manifold absolute pressure
(MAP) sensor, an air temperature sensor, and an expanded
capability of the microprocessor memory. The MAP sensor is
currently used by GM in some of its computer command control
systems and is expected to be used on over half of the vehicles
produced in the future. It is estimated that the sensor's
production cost is about $15 based on data in an unpublished
report by DOT.[6] The air temperature sensor will also be used
in over half of the vehicles. The production cost of this
device is about $1.[5] Assuming about $5 to install both the
MAP and temperature sensors, and $6 for corporate overhead and
profit, the total hardware cost for these two sensors is about
$27. Expanding the adaptive memory involves additional
development of the electronic control unit (microprocessor).
This cost is due mostly to research and development and is
discussed separately under "Fixed Costs" later in this
chapter. Thus, without the cost of expanding the adaptive
memory, the RPE for the ELCS is about $27.
In conjunction with the ELCS, it may be necessary to
incorporate idle speed adjustment for acceptable vehicle
operation, especially for vehicles expected to meet the same
standard at high altitude as well as low altitude. In Chapter
III it was estimated that only those vehicles under the
continuous statutory scenario applicable to 10,200 feet
(scenario la) would require idle speed control. Little data
are available on the cost of this item but one manufacturer has
indicated that this cost is similar to that of an aneroid. As
discussed below, an aneroid costs between $7-9? therefore, this
cost applies to the idle speed control also.
b. Fixed-step aneroid. In Chapter IV, Technology
Assessment, it was projected that a fixed-step aneroid would be
used under fixed-point scenarios requiring that compliance be
demonstrated at a single high-altitude location (2 and the base
scenario). Under these scenarios, only vehicles sold at high
altitude will require modification. Thus, fixed-step aneroids
-------
V-5
Table V-l
Vehicle Hardware Costs (RPE, $1981)
Control Technique cost
Electronic Load Control System $27
Idle Speed Control 7-9
Fixed-Step Aneroid 7-9
Continuous Aneroid 7-9
Turbocharyer 253
Feedback Control 190-243
Expand Function of Existing Electronics
A) Expand Microprocessor Capability 0
3) Add MAP Sensor 26
Change Electronic Modules
A) Recalibrate 0
B) Add MAP for Fuel, Spark, and EGR 26
Enlarged Charcoal Canister 2-3
A) Statutory 10,200 feet scenarios 6
B) Statutory 6,000 feet scenarios 2-3
-------
V-6
will be manufactured at a low production volume which will
result in a relatively higher unit cost.
Fixed-step aneroids are currently used on some car and
truck models to control carburetion, and most other models
could easily be similarly modified. In addition, for some
scenarios aneroids may be used to improve the functions of
power enrichment and EGR. A fixed-step aneroid has been
estimated to cost $4-5 for a production volume of about 1
million per year.[5] At most, about 3 percent of the national
fleet could potentially use fixed-step aneroids (to be
developed by, at most, two suppliers). As discussed earlier,
an economy of scale factor of 1.8 should be a conservatively
high estimate for these low production volumes. At this
production volume an aneroid would cost $7-9.
c. Continuous aneroid. The continuous aneroid is more
likely to be used for scenarios with continuous standards
(numbers 1 and 3) and thus would be incorporated into vehicles
certified and.sold at all altitudes. There is currently no cost
estimate available for continuous aneroids, but the major
differences between this type of aneroid and that of the
fixed-step aneroid is the added cost of a calibration needle
able to make continuous adjustments and the decrease in cost
due to larger anticipated production volumes. These effects
would bring the cost of a continuous aneroid close to that of a
fixed-step aneroid, which was estimated with a conservatively
high economy-of-scale factor. Thus, a cost of $7-S is used
here also.
d. Turbochargers. The cost of turbocharging has been
estimated by EPA for diesel-fueled light-duty vehicles,[7] and
these costs should apply to gasoline-fueled vehicles as well.
From these previous EPA estimates, the cost of turbocharging a
4- and 6-cylinder engine (in 1979 dollars) was $207 and $226.
In 1981 dollars these costs are $241 and $264, respectively.
Included in the estimate is the turbocharger itself, oil lines
and other plumbing, and manifold and exhaust transition
hardware. The cost of turbocharging an S-cylinder engine is
not estimated because very few, if any of these engines are
expected to be produced after 1984. Since the future market
shares of 4- and 6-cylinder engines is unknown, they are
assumed to be produced in equal numbers. Therefore, the
average cost of a turbocharger is $253.
e. Feedback control. The cost of an electronic
feedback control system is difficult to estimate and EPA is not
aware of any documented cost analysis. However, an attempt has
been made by DOT in an unpublished report to determine the cost
of a feedback control system,[6] and these estimates will be
used in this analysis. For 1980 production levels, these costs
come to $190-210 (1981 dollars). This cost estimate includes
the following components: electronic control unit (ECU), ECU
-------
V-7
mounting bracket and screws, closed-loop wiring harness and
straps, Zr03 oxygen sensor, MAP sensor, throttle position
sensor, electromechanical carburetor solenoid, high energy
ignition with electronic spark control, advance and retard
solenoid for spark control revisions, tubing, valves and
mounting brackets, detonation sensor, vacuum switch and idle
deceleration unit, cold start switch, maintenance indicator
lights, closed-loop carburetor, vortex mixer, a proportional
EGR system, intake manifold revisions, and exhaust manifold
revisions. The cost of the feedback control system also
includes credits for the standard carburetor and backpressure
EGR system.
Only vehicle manufacturers who presently produce vehicles
with nonfeedback systems may need to convert their vehicles to
feedback control. These manufacturers include Ford (assumed to
have a 20 percent market share) and some foreign manufacturers
(assumed to have an 11 percent market share). While
certification data show that all of Ford's open-loop vehicles
are equipped with three-way catalysts,[8] some foreign
manufacturers do not currently use them. These latter vehicles
will probably use three-way catalysts when converting to
feedback control. The cost increment of a three-way catalyst is
about $95,[4] so the total cost of feedback control for foreign
manufacturers is $190-305 per vehicle. Based on the assumed
market shares for the manufacturers, the average cost of
converting from nonfeedback to feedback is $190-243.
f. Expanding functions of existing electronics.
Expanding the functions of existing electronics consists of
primarily augmenting the electronic memory (microprocessor)
capability. For some feedback carburetor equipped vehicles, a
MAP sensor will also be added. Expanding the microprocessor
capability is primarily an R&D cost, and this will be discussed
in detail later in the section on fixed costs. As was
estimated for the ELCS above, the MAP sensor costs about $15.
If $5 for installation is assumed along with $6 for corporate
profit and overhead, the total hardware cost of the MAP sensor
is $26.
g. Changing electronic control modules. Modifying the
electronic module involves either: 1) replacing an existing
module with a differently calibrated module to control fuel
metering, spark timing and EGR, or 2) the addition of a MAP
sensor to automatically adjust the fuel metering, spark timing,
and EGR control. There is essentially no variable cost
associated with replacing the existing control module since no
additional hardware is required beyond that already present in
vehicles. An R&D cost will be involved, and this is discussed
below under "Fixed Costs" for engine recalibration. The cost
of the MAP sensor has already been estimated above at $26.
-------
V-8
h. Charcoal canister for evaporative emissions. A
charcoal canister is necessary for the storage of evaporative
emissions. Due to the decrease in air density at high
altitude, a larger canister is expected to be required. Thus,
a larger plastic container and an increased amount of charcoal
are necessary for high-altitude evaporative emission control.
The cost increase is then basically an increase in the cost of
additional charcoal. One estimate shows that a 50 percent
increase in carbon bed costs about $6.[9]
For the continuous statutory standard applicable to 10/200
feet, or scenario la, it is estimated that approximately 50
percent more carbon will be necessary for storage of
evaporative emissions. The cost in this case is $6. For the
continuous statutory standards applicable to 6,000 feet, or
scenarios lb and lc, or scenario 2 and for the fixed-point
statutory standard at 5,200 feet, approximately 25 percent
additional charcoal will be required to meet evaporative
emission standards. This will cost roughly $2-3. For
proportional standards, or scenarios 3a, 3b, and the base
scenario, no additional evaporative emission control should be
necessary, and thus no costs would be incurred.
3. Control Hardware Cost for Each Scenario
In the technology section of this report, EPA projected
the fraction of gasoline-fueled light-duty vehicles expected to
require each of the above-mentioned control hardware components
or modifications. For convenience, this information is
duplicated in Tables V-2 through V-4 for the seven scenarios
being analyzed. With this information, the average hardware
cost for the vehicles affected by high-altitude standards in
each scenario is readily calculated by multiplying the
appropriate market shares (Tables V-2 through Table V-4),
together with the cost of the respective emission control
components as detailed in the preceding section of this chapter
(section (B)(b)(1-9)), and then adding these results. Rather
than present a detailed description of the calculations for
each scenario, an example will be provided below for the base
scenario which is a continuation of the 1982 and 1983 interim
high-altitude standards.
In the base case, only vehicles sold at high altitude need
to comply with high-altitude standards. The effect of this
two-car strategy is that only about 3 percent of all vehicles
sold nationwide (excluding California vehicles) will be
required to be equipped with high-altitude emission control
hardware.
Table V-2 shows that 31 percent of the vehicles sold at
high-altitude dealerships should require a fixed-step aneroid
for the carburetor. Also, 2 percent are estimated to require a
recalibrated electronic control module and 13 percent should
-------
Table V-2
Estimated Exhaust Emission Control Requirements
for Scenarios witn Exemptions
Control
Hardware
Nonfeedback Vehicles
Fixed-Step Aneroid:
A. Carburetor
B. Power Enrichment
C. EGR or air
injection rate
Continuous Aneroid:
A. Carburetor
B. Spark
Feedback Vehicles [b]
Feedback Control w/MAP
(Three-Way Ford and
Foreign Market share)
Expand Function of
Existing Electronic
A. Expand Capability
B. Add WAP Sensor
Scenario lb:
Cont inuous
Statutory,
6,000 Feet
W/Exemptions
N/A[a]
N/A
Scenario 2:
Fixed-Point
Statutory,
5/200 Feet
W/Exempt ions
31% A,B,C
N/A
31%
59% A (TBI)
10% A,B (FBC)
N/A
5 9% A (TBI)
Scenario Ja:
Continuous
Proportiowal,
O/U0U Feet
W/Exemptions
N/A
Base Scenario:
Fixed-POxiit
Proportional,
5,200 Feet
W/Exemptions
31% A
31% A,3
N/A
N/A
N/
13% A (TBI) JL3% A (TBxJ
2% A (FBC)
-------
Table V-2 (cont'd)
Estimated Exhaust Emission Control Requirements
for Scenarios with Exemptions
Control
Hardware
Change Electronic
Modules for FBC
A. Recalibrate Fuel
Meter iny
B. Recalibrate Fuel
Metering, Spark,
EGR Plus Add MAP
Fuel
Scenario lb:
Continuous
Statutory,
o/llOu Feet
W/Exemptions
N/A
anu
for
Scenario 2:
Fixed-Point
Statutory,
5,200 Feet
W/Exemptions
1U% B (FBC)
Scenario 3a:
Cont liiUoUS
Proportional,
6,uUU Feet
WExempt ions
N/A
Base Scenario:
Fixeu-Pomt
Proportional,
i>,2Ul) Feet
W/ Exempt ions
2% A (FbC)
No Chanye FBC
No Chanye TBI
lJ/A
N/A
4 6%
8%
46*
ta] Not applicable.
[b] FBC means feedback carburetor system.
TBI means throttle body injection system.
-------
Table V-3
Control Hard-
ware Before
Modifica-
tion[a]
Estimated Exhaust Emission Control Requirements
for Scenarios Without Exemptions
Control Hard-
ware After
Modifica-
tions[a]
Scenario la:
Continuous
Statutory,
10,200 Feet
W/0 Exemptions
Scenario lc:
Continuous
Statutory,
6 ,000 Feet
W/Q Exemptions
Scenario 3b:
Continuous
Proportional,
6,000 Feet
W/O Exemptions
OL
OL W/Aneroid
0%
0%
23%
OL
Turbocharyed
18%
5%
U%
OL
FBC
13%
26%
b%
FBC
ELCS [b]
21%
20%
10%
FBC
Turbocharyed
2%
0%
0%
FBC
No Change
0%
0%
tf%
TBI
Expansion
39%
59%
j.3%
TBI
Turbocharyed
2U%
U-6
D £
TBI
No Chanye
0%
0%
46%
I a J OL means nonfeedback or open-loop system.
FBC means feedback carburetor system.
ELCS means electronic load control system.
TBI means throttle body injection system.
Expansion means the capability of the existing electronic components is
upgraded.
[b] The feedback carbureted systems which chanye to electronic control systems
include a portion of open-loop systems that have switched to teedback
systems. Therefore, the percentage listed for each scenario does not add to
100.
-------
V-12
Table V-4
Estimated Evaporative Emission Control
Technology Requirements for all Scenarios
No Change to 25% Increase in 50% Increase in
Scenario Carbon Canister Carbon Canister Carbon Canister
la
N/A [a]
N/A
100%
lb
N/A
N/A
N/A
lc
N/A
N/A
N/A
2
N/A
100%
N/A
3a
100%
N/A
N/A
3b
100%
N/A
N/A
Base
100%
N/A
N/A
[a 1 Not applicable.
-------
V-13
require an expansion of their existing electronic controls
primarily for fuel metering. Since these are R&D costs,
however, they are not included here but are discussed later
under "Fixed Costs." The remaining 54 percent should require
no new exhaust emission control hardware or modifications.
Under this scenario, no high-altitude vehicle requires
additional charcoal and larger canister for evaporative HC
controls (Table III—4). With costs taken from Table V-l, the
sales-weighted average cost is $2-3. Again, this cost applies
to only the 3 percent of all nationwide vehicles certified for
high-altitude sale.
The cost increase for each high-altitude vehicle can also
be stated as an average for all vehicles sold in the nation by
amortising the costs over the entire fleet (i.e., both low- and
high-altitude sales) . Stated in this way, the average cost
would be less than 10 cents per vehicle or essentially $0 per
vehicle.
In this analysis the incremental cost of each alternative
scenario is important. This cost is found by calculating the
total average cost of each alternative scenario, as previously
explained, and then including a credit for the costs of the
control hardware associated with the interim standards (base
scenario), if this hardware is no longer needed. Therefore, a
credit of $2-3 is applied toward scenario 2 (two-car strategy),
while no credit ($0) is applied toward the remaining scenarios
(one-car strategies).
It should be noted that for scenarios with exemptions, the
exempted vehicles must be included in the estimates when the
incremental cost is expressed as an average for all vehicles
sold in the nation. In scenario 2, as in the base scenario
example, this is accomplished by amortizing the high-altitude
costs over the entire national fleet (i.e., both low- and
high-altitude sales) . In scenarios lb and 3a it is easily
accomplished by reducing the average sales price by the
respective percentage of exemptions for each scenario.
The average incremental hardware costs for the alternative
scenarios are presented in Table V-5.
B. Fixed Costs
The fixed (or capital) costs of high-altitude control are
examined in this section. These costs include R&D for
expanding the electronic capability, developing engine
calibrations, certification, selective enforcement auditing,
and test facility additions or modifications.
-------
V-14
Table V-5
Incremental Control Hardware Costs for
Each Scenario ($1981)[a]
Avera9e Cost, Average Cost,
Scenario High-Altitude Fleet[b] National Fleet[c]
la 140-148
lb 48-61
lc 67-81
2 8-9 1 [d]
3a 4-5
3b 21-26
[a] Reflects costs of control over the base case scenario,
[t] Applies only to vehicles sold at high altitude in
fixed-point scenarios. Assumes manufacturer has amortized
costs over those vehicles affected (i.e., vehicles sold at
high altitude).
[c] Applies to vehicles sold nationwide. Assumes
manufacturers will recover cost over entire national
fleet, and not only on vehicles affected by a
high-altitude standard.
[d] Although the average cost of scenario 2 on a fleet basis
is less than 50 cents, a value of $1 will be assumed to be
conservative.
-------
V-15
1. Research and Development Costs for Expanding
Electronic Capability
Expanding the capability of electronic functions is
primarily an R&D expense. The total R&D expense will depend
first on the number of suppliers and second on the expense
incurred by each supplier. The total number of suppliers are
estimated to be two or four, depending on whether only
high-altitude vehicles or all vehicles are affected. These
suppliers may be the larger vehicle manufacturers or outside
suppliers. In this analysis, it will be assumed that half of
the total suppliers are manufacturers while the other half are
outside suppliers.
R&D costs for the expansion of the microprocessor's
capability should be much less than that for similar
technologies such as the electronic control unit (ECU), which
Lindgren estimates to be about $2.7 million in 1981
dollars. [4] This is because much of the work on electronic
capability should have been performed by 1984, so that any
futher development of the microprocessoring unit should cause
only a small increase in R&D expenses. A best estimate here is
that a total expense of $500,000 or less will be incurred by
each outside supplier. It will be assumed that this R&D cost
will be recovered during the 5-year period being analyzed, or
1984-89. The total R&D cost for all suppliers is then
estimated to be $1-2 million, depending on the control scenario.
Expanding the memory capability is an integral part of
both the ELCS and upgrading the functional capacity of an
existing electronic-feedback system. All continuous scenarios
are projected to require either ELCS or increased
electronic-control capacity on at least some vehicles. The
percentage of vehicles expected to use either technique is
presented in Tables V-2 and V-3 for each scenario.
Since continuous scenarios require that all vehicles sold
nationwide must meet the applicable standards it is assumed
that manufacturers will recover their R&D costs through
nationwide sales. The average annual nationwide sales has been
discussed in the Environmental Impact chapter of this report
and these sales projections are repeated in Table V-6. Between
the years 1984-88, or the first five years after the
high-altitude standard would become effective for light-duty
vehicles, the average annual nationwide sales of
gasoline-fueled light-duty vehicles would be about 9 million.
Over the 5-year period, total sales would amount to about 45
million vehicles. This estimated sales volume is used for
predicting amortized costs later in section "g."
As shown in Tables V-2 anci V-3, the number of vehicles
requiring expanded memory capability for ELCS or for upgrading
the electronic control system varies according to each
-------
V-16
Table V-6
Year-by-Year Projections of the Sales of Gasoline-Fueled
Light-Duty Vehicles (reduced by 10 percent to
eliminate California vehicles)
Model Year
1984
1985
1986
1987
1988
LDV Sales
(millions)
9.41
9.26
8.90
8.56
8 .45
High-Altitude
LDV Sales
(millions)[a]
0 .29
0.28
0 .28
0.26
0.26
[a] Estimated to be 3.1 percent of all vehicles in each
vehicle group.
-------
V-17
scenario. In scenario la, approximately 60 percent of all
vehicles will require an expansion of the electronic
microprocessor either for the ELCS or the existing throttle
body injection (TBI) system, and EPA estimates that a total of
four suppliers will perform the necessary R&D at a total cost
of $2 million. In scenario lb and lc, approximately 70-80
percent of all vehicles will require R&D for expanded
electronics, and again total R&D costs is projected to be $2
million. In scenario 3a, only 15 percent of vehicles certified
for high-altitude sale will require expanded electronics, and
this should only require two suppliers and $1 million of R&D.
In scenario 3b, approximately 23 percent of vehicles will
require R&D, and EPA estimates this work will be done by three
suppliers for a total cost of $1.5 million.
The R&D cost to manufacturers (which is assumed to be half
of the total number of suppliers) is shown for each scenario in
Table V-7. Included in this table is the cost of capital, which
is assumed to be 15 percent.
2. Development Costs for Recalibrating Existing Hardware
A detailed analysis of development costs for high-altitude
engine recalibration can be found in Appendix III of this
report. A summary of development costs is shown in Table V-8
for families with unique high-altitude calibrations.
Development costs for scenarios other than the base scenario
are estimated to be $3.2-8.2 million.
3. Certification Costs
This analysis assumes that a full-certification program is
in effect for motor vehicles sold at higher elevations. Such a
program requires that actual vehicular emission tests be
conducted under high-altitude conditions. The high-altitude
certification requirements could incorporate a less rigorous
procedure by allowing engineering evaluations by the
manufacturers to be substituted for actual test aata. Such
"self-certification" requirements are now being used in
conjunction with the 1982 and 1983 interim high-altitude
standards in response to President Reagan's regulatory relief
initiatives. This type of program would be somewhat less
costly and less time consuming than the program assumed in this
analysis. Nevertheless, the more rigorous certification
requirements are evaluated in this document to ensure that the
certification costs were not understated.
A detailed analysis of certification costs can be found in
Appendix III. A summary of the certification costs,
incremental to the base scenario, is shown in Table V-9. The
incremental costs are $0-18,773,000 for the first year, and
$0-14,000 for each year after. In addition to these costs,
there would be a 15 percent cost of capital. The large costs
-------
V-18
Table V-7
R&D Costs for Expanding Memory
Capability ($1981)[a]
Scenario
R&D
R&D with
Cost of Capital
la
1,000,000
1,150,000
lb
1,000,000
1,150,000
lc
1,000,000
1,150,000
2
0
0
3a
500,000
575,000
3b
750,000
862,000
T It is assumed outside suppliers will incur costs equal
that shown for each scenario in this table.
-------
V-19
Table V-8
High-Altitude Development Costs ($1981)
Scenario
Recalibrated
Engine Families
Total
Tests[a]
Costs L b J
With Cost
of Capital[c]
la
109
16/350
8,175,000
9,401,000
lb
82
12/300
6,150,U00
7 ,073 ,000
lc
109
16/350
8 ,175,OUU
9,401,000
2
93
13,450
6,975,00U
8,021,000
3a
45
6,750
3,375,000
3,881,000
3b
50
7,500
3,750,000
4,313,000
[a] 150 FTP tests for gasoline vehicles.
[b] $500 per FTP development test.
[c] Includes 15 percent cost of capital.
-------
V-2U
Table V-9
Incremental Certification Costs ($1981)[a]
Annual
Cert. Cost
Scenar io
Cert. Cost
First Year
(1984 models)
With Cost
of Capital[b]
Subsequent to
First Year
(1985-88 models)
With Cost
of Capital[b]
la
18,773/000
21,589,000
14,000
16,000
lb
6,104,000
7,020,000
0 [c]
0
lc
12,624,000
14,518,000
14,000
16,000
2
0
0
0
0
3a
0
0
0 t c ]
0
3b
3,107,000
3,573,000
14,000
16,000
[a] Reflects certification costs over base scenario.
[b] Includes 15 percent cost of capital.
[cj Incremental cost is calculated to be actually less than
$0, or a potential cost savings over the base scenario.
-------
V-21
shown in Table V-9 for the first year are due to the fact that
the continuous scenarios (numbers 1 and 3) will require the
recertification of many or most low- and high-altitude vehicles
in 1984. These incremental certification costs will be carried
out through the remainder of this report.
The promulgation of a high-altitude standard in 1984
should not affect the number of Selective Enforcement Auditing
(SEA) tests performed on LDVs. High-altitude vehicles count
toward a manufacturer's annual quota and are merely substituted
for low-altitude audits. Manufacturers may experience a slight
cost increase due to transporting vehicles to SEA sites.
However/ these costs should be negligible and shoula not
adversely affect manufacturers of high-altitude vehicles.
4. Test Facility Modifications
There is difficulty in forecasting a manufacturer's needs
for test facilities. The possibility that new facilities are
needed should not be ruled out, especially considering that it
will be more difficult to comply with some scenarios than
others. For scenarios la, lb, and lc it is estimated that one
new facility may have to be built for each large manufacturer
(assumed to be GM, Ford, Chrysler, AMC, and Toyota in this
analysis). The cost of building a new high-altitude testing
chamber is approximately $2-4 million (assuming that a building
already exists for the testing chamber).[10] An average of $3
million will be used here. The additional equipment, which
includes the dynamometer, a CVS system, analyzers, software and
computer hookup, and other emission related test equipment,
costs about $1 million. The total cost for a fully-equipped
facility is estimated to be about $4 million. The total maximum
cost for all five large manufacturers is then $20 million for
scenarios la, lb, and lc.
For scenarios 2, 3a, 3b, and the base case, the
promulgation of a 1984 high-altitude standard should not
require manufacturers to purchase new equipment for
constructing new test facilities or for modifying existing
emission test cells. The existing facilities which
manufacturers use to measure emissions for the interim
standards should be adequate for any future standards. For
scenarios 3a, ana 3b, where emission standards apply
continuously between low and high elevations (i.e., between
1,800 feet and 6,000 feet), it is assumed that good engineering
judgment can be used to estimate emission results. Thus, a
cost of $0 will be used for these scenarios.
The test facility costs are shown in Table V-10. As with
the R&D costs above, a 15 percent cost of capital is used.
-------
V-22
Table V-10
Test Facility Costs ($1981)
Test With Costs
Scenario Facility Costs of Capital[aJ
la 20f000,000 23,000,000
lb 20,000,000 23,000,000
lc 20,000,000 23,000,000
2 0 0
3a 0 0
3b 0 0
[aj Includes 15 percent cost of capital.
-------
V-23
5. Summary of Capital Costs to Manufacturers
The total capital costs to manufacturers consist of the
"development, certification, and R&D efforts required by a
change in standards and any new test facilities required.
Other capital investments should be incurred by outside
suppliers. A summary of the manufacturer's fixed or capital
costs are shown in Table V-ll.
6. Amortized Cost of Capital
To estimate the development and certification, R&D for
expansion of memory capability, and test facility modification
costs on a per vehicle basis, first the production volume of
gasoline-fueled vehicles not sold in California must be
projected for the first five years after a 1984 high-altitude
regulation is promulgated. These sales projections are shown
in Table V-6 for the years 1984-88. Next, the development and
certification, R&D, and test facility costs previously-
determined must be amortized over these production volumes.
Certification and development should take place one year before
the first year these standards would become effective, or in
1983. After that year, only certification costs will occur,
again one year prior to the actual year each vehicle is sold.
The final cost over the five-year period would then be
calculated at the present value when the regulation first takes
effect, or 1984 for LDVs, using a 10 percent discount rate for
each year's fixed cost. This cost is then amortized over
1984-88 production and is weighted to result in an equal cost
per vehicle over the years of production cited. Expenses are
assumed to occur on January 1 of the given year and revenues
are assumed to be received on December 31 of the given year.
The total development and certification costs ana the
costs per vehicle are shown in Table V-12. Manufacturers may
choose to recover their costs over the 3 percent of total
vehicles sold at high altitude (for scenario 2 only), or they
may choose to recover costs over the entire national fleet..
The results of these costs for the different amortization
strategies are also shown in Table V-12. The fixed costs per
vehicle should be about $1 for scenario la, lb, and lc,- and
much less than $1 for scenarios 3a ana 3b. For scenario 2,
fixed costs amount to $6 per vehicle if amortized over vehicles
sold at high altitude only.
III. COSTS TO USERS
The total user cost of the various alternative control
scenarios includes changes in purchase price, maintenance, and
fuel economy.
-------
Table V-ll
Total Capital Costs to Vehicle Manufacturers ($1981)[aj
Scenar io
Development
Certification
Test
Facility
R&D[b]
Total
la
9,401,0 00
21,653,000 .
23,000,000
1,150,000
55,204,000
lb
7,073,000
7,020,000
23,000,000
1,150,000
3b, 243,000
lc
9,401,000
14,582,000
23,000,000
1,150,000
48,133,000
2
ii,021 ,000
0
0
0
8,021,000
3a
3,831,000
0
0
b75 ,000
4,456,000
3b
4,313,000
3,037,000
0
862 ,000
8,71.L ,u00
<
[a] Includes a cost of capital, which is estimated to oe 15 ^
percent. *»¦
[b] Tnis coat is for scenarios which require the use of
expanded microprocessor capability, and assumes tnat nuif
of total R&D is performed by manufacturers.
-------
Table V-12
Amortized Capital Costs Per Vehicle ($1981)
Scenar io
Certification,
Development, Test
Facility, and R&D[a]
(1983)
Certification
(1984-87)
Total Cost
(Present Value
in 1984)
Cost Per
HA
Fleet[cJ
Vehicle l b J
Nat.
Fleet
la
56,290,000
56,000
61,968,000
—
l
lb
39,393,000
0
43,334,000
—
1
lc
49,219,000
56,000
54,190,000
—
1
2
0,021,000
0
8,823 ,000
6
1L d J
3a
4,456,000
0
4,902,000
—
1 [ d J
3b
8,748,000
56,000
y,671,U00
±[dj
[a] The cost for R&D in this taole includes that cost wnich is incurred by
outside suppliers as well as manufacturers, as an example, tne total r&d
cost for scenario la is $2.3 million instead of the $1.15 million shown
in Table V-ll.
[b] Amoritization weighted to result in an e^ual cost per vehicle over tne
years of production cited. Discount rate assumed to oe iU percent.
Expenses are assumed to occur in January 1 of the given year anu revenues
are assumed to be received on December 31 of each year.
[c] For fixed-point scenarios only. Assumes 3.1 percent of all vehicles are
sold at high altitude.
[d] Although these costs are much less than $1, a cost of $1 will be carried
through the remainder of this report as a conservatively high estimate.
-------
V-26
A. Sticker Price Increase
Vehicle purchasers will have to pay for the costs of
emission control hardware and engine modifications on vehicles
affected by high-altitude standards. In addition, they will
have to pay for the costs of the capital investment required of
manufacturers. The vehicle manufacturers will pass on these
costs to the purchaser by increasing the retail price or
"sticker price" of the vehicle.
The average costs for control hardware has already been
estimated for each scenario on a per vehicle basis and is
summarized in Table V-5. To these must be added the
amortization of the fixed costs of control, which are
summarized in Table V-12. The average sticker price increase
for vehicles affected by the high-altitude standards is
approximately $141-149 in scenario la, $49-62 in scenario lb,
$68-82 in scenario lc, $5-6 in scenario 3a, $22-27 in scenario
3b, and $14-15 in scenario 2.
B. Maintenance Costs
Two control hardware items that will probably result in
additional maintenance costs or operating costs are the
turbocharger and the feedback control system.
The turbocharger hardware itself should require no
maintenance. However, the addition of a turbocharger requires
that the engine oil and the filter be replaced approximately
every 3,000 miles that the vehicle is driven[ll] rather than
every 6,000-7,500 miles for a naturally aspirated engine. Over
the 100,000 mile useful life of a light-duty vehicle this would
mean that a turbocharged vehicle would require roughly 15-20
more oil changes than its naturally aspirated counterpart.
With the cost of an oil change around $15, this would amount to
a lifetime cost of about $225-300. The average lifetime of a
light-duty vehicle is 10 years, and assuming that the oil
changes occur at their regular intervals and using a 10 percent
discount rate, the maintenance cost is $150-185, discounted to
the time of vehicle purchase. This cost should be included in
the overall cost for each scenario requiring turbochargers
(i.e., 40 percent in scenario 1 and 5 percent in scenario lc).
The feedback control system may also require some
maintenance even though manufacturers are currently not
recommending any maintenance intervals for these systems.
Recent data from EPA's I/M program implies that roughly 15
percent of all vehicles will require some maintenance of the
oxygen sensor.[12] A survey of local dealerships indicate that
the replacement cost of an oxygen sensor with labor is about
$20. This would probably occur once during the vehicle's
lifetime, and for purposes of this analysis, EPA assumes that
this will occur halfway through the vehicle's life, or about
-------
V-27
five years after the vehicle was purchased. Thus, this cost
discounted to the year of vehicle purchase is about $14. A
sales-weighted average cost is then 15 percent of $14, or about
$2 per light-duty vehicle. This cost would apply to 13 percent
of the vehicles in scenario la, 31 percent in lb, 26 percent in
lc, and 8 percent in 3b. These vehicles are estimated to
change from nonfeedback to feedback systems to comply with the
high-altitude standards.
It is also possible that the microprocessor will require
some maintenance. However, by 1984, or the first year of
high-altitude standards for LDVs, the development of a
microprocessor should be improved so that no maintenance is
necessary.
If the above costs are sales-weighted appropriately for
each scenario in which they occur, then the total maintenance
cost is $61-75 in scenario la, $1 in scenario lb, $9-10 in
scenario lc, and essentially $0 in scenario 3b.
C. Fuel Economy
Purchasers may also benefit from fuel economy savings due
to implementation of high-altitude emission control technology,
as was already discussed in detail in Chapter IV. The fuel
economy benefits summarized here will be incremental to those
experienced by continuing the 1982 and 1983 interim rule, the
continuation of which is the base scenario. These fuel economy
benefits are best EPA estimates based on a wide range of data
indicating the fuel economy improvements or penalties
associated with the various control techniques. For
turbochargea vehicles, an average fuel economy benefit of 5
percent was observed compared to naturally aspirated vehicles
of similar driving performance. Vehicles with aneroids will
also experience some fuel economy benefit at high altitude. In
the baseline case, one aneroid is assumed to already exist on
some vehicles. If an additional aneroid is installed (see
scenario 3a and 3b), then a fuel economy benefit of 0 percent
is expected. If two aneroids are added (scenario 2) the data
indicate an improved fuel economy of 2 percent. For vehicles
with feedback control, a 3 percent fuel economy benefit should
be possible over those vehicles in the baseline case using
open-loop (nonfeedback) control. Expanding the
microprocessor's capability for some vehicles may also improve
fuel economy, but data confirming this effect could not be
found, so no fuel economy benefit will be projected here.
These fuel economy benefits can be expressed in terms of
cost savings if the gallons of fuel saved and the price per
gallon are estimated for the full lifetime of a vehicle.
First, the lifetime of a light-duty vehicle is assumed to be
100,000 miles accumulated over a period of 10 years.[13]
Second the corporate average fuel economy (CAFE) standard is 27
-------
V-28
mpg in 1984 and 27.5 thereafter for LDVs. This latter CAFE
standard (27.5 mpg) is used in this analysis. Based on this
information, the average amount of fuel consumed by LDVs is
estimated to be 3,600 gallons for vehicles affected by this
regulation. A fuel economy benefit of 1, 2, 3, and 5 percent
translates to a savings of 36, 72, 106, and 173 gallons,
respectively, for LDVs. Finally, the price of unleaded
gasoline is currently (at the time of this writing) about $1.30
per gallon.
The total fuel economy savings must be discounted back to
the original year or purchase. A 5 percent discount rate is
used for fuel costs, instead of a 10 percent rate which is used
elsewhere in this analysis, to indicate that the expectec
inflation of fuel costs will be much greater relative to all
other goals. This procedure has been cone in a recent EPA
regulatory analysis.[3 ] If it is assumed that fuel usage
occurs equally for each year during a vehicle's lifetime, then
one-tenth of the total fuel consumption is usea each year for
LDVs. Thus, based on the discount rate, the total fuel
consumption, the price of unleaded gasoline, and the percent
savings of fuel as a result of using each control hardware
component, the following fuel economy benefits are observed for
LDVs when compared to the baseline case: 2 additional
aneroids, $80; turbocharger, $190; and feedback control, $115.
A summary of these fuel economy savings is shown in Table V-13.
Of course, these cost savings will not apply to all
vehicles in each scenario, since all vehicles will not be
equipped with each of these control hardware components.
However, a sales-weighted average savings can be calculated for
each scenario, again based on the percentage of control
hardware expected to be used for each scenario (Tables V-2 and
V-3).
b. Net Cost to Consumer
The net cost to the consumer for each scenario is shown in
Table V-14. These costs include the control haraware price
increase, the R&D for expanded memory capability, the
maintenance cost, the certification and development cost, test
facility costs, and the fuel economy savings. The costs shown
in Table V-14 apply either to those vehicles sold at high
altitude only (scenario 2), or to those vehicles sold
throughout the nation.
IV. AGGREGATE COSTS
The aggregate costs to the nation of complying with the
1584 high-altitude standards consist of the sum of increased
costs for development, certification, emission control
hardware, engine modifications, R&D for expanded memory
capability, test facility modifications and additions, and
-------
V-29
Table V-13
Fuel Economy improvement
Control Hardware
Two Aneroids
Turbocharger
Feedback Control
Percent
Improvement
(best estimate)
2%
5%
3%
Cost Savings
($1981)[a]
8U
19U
115
Estimated using a 5 percent discount rate, for a vehicle
lifetime of 10U,00U miles over period of 1U years.
-------
V-29
Table V-14
Net Cost to Consumer (LDVs) ($1901 per vehicle)
Certification,
Net Venicle
Cost
Scenario
Hardware
Maintenance!a]
Development, R&D,
and Test Facilities!a]
Fuel
Economy[aJ
HA
Fleet
Nat.
Fleet
la
140-148
61-75
1
-91
W/a LbJ
111-133
lb
48-61
1
1
-36
N/A
14-27
lc
67-81
9-1U
1
-40
N/A
37-51
2 [b]
8-9
N/A
6
-25 -11
to -luLcJ
N/A
3a
4-5
N/A
1
0
N/A
5-6 <
3b
21-26
0
1
-9
N/A
12-17 £
[a] Discounted to year of purchase.
[bj Not applicable.
[c] Costs for scenario 2 apply to vehicles sold at high altitude only, or
about 3.1 percent of the national vehicle fleet.
-------
V-31
changes in fuel consumption and maintenance. This cost is
simply the net cost to the consumer times the number of
vehicles sola. These costs will be calculated for the first
five years after which the high-altitude standaro becomes
effective. The year 1984 will be used as the base year to
compare present values of the money throughout the time period
of concern. A discount rate of 10 percent is used for all
costs when calculated in this manner. All costs are expressed
in terms of 1981 dollars.
The aggregate cost to the nation is dependent on the
number of light-duty vehicles sold during the time period.
Although any sales projection of this type will be rough due to
the many social and economic factors involved, the sale
projections shown in Table V-6 are suitable for this analysis.
Only the percentage of the national fleet to which net costs
are applicable are used in calculating aggregate costs.
The aggregate cost for each alternative scenario is shown
in Table V-15. As can be seen, for LDVs the scenarios range
from a net savings to a net cost of $4,S74 million. This large
range is due to the fact that fuel economy benefits are the
predominant factor in determining the net cost estimate of some
scenarios relative to others. This effect is explored further
in the sensitivity analysis at the end of this chapter.
V. SOCIOECONOMIC IMPACT
In this section, the socioeconomic impact on
manufacturers, dealers, and users will be discussea.
A. Impact on Manufacturers
The impact on manufacturers will be analyzed in two
separate categories. First, the capital expenditures which
manufacturers must confront could be burdensome and will thus
be investigated. Second, the additional cost of each vehicle
could affect demand and, subsequently, the sales of each
manufacturer.
1. Capital Expenditures
Capital expenditures consist of development costs for
high-altitude engine calibration, research ana development for
expanded memory capability, and certification. The total
capital costs have already been calculated and are shown in
Table V—11. The bulk of the capital expenditures will occur in
the first year of this regulation, and the real burden of a
high-altitude standard can be viewed as raising the first
year's fixed costs before vehicle sales begin to repay the
investment. In this section, the first year cost is examined
with a 15 percent cost of capital.
-------
V-32
Table V-15
Aggregate Costs[aJ
Scenario
Aggreyate Costs
(millions of $1981)
la
4151-4974
lb
524-1010
lc
1384-1907
2
Wet Savings [b]
3a
187-224
3b
449-636
[a] Present value in 1984.
[b] If the estimated fuel economy benefit is excluded, the
aggregate cost would range up to $17 million.
-------
V-33
It is possible that other capital expenses will result
from a high-altitude standard. However, as previously stated,
all capital costs associated with the control hardware, such as
tooling, machinery, and building expenses, are expected to be
borne by outside suppliers with the exception of expanding
memory capability, where approximately half of the total R&D
expense is likely to be incurred by manufacturers.
The first year capital costs are shown for each scenario
in Table V-16 and are simply the sum of development, first year
certification, test facility additions, and R&D for expanded
memory capability, multiplied by 1.15 to account for the
predicted cost of capital. Most of the costs will be incurred
by manufacturers with a large number of engine families for
certification and development. Of course, the impact of
capital cost will vary according to each scenario. Admittedly,
the largest capital expenses, which are necessary in scenarios
la, lb and lc, may be significantly more burdensome to
manufacturers than those capital expenses resulting from the
other scenarios.
Smaller LDV manufacturers will have the most trouble
raising capital than will the. larger manufacturers, since they
have less vehicles to spread their cost over. In particular,
scenarios la, lb, and lc could significantly affect the abil'ity
for each manufacturer to finance the required investment. In
contrast, scenarios 2, 3a, and 3b would be much less burdensome
and should not significantly affect small manufacturers.
It is true that some of these large , manufacturers will
absorb a higher percentage of the total capital costs than will
.other manufacturers, but usually a manufacturer which pays a
higher capital expense has earned a larger profit when compared
to manufacturers spending less on capital. Thus, even for the
scenarios requiring more capital, larger manufacturers should
not be severely impacted.
2. Effects on the Demand for High-Altitude Vehicles
The impact of sales can be evaluated by examining the
sticker price increases per vehicle for each scenario. With
the exception of scenario 2, the sticker price increase will
probably occur for all vehicles in the nation, even though the
air quality benefit is obtained at altitudes above 1,800 feet.
Scenarios 1 and 3 require all vehicles to be be equipped with
special control haroware whether or not they are ever used at
higher elevations. In scenario 2, cost increases will probably
occur only for vehicles sold for principal use above 4,000
feet, since only high-altitude vehicles need be equipped with
control hardware.
-------
Table V-16
First Year Capital Costs to All Manufacturers
of Light-Duty Vehicles ($l98l)[a]
Scenario
Development
Certification
Test
Facility
K&D
Total
la
9/401/UUU
21/589,000
23,000,000
1,15U,00U
55,140 ,000
lb
7/073/000
7 /020 ,000
23,000,000
1,150,000
38,243,000
lc
9/401/000
14/158/000
23,0U0,000
1,i50,000
48,069,000
2
8/021/000
0
0
—
8,021,000
3a
3/881/000
0
0
575/000
4,456,000
3b
4/313/000
3,573,000
0
862,000
8,748/000
L a J A 15 percent cost of capital is included.
-------
V-35
The price increase for each scenario can be applied in
conjunction with the following equation to estimate the impact
of sales:
% Change in Vehicle Sales = [price elasticity]
[0.5 (% change in vehicle price)]
In the above equation, the price elasticity for vehicles
during 1984-88 is assumed to be the same as that for 1982-83,
cr 0.35.[13] Next, the total sales must be determined, and
according to Table V-6, the total 5-year sales for gasoline
LDVs (that are not sold in California) is 44.58 million. The
average cost of a vehicle is roughly $7,000 in 1S81 dollars.[13]
The maximum impact of high-altitude sales would occur for
scenario la, where the average sticker price increase would be
about $141-149 (hardware costs plus fixed costs). This impact
would reduce sales by as much as 0.37 percent, or by 165,000
vehicles over a perioa of five ^ears. Sales by the smaller
manufacturers may decrease at a higher rate than that by the
larger manufacturers, due to a smaller manufacturer's lower
production volume with which to amortize fixed costs. Thus,
while larger manufacturers may not be greatly affected by the
increase in price under scenario la, the smaller manufacturers
may be affected somewhat more severely under this scenario.
The next largest impact would occur under scenario lc,
where a sticker price increase of $68-82 could reduce sales by
as much as 0.21 percent, or by 8S,000 vehicles. Although the
loss of sales here is estimated to be over 75,000 less than the
5-year loss of sales for scenario la, this loss of sales could
still adversely affect the smaller manufacturers. Under
scenario lb, which has a sticker price increase of about
$49-62, or close to that of scenario lc, the estimated 5-year
loss of sales is about 69,000, and the impact on manufacturers
would be nearly the same as that projected for scenario lc.
Scenario 3b could increase the sticker price up to $27 with a
potential sales reduction of about 30,000 vehicles. Such a
sales loss could also have an adverse impact on some small
manufacturers. Thus, scenarios la, lb, lc, and 3b could
adversely affect sales, particularly for the smaller
manufacturers.
For scenario 3a, the maximum sticker price increase is $6,
and this could cause a decrease of about 7,000 vehicles over a
5-year period. This is small compared to the loss of sales
determined for scenarios la, lb, lc, and 3b, and would probably
not significantly affect any of the manufacturers. For
scenario 2, the sticker price increase for vehicles sola at
high altitude is about $15, and based on a 5-year high-altitude
sales projection of 1.4 million, the potential sales loss is
525 vehicles. This represents about 105 lost sales per year,
which is less than the amount of lost sales estimated in the
-------
V-36
regulatory analysis for the 1982 and 1983 interim
standards.[13] As was concluded in the interim standards
analysis, this loss of sales should not noticeably affect the
sales of any manufacturer. Thus, scenarios 2 and 3a should
not significantly affect a single manufacturer's sales due to
the sticker price increases examined above. Also, there should
be no impact on employment or productivity in the industry.
B. Impact on High-Altitude Dealers
The effects of a 1984 high-altitude standard on
dealerships can be divided into two general areas: reduced
model availability and higher vehicle prices. These changes
arise from each scenario and affect vehicle sales, and hence,
dealership profitability.
1. Model Availability
As previously stated, the 1977 high-altitude regulations
resulted in the unavailability of many models and optional
engine configurations in high-altitude areas. At that time,
manufacturers chose to limit model availability in
high-altitude areas because the small percentage of the market
represented by high-altitude sales (about 3 percent) did not
justify the development costs required to certify the emission
control capabilities of all their vehicle configurations. Some
high-altitude dealers alleged that this resulted in lost sales.
Model availability should not be a problem with each of
the scenarios where exemptions from high-altitude sales are 10
percent or less (scenarios la, lc, 3a, and 3b). Since almost
all new vehicles in each scenario will be certified for sale at
high altitude, each manufacturer will be more likely to make a
substantial amount of his product line available to
high-altitude purchasers. Conceivably, a manufacturer might
comply with the regulations by certifying all models for
high-altitude sales but choose not to offer certain models to
high-altitude purchasers. However, EPA believes that due to
the expense involved with certifying and developing each
vehicle, manufacturers will offer almost all of the vehicle
configurations for sale at high altitude. Also, vehicles that
are exempted from sale at high altitude are most likely
low-power vehicles that would normally not be sold at high
altitude. Thus, the model availability at low altitude should
remain unchanged and the model availability a high altitude
should not be noticeably affected for scenarios where
exemptions are 10 percent or less.
For scenarios where exemptions are estimated to be greater
than 10 percent (scenarios lb, 25 percent and 2, 15 percent),
model availability may be a problem for dealers at high
altitude. This greater number of exempted vehicles may curb
availability of the more popular fuel efficient vehicles. If
-------
V-37
it is determined that these exemptions would affect model
availability significantly, methods could be introduced to ease
this problem, such as allowing waivers for particular exempted
models, or specifying different criteria so that fewer vehicles
would be exempted. Also it is believed that these control
strategies combined with the manufacturers* increased
experience with altitude-compensating emission control systems
during 1S82 and 1983 will keep availability to acceptable
levels.
2. Higher Vehicle Prices
The cost of a high-altitude vehicle depends on whether the
dealer acquires the new vehicle by ordering it as original
equipment from the factory or through a "dealer trade" with a
low-altitude dealer. Under scenario 2 some low-altitude
vehicles acquired in dealer trades must be modified into the
proper high-altitude configuration before they are sold.
The cost of factory-built high-altitude vehicles depends
on the manufacturer's pricing strategy. Manufacturers may
choose to amortize the cost of these standards across vehicles
sold at high-altitude only (for scenario 2), or over the entire
national production.
In scenarios la, lb, lc, 3a, and 3b, it is likely that
manufacturers will recover these costs over nationwide sales.
Although the high-altitude market represents only a small
percentage of total sales, this small amount may be more
significant for manufacturers during their ascent from recent
economic difficulties and as the entire market shifts to more
competitive smaller cars than in the 1S82 and 1983 model years.
Therefore, competition for high-altitude sales among
manufacturers could be quite intense. Additionally, the
industry's historical price leader, General Motors, will likely
incur the least additional cost no matter which of these
scenarios is used. Therefore, because of competition with such
companies as GM, other manufacturers may indeed raise
high-altitude vehicle prices less than the sticker prices
indicated previously in this analysis in order to remain
competitive.
In scenario 2, manufacturers may choose to recover their
costs only on high-altitude sales, and the estimated average
price increase is about $15. This represents approximately 105
lost sales per year. As stated in the regulatory analysis for
the interim standards,[13] there are about 1,000 high-altitude
dealerships. However, only those dealers representing
manufacturers whose vehicles must be recalibrated to meet the
high-altitude standards (41 percent of the fleet) will be
affected by significantly higher vehicle prices. The
manufacturers building LDVs that generally will not require
recalibration are GM, AMC, Nissan, Volkswagen, Volvo, JRT, BMW,
-------
V-38
Peugeot, Porsche, and Saab. The actqal number of high-altitude
dealers selling recalibrated vehicles is not readily
available. Nevertheless, it is possible to reasonably estimate
the number of high-altitude dealerships selling vehicles with
significantly higher prices based on the national fraction of
dealer outlets representing manufacturers which build
recalibrated LDVs. Using this analogy, EPA estimates that 50
percent of the 1,000 high-altitude dealers potentially may be
affected by significant first price increases. Since only 105
lost sales should occur, most of the 500 potentially affected
dealers will not experience any sales reduction. Therefore,
the potential price increase for original equipment vehicles
should have no significant economic impact on individual
high-altitude dealerships. Of course, in scenario 2, if a
manufacturer chooses to amortize his cost over the entire
national fleet, then the cost increase would be so small that
sales should not be affected at all.
In some cases, dealer trades may be adversely affected by-
each of the scenarios. The impact on sales, however, remains
conjectural. Dealer trades generally involve small rural
dealers who cannot stock a wide variety of vehicles and must
trade with large metropolitan area dealers to satisfy customer
demand. Dealer trades were estimated by the Coloraoo
Automobile Dealers Association to involve from 10 to 15 percent
of sales by small rural dealers. Therefore, the potential
impact will predominantly apply to high-altitude dealerships
which are isolated from high-altitude metropolitan areas. EPA
is unable to estimate the number of such isolated dealerships,
but believes it is reasonable to postulate that the number is
relatively small since most high-altitude areas are within
"trading" distance (a few hundred miles) of a high-altitude
metropolitan area. Also, not all manufacturers will have
special high-altitude vehicles, so some dealers should not have
any problem. Nevertheless, even though the number of
high-altitude dealerships which may trade with low-altitude
metropolitan dealerships may be relatively small, the potential
impact on these dealers needs to be explored further.
First, for the continuous proportional and the continuous
statutory scenarios (scenarios la, lb, lc, 3a, and 3b),
vehicles sold at high altitudes should have identical
configurations as their low-altitude counterparts and, thus,
dealers should have no problem obtaining a desired vehicle.
This of course assumes that the aesired vehicle is a non-exempt
vehicle. As was discussed in the previous section, scenarios
with a significant percentage of exemptions may cause model
availability problems. For the fixed-point statutory scenario,
or scenario 2, about 41 percent of the high-altitude vehicles
will differ from their low-altitude counterparts. However,
dealers should generally have access to all high-altitude
models from the factory. But, if models are available from the
factory, why be concerned with dealer trades at all?
-------
V-39
In the past, for fixed-point strategies, high-altitude
dealers have stated that their primary concern is being able to
obtain vehicles that are in high demand. Apparently, in 1977
when most vehicles involved factory installed high-altitude
modifications, there were sometimes long delays in obtaining
vehicles and sales were lost. EPA has addressed this problem
for the interim standaras by requiring all vehicles that do not
automatically comply with the standards, to be capable of being
modified to do so. This will also hold true for scenario 2 of
the 1S84 standards. This will help ensure that the small
number of isolated, rural dealerships which trade with
low-altituoe dealers can obtain vehicles on a timely basis and
modify them into the proper configuration before sale. The
only potential barrier could be that the modification might be
expensive. The Colorado Automobile Dealers Association
estimated that modifications costing perhaps up to $150 per
vehicle would not affect sales. As discussed in the Summary
and Analysis of Comments for the 1982-83 standards,[14] EPA
expects many vehicles will be modifiable for less than that
amount. Since dealer trades appear to be most critical for
high-demand vehicles for which long ordering delays may be
experienced, the real potential impact of the high-altitude
standards is whether or not dealers will lose sales for those
few vehicles that are in high demand and are expensive to
modify.
Looking closer at scenario 2, if it is first assumed that
by the time a prospective customer contacts a dealership the
customer has previously decided that a specific new car is
necessary and that a substitute (i.e., one that is more
available) is not suitable, there are two fundamental problems
in the "worst case." First, the vehicle of choice must be
ordered from the factory but there will be a delay. Second,
the vehicle of choice may be available sooner but must be
modified at an extra cost of a few hundred dollars.
Since under scenario 2 it will be illegal for the
prospective customer to purchase a low-altitude vehicle, a
decision based primarily on economics must be made (i.e., is it
worth the extra cost to have the specific vehicle sooner), or
is it better to wait and, in the process, save money. No
matter which choice is made, the sale is not lost in this
example.
Of course, under scenario 2 a prospective customer may not
have previously decided on a particular high-demand vehicle
that is in short supply. If this is the case he may shift to
another more available vehicle from the same manufacturer. In
this case the sale would not be lost. The customer may also
decide to purchase a comparable vehicle from another dealer.
In this case the potential high-altitude sale would not be
lost. Or, the customer may have only been marginally
-------
V-40
interested in the particular "problem" vehicle and decide not
to buy any vehicle. In this case the potential sale would be
lost.
In summary, the regulations under any of the scenarios
should not significantly affect overall high-altitude sales
unless a large number of vehicles are exempted, as was
discussed in the previous section. The potential for adversely
affecting sales is predominantly limited to relatively
isolated, rural high-altitude dealerships which must "modify"
low-altitude vehicles acquired in dealer trades with
low-altitude dealerships and this would probably occur only
under scenario 2. For these isolated dealers, the potential
problem should be limited to the relatively few "high-demand"
vehicles which are expensive to modify into the proper
high-altitude configuration. Even in these instances, however,
only a portion of such potential sales would be lost.
Therefore, it is reasonable to assume that any single
high-altitude dealership will not be greatly affectea by
high-altitude standards.
C. Impact On User
Users will be affected by higher new vehicle prices.
Along with this initial price increase the purchaser will pay
for additional maintenance and benefit from fuel economy
savings. The average sticker price increase would be greatest
for scenario la, which has a cost of $141-149. This represents
about 2 percent of the total vehicle cost, and could affect a
consumer's ability to purchase new vehicles. Also, under this
scenario, maintenance costs are high, although this increase
should be offset by a fuel economy benefit. Thus, some
purchasers may have trouble financing their desired vehicle.
For scenario lb and lc, the sticker price increase is
approximately 1 percent of the total vehicle cost, and would
have a lesser effect than scenario la on the ability for
consumers to purchase new vehicles. In scenarios 2 and 3a, the
sticker price increase is 2 percent or less of the initial
vehicle price. Purchasers should have no problem paying for
the desired vehicles in these scenarios. Looking at scenario
3b, the sticker price increase is about 0.4 percent. These are
small percentages when compared to scenarios la, lb, and lc,
and would not affect purchasers of high-altitude dealers.
Thus, scenarios la, lb, and lc could affect a purchaser's
ability to buy a new vehicle, while scenarios 2, 3a, and 3b
would affect very few, if any, purchasers.
VI. SENSITIVITY ANALYSIS
The net cost to consumers shown in Table V-14 was based on
conservative estimates for fuel economy benefits and was also
based on the assumption that the control hardware costs were
accurately determined.
-------
V-41
Earlier in this chapter, the following best estimates of
fuel economy improvements were used for vehicles compared to
the base-line case: 2 additional aneroids, 2 percent,
turbocharging, 5 percent; and feedback control, 3 percent.
These numbers are conservative estimates based on a range of
data for fuel economy improvements or penalties that were
observed. Based on an analysis of fuel economy ciata as
explained in Chapter III of this report, it is not unreasonable
that fuel economy changes for the above control technologies
could be expanded to the following ranges: 2 additional
aneroids, 0-4 percent, turbocharging, S-10 percent, and
feedback control, 0-5 percent. These fuel economy changes
would lead to the following savings over the lifetime of a
light-duty vehicle: 2 additional aneroids, $0-150;
turbocharging, $190-380; and feedback control, $0-190. In
addition, these savings could be expanded even further if other
control technologies which were discussed in Chapter III were
also included (i.e., expanding the microprocessor capability
and adding 1 aneroid). However, the sensitivity analysis will
be conducted only with the potential fuel savings for the
technologies already included in the analysis since any fuel
economy impacts of the remaining control hardware has already
been rejected. The effect of including the range of savings
can be seen in Table V-17, where the net cost due to the
sensitivity analysis is compared to the net cost under the best
estimate of fuel economy savings.
The sensitivity analysis on fuel economy shows that the
potential fuel savings could significantly affect the net cost
of each scenario with the exception of scenario 3a. Table V-17
shows that a small change in fuel economy benefit can cause a
large change in the net cost. This is especially apparent in
scenario 2, which is extremely sensitive to the estimated fuel
economy increment. Instead of a net savings, implementing the
standards in this scenario may actually result in a net cost.
Interestingly, while the estimated, fuel economy benefits are
crucial to the outcome of the net cost under each scenario, the
ranking of the scenarios in order of costs remains essentially
unchanged. In conclusion, this sensitivity analysis generally
shows that the net cost estimates in Table V-14 should be
conservative and thus should be referred to when considering
the net cost of a high-altitude standard. For scenario 2,
however, this analysis shows that the potential to
underestimate the actual cost of the standards is such that it
would be better to consider a range for the net vehicle cost
based on the best fuel economy estimate and no assumed
benefit. Therefore, a net vehicle cost of $-11 to $15 will be
considered for scenario 2 in the succeeding chapters of this
report to retain the conservative nature of the analysis.
Next, the sensitivity of control hardware costs needs to
be examined. For purposes of this analysis, a sensitivity of
+30 percent will be used for hardware costs so that the effect
-------
Table V-17
Net Cost Pee Vehicle Due to Fuel Dconumy
Sensitivity Analysis (LDVs)($1981)
Best Kstihidtes
Certification, Net Vehicle Cost Net Venicie Cost
Control Development, R&D Fuel HA Nat. Ha Nat.
Scenario Hardware Maintenance!a] and Test Facility Economy1 aJ Fleet Fleet Fleet Fleet
la $140-148 61-75 1 -177 to W/AlbJ 25-148 N/A lil-UJ
-76
lb $48-61 1 1 —S9 to N/A -9 to N/A ±4-27
U oJ
lc $67-81 9-10 1 -68 to N/A 9-U2 N/A J7-'ji
-10
2 IcJ $ 8-9 N/A 6 -47 to -J3 to N/A -11 to N/A
0 15 -10
3a $4-5 N/A 1 U/A J-6 W/A a-o
0
Jb $21-26 N/A 1 -15 to N/A 7-27 N/A iz-i7
0
[a] Discounted to year of vehicle purchase,
lb) Not applicable.
Ic| Costs for scenario 2 apply to vehicles sold at liiyh altitude only, or
about 3.1 percent of the national vehicle fleet.
-------
V-43
of the net cost can be observed. The results of a +30 percent
control hardware cost sensitivity can be seen in Table V-18.
The effect of a +30 percent sensitivity does appear to affect
net costs and widens the range of cost for each scenario
considerably. However, as for fuel economy, each scenario
remains the same with regard to its cost rank. In addition,
scenario 2 is less sensitive to reasonable changes in the
hardware cost estimates than for fuel economy.
In summary, the effect of a change on fuel economy and
control hardware costs could affect significantly the net cost
for each scenario. The most dramatic effect is observed for
fuel economy, where a small percent change in fuel economy
improvement or penalty leads to a larger percent change in the
net cost of each scenario. A change in control hardware costs
leads to about an equal change in net costs and this would
probably not affect the determination of which scenario has the
least economic impact. A change in fuel economy, on the other
hand, has greater potential to affect the determination of
which scenario is the most desirable economically and thus
neeas to be estimated accurately. Because of the extreme
sensitivity of scenario 2 to the. fuel economy estimates, a
range cf $-11 to $15 will be considered for the net vehicle
cost under this scenario in subsequent chapters of the report.
-------
Table V-lB
Met Cost Per Vehicle Due to Fuel Ecoiioi.i/
Sensitivity Analysis (LPVs) ( $198i )
Cer tit" icat ion,
best
klbtiuate
Net Vehicie Cost
best ttstiuutes
Net Vehicle Cust
Sceuar10
Control
Hardware
Maiuteuancela 1
Development, H&D
and Test Facility
Fuel
£couoia/ld J
HA
Fleet
Nat.
Fieet
Ha
Fleet
Not.
Fleet
la
$98-192
61-75
1
-91
N/AlbJ
09-177
N/A
Ij.1-11.
lb
$34-79
1
1
-36
N/A
0-45
N/A
14-2/
lc
$47-105
9-1U
1
-41)
N/A
17-76
N/A
J7-S1
2 [c]
$ 0-12
N/A
6
-25
1 1
-J t-
U
rr
C
—
-11 to
lu
n/ h
3a
$ 3-7
N/A
1
0
N/A
4-tJ
W/A
l>-0
3b
$15-34
N/A
1
-9
N/A
7-27
N/A
12-17
I a ] Discounted to year o£ vehicle purchase.
(b) Not Applicable.
IcJ Coats for scenario 2 a^ply to vehicles sold at hiyh altitude only, or
about J.l percent of the national fleet.
-------
V-45
References
1. "Regulatory Analysis and Environmental Impact of
Final Emission Regulations for 1984 and Later Model Year
Heavy-Duty Engines," U.S. EPA, OANR, OMS, ECTD, SDSB/ December/
1979 .
2. "Draft Regulatory Analysis - heavy-Duty Diesel
Particlate Regulations," U.S. EPA, OANR, OMS, ECTD, SDSB,
December 23, 1980.
3. "Regulatory Analysis of the Light-Duty Diesel
Particulate Regulations for 1982 and Later Model Year
Light-Duty Diesel Vehicles," U.S. EPA, OANR, OMS, ECTD, SDSB,
February 20, 1980.
4. "Cost Estimations for Emission Control Related
Components/Systems and Cost Methodology Description," Rath and
Strong, Inc., Leroy H. Lindgren, EPA-460/3-78-002, March 1978.
5. "Cost Estimations for Emission Control Related
Components/Systems and Cost Methodology Description Heavy-Duty
Trucks," Rath and Strong, Inc., Leroy H. Lindgren,
EPA-460/3-80-001, February, 1980,.
6. "Closed-Loop Control, Three-Way Catalyst Automotive
Emission System Costs and Issues," U.S. Department of
Transportation, National Highway Traffic Safety Administration,
Contract No. DOT-HS-8-01912, September, 1979.
7. "Summary and Analysis of Comments on the Notice of
Proposed Rulemaking for the Control of Light-Duty Diesel
Particulate Emissions from 1981 arici Later Model Year Vehicles,"
U.S. EPA, OANR, OMS, ECTD, SDSB, October, 1979.
8. Test Car List, 1981, First Edition, U.S. EPA, OANR,
OMS, CD.
9. "Investigation and Assessment of Light-Duty Vehicle
Evaporative Emissions Sources ana Control," EPA 460/3-76-014,
June, 197 6.
10. Telephone Conversation with Mike Brown of
Environmental Tectronics Corporation.
11. "Buick's Turbocharged V-6 Powertrain for 1978,"
General Motors Corporation, Buick Motor Division, Wallace,
T.F., SAE Paper No. 780413.
12. "Summary Discussion of Results from Three-Way
Catalyst Vehicles," U.S. EPA, OANR, OMS, ECTD, IMS, Hughes,
David, September, 1980.
-------
V-46
13. "Final Regulatory Analysis - Environmental ana
Economic Impact Statement for the 1982 and 1983 Model Year
High-Altitude Motor Vehicle Emission Standards," U.S. EPA/
OANR, OMS, ECJTD, SDSB, October, 1980.
14. "Summary and Analysis of Comments on the Notice of
Proposed Rulemaking for High-Altitude Emission Standards for
1S82 and 1983 Model Year Light-Duty Motor Vehicles," EPA, OANR,
OMS, ECTD, SDSB, October 1980.
-------
Chapter VI
Cost Effectiveness1
I. INTRODUCTION
Cost effectiveness refers to an analytical method by which
several alternative means of achieving a desired goal are
evaluated based on their costs (usually in dollars) and a
separate measure of effectiveness. In this report, the goal is
to reduce automotive emissions of hydrocarbons and carbon
monoxide.
II. METHODOLOGY
The costs of' meeting this goal for each scenario were
determined in Chapter V. For the purposes of determining cost
effectiveness, total costs (i.e., the costs for both the
manufacturer and the operator), are used on a per vehicle basis
and discounted to the year of vehicle purchase. These costs
are then equally allocated between the two pollutants being
controlled.
In order to measure the effectiveness of each scenario,
total lifetime emission reductions are used on a per vehicle
basis. Since the primary purpose of this report is to
determine if the described alternative approaches to continuing
the 1982 and 1983 interim high-altitude standards are cost
effective enough to warrant further consideration, it is the
incremental cost effectiveness of each scenario over the base
scenario that needs to be examined. Thus, the relative costs
and emission' reductions are the differences between those under
the interim high-altitude standards (base scenario) and each of
the other scenarios. The incremental cost effectiveness is
determined, then, by dividing the added cost of control by the
additional amount of pollutant removed from the atmosphere.
III. ANALYSIS
The incremental cost, emission reduction, and the
resulting cost effectiveness are presented in Table VI-1 for
each alternative scenario. Before discussing the overall
results of the analysis, a brief explanation of the
cost-effectiveness values listed for scenario 1 is necessary.
As. previously stated in Chapters III and IV, the stringent
requirements of meeting the statutory standards at all
altitudes are expected to cause nonfeedback systems to be
replaced by feedback systems in scenario 1. The emission
factors used in this analysis assume that a significant number
of feedback systems will experience catastrophic failures.
These in-use failures result in excessive CO emissions well
beyond allowable limits. . For this reason, the air quality
analysis for scenario 1 shows a significant increase in CO
-------
Table VI-1
Incremental Cost Effectiveness
of LDGV Control Strategies
Costs[a]
(dollars
Reductions[b]
(10~3 metric tons
Incremental
Cost Effectiveness
per vehicle)
per
vehicle)
HC
CO
Scenario
Low
High
HC
CO
Low
High
Low
High
la
2120
2530
10 .5
-181.6
101,000
121,000
lc]
t c J
lb
265
515
10.5
-18i.6
12,600
24,500
[Cj
[C]
lc
705
970
10.5
-181.6
33,600
46,200
[c]
IC J
2
-ll[d]
15
13 .0
239.9
neg.
575
neg.
30
3a
95
115
0.5
7.7
95,000
115,000
6,200
7,500
3b
230
325
0.5
7.7
230,000
325,000
14,900
21,100
[a] 1981 dollars discounted to year of vehicle purchase. Costs were
allocated to the vehicles above 1,800 feet for scenarios 1 and 3 and to
the vehicles above 4,000 feet for scenario 2 in order to correspond to
the same vehicles used to determine emission reductions.
[b] Emission reductions were calculated by dividing fleetwide emission
reductions in Tables IV-9 and IV-10 by LDGV sales in Table IV-5.
[c] The cost-effectiveness values for CO under scenario 1 were not presented
in the table since emissions of this pollutant may increase under tnis
strategy. However, it is also possible that CO emissions may decrease by
about the same total amount listed for botn scenarios 2 and 3. Using
this assumption, the cost effectiveness would range from $535 to $5,10U
per metric ton for CO.
[d] The great uncertainties associated with the expected fuel economy benefit
make any conclusion that a savings may result from implementing scenario
2 very tentative.
-------
VI-3
emissions, since the nonfeeaback systems that were replaced by
feedback systems do not exhibit such catastrophic failures.
The data base with which the emission factors were
developed, however, is quite limited and extrapolating the
current trend of catastrophic failures into the future is
speculative at this time. It is very likely that new
information will show different failure rates and that the
lifetime emissions for feedback systems may possibly be more
like those for nonfeedback systems. Therefore, instead of
increasing CO emissions, scenario 1 (with exclusively feedback
systems) may actually reduce this pollutant by an amount
similar to that calculated for scenario 2 (with a mixture of
both systems) in areas above 4,000 feet. At intermediate
altitudes (1,800 feet to 4,000 feet) the incremental benefits
may be greater than for scenario 3 which controls vehicles to
only proportional levels. The possibilities that CO emissions
may increase or decrease under scenario 1 are considered in the
following discussion.
Table VI-1 shows that scenario 2 is predicted to be the
most cost effective of the alternative scenarios analyzed in
this report. The cost-effectiveness values range up to $575
per metric ton of HC, compared to $12,600 to $121,000 per
metric ton for scenario 1 and $95,000 to $325,000 per metric
ton for scenario 3. under scenario 2, the cost effectiveness
for CO ranges up to $35 per metric ton. The cost effectiveness
of CO emission reductions from scenario 3 ranges from $6,200 to
$21,100 per metric ton. as previously stated, CO emissions
under scenario 1 could either increase or decrease. The cost
effectiveness was not calculated for the possibility that CO
emissions might increase. The cost effectiveness of potential
CO reductions under scenario 1 was determined by assuming that
any benefit would be the total of those listed for scenarios 2
and 3 (0.25 metric tons per vehicle). using this value, the
cost effectiveness of reducing CO emissions under scenario 1
ranges from $535 to $5,100 per metric ton.
Scenarios 1 and 3, as analyzed in this report, are
predicted not to be cost effective for the primary reason that,
while emission reductions are achieved solely on vehicles above
1,800 feet (roughly 5 percent of the fleet), vehicles sold
below this altitude (95 percent of the fleet) must also bear
the additional cost of applying high-altitude control systems
on them as well. This approach tends not to be cost effective.
In Chapter IV, the environmental consequences of a
high-altitude fleet comprised of one low-altitude vehicle out
of every ten vehicles in high-altitude areas was discussed as
part of the sensitivity analysis. As explained in that
chapter, emissions at high-altitude would be greater under the
fixed-point strategies (scenario 2 and the base scenario) if a
certain fraction of the vehicles were low-altitude cars, as
-------
VI-4
opposed to all high-altitude vehicles, because low-altitude
cars are unable to adequately compensate for the effects of
less dense air under these scenarios. However, if a continuous
strategy were implemented (scenarios 1 or 3), this trend would
not be true since low-altitude vehicles at higher elevations
would be able to adjust.
Table vi-2 presents the incremental cost effectiveness of
the alternative control options based on cost estimates in
Chapter V and the emission reductions determined in Chapter IV
for a fleet containing one low-altitude vehicle for every nine
high-altitude vehicles (refer to Table IV-17). As can be seen,
even in this extreme case scenario 2 is still by far the most
cost effective of the analyzed alternatives to the base
scenario.
Cost-effectiveness figures for other control strategies
already adopted are provided for comparison in Table VI-3. The
values listed in that table and Table VI-1 show that even the
high cost estimate for scenario 2 is comparable to many of the
other HC control strategies. The high estimate for CO control
is less than that for the LDV statutory standards or
inspection/maintenance programs.
Compared to the interim (1982-83) high-altitude standards,
scenario 2 may be less cost effective. This is to be expected
since the benefits of each succeeding increment of pollution
control is generally more costly to attain. The important
point to consider is that if, after additional study, further
emission reductions are necessary to assume attainment and
maintenance of the NAAfcS in high-altitude areas, further
control of motor vehicle emissions in these areas is predicted
to be cost effective.
It should be pointed out that at the time this document is
being prepared, the cost-effectiveness figures reported in
Table VI-3 for the HDG evaporative strategy are based on a
proposal that is not yet final. Also, this strategy may not be
strictly incremental because there could have been intermediate
control levels chosen. Thus, the cost effectiveness was
determined over a wide range of emission reductions rather than
just the last increment. This approach tends to yield low
cost-effectiveness values.
While scenario 2 is the most cost effective of the six
alternative scenarios being studied, the final decision as to
the viability of scenario 2 will be reached in the next
chapter. This decision will depend on such factors as the
overall costs of compliance, the overall emission reduction
potential, the effect on air quality, and the effect of
potential errors in estimates and assumptions used in the
analysis.
-------
Table VI-2
Cost Effectiveness Comparison
Adjusted for Low-Altitude Vehicles Above 4/000 Feet
[aj
Scenar io
COStS [ L) ]
(dollars
per vehicle)
Low High
Reduct ions[c]
(10~3 metric tons
per vehicle)
Incremental
Cost Effectiveness
(dollars/metric ton)
HC
CO
HC
CO
Low
Low
Hiyh
la
2120
2530
11.1
-175 .2
95,500
114,000
LdJ
luj
lb
265
515
11.1
-175 .2
11,900
23,20U
LdJ
I d J
lc
705
970
11.1
-175 .2
31,800
43,700
LdJ
Ldj
2
-11
15
12.0
215 .9
ney.
625
ney.
3b
3a
95
115
1.1
14 .1
43,200
52,300
3,400
4,100
3b
230
325
1.1
14 .1
105,000
148,000
8 ,20U
11,500
[a] Assumes that 1 out of every 10 vehicles sold above 4,000 feet emits as if
it were a low-altitude vehicle.
[b] 1981 dollars discounted to year of vehicle purchase. Costs were
allocated to the vehicles above 1,800 feet for scenarios 1 and 3 and to
the vehicles above 4,000 feet for scenario 2 in order to correspond to
the same vehicles used to determine emission reductions.
[c] Emission reductions were calculated by dividiny fleetwide emission
reductions in Tables IV-1G and IV-17 by LDGV sales in Table IV-5.
[d] The cost-effectiveness values for CO under scenario 1 were not presented
in the table since emissions of this pollutant may increase under this
strategy. However, it is also possible that CO emissions may decrease by
about the same total amount listed for both scenarios 2 and 3. Usiny
this assumption, the cost effectiveness would ranye from $58U to $5,5UU
per metric ton for CO.
-------
VI-6
Table vi-3
Cost Effectiveness Comparison with
Other Emission Control strategies
Control Program
Baseline
Emissions [a]
Emissions
After Control
Strategy
Cost
Effectiveness
($/iaetcic ton)
HC CO
LDV statutory[2]
HC
0.9
HC
0.41
734
67
Standards
CO
15
CO
3.4
LDV I/M [3]
—
—
640
58
LDT 1984 [4]
HC
1.7
HC
0.8
195
15
Standards
CO
18
CO
10
HDE 1984
Standards[5][b]
(gasoline)
HC
1.5
HC
1.3
305
10
CO
25
CO
15.5
(diesel)
HC
1.5
HC
1.3
325
--
CO
25
CO
15 .5
Motorcycle
HC
9
HC
8-22 .
5 [c]
582
Neg
Standards [6]
CO
34.67
CO
27 .4
HDG Evap. [7][d]
HC
1.8
HC
0.17
200
Interim 1982-83
HC
1.47(cars)
HC
1.33
(cars)
393
12
HA Standards [8]
4 .19(trucks)
3.78
(trucks)
CO .
16 .23(cars)
CO
13 .21
(cars)
73 .02(trucks)
55.65
(trucks)
[a]Emission levels are expressed as a standard in grams per mile
except for the HDE 1984 levels wnich are in yrams per brake
horsepower-hour.
[b] The baseline and after control strategy emission values were
based on different test procedures (see reference 5).
[c] Sliding scale based on engine displacement (cubic
centimeters).
[d] The evaporative standard is in terms of grams/test and
converted to g/mi here to facilitate comparison.
-------
VI-7
References
1. "Final High-Altitude Emission Standards for 1982 and
1963 Model Year Light-Duty Motor Vehicles," U.S. EPA, 45 FR
66984, October 8, 1980.
2. "Interagency Task Force on Motor Vehicle Goals
Beyond 1980," U.S. EPA, Department of Transportaiton, Prepared
by the Panel of Automotive Manufacturing and Maintenance, March
1976.
3. "Update on the Cost Effectiveness of Inspection and
Maintenance," U.S. EPA, OANR, OMS, ECTD, IMS, EPA-AA-IMS/81-9,
April 1981.
4. "Regulatory Analysis and Environmental Impact of
Final Emission Regulations for 1984 and Later Model Year
Light-Duty Trucks," U.S. EPA, OANR, OMS, ECTD, SDSB, May 20,
1980.
5. "Regulatory Analysis and Environmental Impact of
Final Emission Regulations for 1984 and Later Model Year
Heavy-Duty Engines," U.S. EPA, OANR, OMS, ECTD, SDSB, December
1979.
6. "Environmental and Economic Impact Statement
Exhaust and Crankcase Regulations for the 1978 and Later Model
Year Motorcycles," U.S. EPA, OANR, OMS, ECTD, SDSB, 1976.
7. "Regulatory Analysis for the Proposed Evaporative
Emission Regulations for Heavy-Duty Vehicles," U.S. EPA, OANR,
OMS, ECTD, SDSB, April 1980.
8. "Environmental and Economic Impact Statement for the
1982 and 1983 Model Year High-Altitude Motor Vehicle Emission
Standards," U.S. EPA, OANR, OMS, ECTD, October 1980.
-------
Chapter VII
Comparing the Alternative Control Scenarios
for Light-Duty Gasoline-Fueled Vehicles
I. INTRODUCTION
The primary purpose ot this report is to oetermine if any
of the alternatives to continuing the current fixed-point
proportional standards (the base scenario), which are analyzed
in this report, deserve further consideration. This chapter
evaluates the seven control scenarios identified in Chapter II
on the basis of the previous analyses (Chapters III-VI) for
light-duty gasoline-fueled vehicles (LDGVs) . It then
identifies what appears to be the most desirable of the six
alternative high-altitude control scenarios. Subsequent
chapters will evaluate this single alternative scenario for its
effects on light-duty, gasoline-fueled trucks (LDGTS) in
addition to aiesel-powered light-duty vehicles and light-duty
trucks (LDDs).
Using LDGVs to identify a viable control strategy-
simplifies the analysis by eliminating the need to review LDGTs
and LDDs in aetail for each scenario, but ooes not compromise
the final selection. Any alternative scenario which is
predicted to be inappropriate for LDGVs will also be
inappropriate for the overall motor vehicle fleet because this
category forms the bulk of the total vehicle market.
A detailed description of the control scenarios analyzed
in the report was originally presented in Chapter II, but is
repeated here for clarity.
A. Base Scenario: Fixed-Point Proportional Standards with
Exemptions
This scenario will require vehicles to comply with
high-altitude standards of 0.57 g/mi HC, 7.8 g/mi CO, 1.0 g/mi
NOx, and 2.6 g/test evaporative HC at only one elevation (i.e.,
5,300 feet). It is essentially a continuation of the current
high-altitude requirements for 1982 and 1983 model year
vehicles. Exemptions may be granted for certain low-power
vehicles that would perform unacceptably at high altitude and
that may have technical difficulty in meeting the standards
cost effectively.
B. Scenario 1: Continuous Statutory Standards
This alternative scenario will require vehicles to comply
with standards of 0.41 g/mi HC, 3.4 g/mi CO, 1.0 g/mi NOx, and
2.0 g/test evaporative HC (the current low-altitude standards)
is required at all elevations up to a maximum altitude. This
scenario is subdivided further, depending on the maximum
-------
VII-2
altitude to which compliance must be demonstrated and on
whether performance-based exemptions are provided:
1. la - Compliance required up to 10,200 feet and no
exemptions allowed;
2. lb - Compliance required up to 6,000 feet with
exemptions allowed; and
3. lc - Compliance required up to 6,000 feet and no
exemptions allowed.
C. Scenario 2; Fixed-Point statutory Standards with
Exemptions
This strategy is similar to the base scenario, except that
at 5,300 feet vehicles must meet the low-altitude statutory
emission standards (0.41 g/mi HC, 3.4 g/mi CO, 1.0 g/mi NOx and
2.0 g/test evaporative HC) instead of the proportional
standards presented in the base scenario.
D. Scenario 3: Continuous Proportional Standards
Under this control strategy, vehicles must meet standards
that increase proportionally with altitude up to 6,000 feet.
At 1,800 feet, the emission standards are the low-altitude
standards; at 5,300 feet, they are the high-altitude standards
outlined in the base scenario. The standards vary linearly in
between these two altitudes and up to 6,000 feet. Like
scenario 1, this scenario has two variations:
1. 3a - Exemptions are allowed; and
2. 3b - No exemptions are allowed.
These seven scenarios differ in their approach to solving
the high-altitude emissions problem. The base scenario and
scenario 2 are termed "two-car" strategies since they allow
vehicle modifications performed on vehicles sold for principal
use above 4,000 feet. This is not the case for scenarios la,
lb, lc, 3a, and 3b. Any modifications which are necessary to
satisfy the high-altitude requirements in these scenarios must
be performed on all affected vehicles regardless of the
altitude at which they are sold. These scenarios are termed
"one-car" strategies.
II. METHODOLOGY
Identifying the most desirable alternative high-altitude
control strategy is done in three parts. First, the
alternative control scenarios are evaluated by examining the
aggregate cost, total emissions reduction, and cost
effectiveness from the previous chapters. The control
-------
VII-3
scenarios are then ranked according to their ability to reduce
high-altitude emissions in a cost-effective manner. second,
the preliminary ranking is reviewed to determine if further
consideration of the underlying assumptions would change the
order and to determine if the first ranked scenario merits
further consideration. Third, the most desirable alternative
scenario is evaluated further based on its air quality impact
in comparison to the base scenario and with regard to the need
for additional pollution control in high-altitude regions.
III. REVIEW OF THE ECONOMIC IMPACT, ENVIRONMENTAL IMPACT, AND
COST EFFECTIVENESS
A. Economic Impact
The economic impacts of the six alternative high-altitude
control scenarios were calculated for a 5-year period (1984-89)
as an increment beyond the costs of continuing the current 1982
and 1983 high-altitude standards (base scenario). The
aggregate cost to the nation for each of the six alternative
strategies appears in Figure VII-1. These costs are primarily
based on estimates of the requisite hardware (see Chapter III),
and include expenses for development, certification, emission
control hardware, engine recalibration, R&D for expanded
microprocessor capacity in the electronic engine control unit,
and changes in fuel consumption and maintenance. A range of
costs was estimated for each scenario because of uncertainties
in defining the production costs.
Scenario 2 is by far the least costly of the alternative
scenarios (Figure VII-1). However, the exact cost of this
scenario is difficult to determine. A net savings to the
nation may result if the estimated fuel economy benefit, which
is larger than the relatively low hardware costs, is included
in the cost calculation. This estimated benefit is based on
very limited information and is considered tentative at this
time. Nevertheless, even without including this estimated
benefit, the incremental cost of scenario 2 is only about $17
million.
The primary cause of the large price differential between
this scenario and the other scenarios is that only scenario 2
is a two-car strategy. Therefore, under this scenario only-
vehicles sold at high altitude (above 4,000 feet) need to be
equipped with additional emission controls. The other
scenarios are one-car strategies and require emission control
modifications on all vehicles throughout the nation regardless
of where they are sold.
Scenarios 3a and 3b have the lowest net cost with a range
of about $190 million to $640 million. The cost difference
between allowing and not allowing exemptions is clear.
Granting exemptions for low-power vehicles in scenario 3a
-------
VII-4
FIGURE VII-1
Incremental Aggregate Costs to the Nation
(1981 Dollars)
6000
Key:
CO
CC
<
-I
-I
o
o
u_
O
(/)
z
o
5000
4000
3000
2000
1000
4974
4151
1907
1010
1
o Low Estimate
High Estimate
636
Uncertain [al
187
224
1a
1b
1c
3a
3b
SCENARIOS
[a] SCENARIO 2 MAY RESULT IN A NET SAVINGS IF THE INCREMENTAL PURCHASE PRICE INCREASE
(ABOUT $15.00 PER HIGH-ALTITUDE VEHICLE) IS OFFSET BY A POTENTIAL FUEL ECONOMY BENEFIT
(ABOUT $25.00 PER HIGH-ALTITUDE VEHICLE). IF THE POTENTIAL FUEL ECONOMY IMPROVEMENT
IS EXCLUDED, THE COST WOULD BE ABOUT $17 MILLION. THE ESTIMATED FUEL SAVINGS IS
TENTATIVE AT THIS TIME BECAUSE OF THE LIMITED DATA BASE.
-------
VII-5
reduces the cost of the continuous proportional standards by
about $260-410 million.
Scenarios la, lb, and lc are the most stringent of the
alternative scenarios and their higher costs reflect the
difficulty in achieving them. Meeting the statutory standards
at all elevations up to 10,200 feet without exemptions makes
scenario la the most expensive at $4.15 billion to $4.97
billion. Reducing the elevation to 6,000 feet (scenario lc)
decreases the cost by about 65 percent to between $1.38 billion
and $1.91 billion. Scenario lb is the least stringent
continuous statutory standard with a cost of $524 million to
$1.01 billion. It differs from lc in that exemptions for
low-power vehicles are allowed. Providing exemptions reduces
the cost in this scenario by about $900 million.
B. Environmental impact
The incremental lifetime emissions from LDGVs which are
affected by the 1984 high-altitude standards were calculated
for a 5-jear period. Three basic factors were used: 1) the
number of miles traveled by a vehicle in its lifetime, 2) the
emission factor for each pollutant (amount of pollution per
mile), and 3) the number of vehicles affected'by the standards.
The emission reductions achieved over the lifetimes of
those vehicles sold in the first 5 years of regulation for each
alternative control strategy are presented in Figure VII-2.
Scenarios la, lb, and lc show the greatest reductions in
hydrocarbon (HC) emissions. These strategies also appear to
increase carbon monoxide (CO) emissions in high-altitude
areas. A similar penalty would occur at low altitude (not
shown) and is caused by the expected catastrophic failure of
some feedback systems which are used as emission control
hardware in these scenarios. However, the emission factors
that generated these results for scenarios la, lb, and lc are
considered to be very preliminary at this time because of the
limited number of vehicle tests upon which the catastrophic
failure rate is based. The vehicles in this sample were also
"early" feedback systems and it is difficult to extrapolate the
results from these tests to future systems. (This same general
qualification also applies to the other scenarios because they
also rely on estimates of emission factors for future feedback
emission control systems, although to a lesser degree.)
EPA believes that because of the limitations in the
original data base, the emission factors may change as further
tests are conducted in the Agency's surveillance programs. The
direction of this change is, of course unknown, but as stated
in Chapter IV, it is most likely that future sistems will
exhibit fewer catastrophic failures. if this occurs, the
emission factors for feedback systems may be more like those
for nonfeedback systems which show a significant reduction in
-------
V11-6
FIGURE VII-2
Incremental Emission Reductions [a]
400
24.6
-425
1b
24.6
-425
1 c
337
17.9
1.1
18.0
|HC
~ co
3a
1.1
18.0
3b
SCENARIOS
[a] ALTHOUGH CO EMISSIONS MAY INCREASE IN SCENARIO 1. IT IS ALSO
POSSIBLE THAT THIS POLLUTANT MAY BE SIGNIFICANTLY REDUCED.
-------
VII-7
HC and CO emissions when controlled to the statutory levels at
high altitude. Therefore, instead of producing an adverse
environmental impact, it is possible scenarios la, lb, and lc
may substantially reduce CO emissions in areas above 1,800
feet. For intermediate altitudes (1,800 feet to 4,000 feet)
these scenarios may produce an incremental CO benefit that is
greater than that calculated for the proportional standards in
scenario 3 (18,000 metric tons). In areas above 4,000 feet,
the CO reduction may be much like that for scenario 2 which
also controls emissions to statutory levels (331,000 metric
tons). These potential reductions for scenarios la, lb, and lc
total approximately 349,000 metric tons (Figure VII-2).
If the catastrophic failure rates assumed for scenario 1
are valid, scenario 2 offers the greatest overall reduction in
emissions of all the alternative scenarios (Figure VII-2). In
this scenario, it is unnecessary to replace nonfeedback systems
with feedback systems as in scenario 1. Therefore, the
possibility of additional catastrophic failures is avoided
along with any potential CO penalty. If the failure rates
prove to be significantly less, however, scenario 1 may offer
slightly greater benefits than the combined total of scenarios
2 and 3 (Figure VII-2).
The emission reductions for scenarios 3a and 3b are
relatively small (Figure VII-2). The proportional standards in
these scenarios have a benefit only at intermediate altitudes
from 1,800 to 4,000 feet. There is no additional control above
4,000 feet since the base scenario already provides emission
control to proportional levels at those elevations.
C. Cost Effectiveness
The cost effectiveness of further reducing LDGV emissions
in high-altitude areas of the country was calculated in Chapter
VI by dividing the total cost per vehicle for each alternative
control strategy by the respective pollutant reductions.
Figure VII-3 shows the incremental cost-effectiveness values
($/metric ton) for each control scenario. The cost
effectiveness of scenario 1 was not calculated because of the
possibility that CO emissions might increase, although it was
calculated tor a possible decrease. The cost-effectiveness
calculation for scenario 2 was expanded to evaluate a "worst
case" assumption that no fuel economy benefit would accompany
implementing the more stringent high-altitude standards of this
control strategy. Therefore, the range of values for scenario
2 is based on the inclusion or exclusion of the estimated fuel
increment. Figure VII-3 also contains cost-effectiveness
values of other mobile source emission regulations for
compar ison.
Scenario 2 is the most cost-effective option being
considered (Figure VII-3). It would reduce HC emissions at a
-------
400,000
300,000
200,000
100,000
70,000
50,000
40,000
30,000
20.000
10,000
7000
5000
4000
3000
2000
1000
700
500
400
300
200
100
70
50
40
30
20
10
ECOSTE
iSIBILIT
LY THE I
FIGURE VII-3
Incremental Cost Effectiveness of Control
Scenarios and Other Emission Control Strategies [a]
(1981 Dollars Per Metric Ton)
¦ hc
Qco
i
00
1c 2(b) 3a
SCENARIOS
LDV LDV
Stat. I/M
Stds.
:FECTIVENESS FOR CO CONTROL IN SCENARIO 1 IS NOT PRESENTED BECAUSE OF THE
THAT EMISSIONS OF THIS POLLUTANT MAY INCREASE BEYOND THE BASE SCENARIO.
IGH-COST ESTIMATE IS SHOWN FOR SCENARIO 2. THE LOW ESTIMATE WAS NOT SHOWN
INCLUDES AN ESTIMATED FUEL ECONOMY BENEFIT, WHICH IS TENTATIVE AT THIS TIME.
HOE HDE Motor- HOG Interim
1984 1984 cycle Evap. 1982-83
Stds. Stds. Stds. Stds. H.A. Stds.
(gasoline) (diesel)
-------
VII-9
cost of up to $575 per metric ton, compared to $12,600 to
$121,000 per metric ton for scenario 1 and $95,000 to $325,000
per metric ton for scenario 3. Under scenario 2, CO emissions
would be reduced at a cost of up to $30 per metric ton. The
cost effectiveness of potential CO reductions under scenario 1
was found by assuming that the benefit would be the total of
those listed for scenarios 2 and 3 (0.25 metric tons per
vehicle). Using this value, reducing CO emissions under
scenario 1 ranges from $535 to $5,100 per metric ton. The cost
effectiveness of CO emission reductions under scenario 3 ranges
from $6,200 to $21,100 per metric ton.
Scenarios 1 and 3 have extremely poor (i.e., high)
cost-effectiveness values for one basic reason. While emission
reductions are achieved solely on vehicles above 1,800 feet
(roughly 5 percent of the fleet), vehicles sold below this
altitude (95 percent of the fleet) must also bear the
additional cost of applying high-altitude control systems.
This approach generally tends net to be cost effective because
the added cost is not offset with an attendant emissions
benefit.
Figure VII-3 shows that scenario 2, even under the high
cost estimates, is the only alternative control scenario
analyzed which is comparable to other mobile source emission
control regulations. Therefore, scenario 2 appears to be be a
cost-effective approach to reducing high-altitude emissions
from LDGVs.
Compared to the interim (1982-83) high-altitude standards,
scenario 2 is less cost effective. This is to be expected
because the benefits of each succeeding increment of pollution
control are generally more costly to attain. The important
point is that if further emission reductions are necessary to
attain and maintain the NAAfcS in high-altitude areas, further
control of motor vehicle emissions in these area appears to be
cost effective.
IV. RANKING OF THE ALTERNATIVE HIGH-ALTITUDE SCENARIOS
Although alternative control strategies are not directly
comparable based solely on cost and benefits, they generally
can be compared based on their cost effectiveness. Cost
effectiveness is, therefore, the primary oecision-making
criterion.
The alternative control scenarios analyzed in this report
are ranked as follows in order of their ability to reduce
high-altitude emissions and their incremental cost
effectiveness:
1. Scenario 2 - Fixed-point proportional standard with
exemptions;
-------
VII-10
2. Scenario 3a - Continuous proportional standard with
exemptions and a ceiling of 6,000 feet;
3. Scenario 3b - Continuous proportional standard
without exemptions and a ceiling of 6,000 feet;
4. Scenario lb - Continuous statutory standard with
exemptions and a ceiling of 6,000 feet;
5. Scenario lc - Continuous statutory standard without
exemptions and a ceiling of 6,000 feet; and
6. Scenario la - Continuous statutory standard without
exemptions and a ceiling of 10,200 feet.
Scenario 2 ranks first. it is by far the most cost
effective and is projected to provide the largest air quality
benefit of the alternative scenarios. Scenarios 3a and 3b are
ranked second and third, respectively, in order of increasing
cost. Although these scenarios reduce HC less than scenarios
la, lb, and lc, they ranked higher because of the potential for
increased CO emissions which result in scenarios la, lb, and
lc. This potential penalty, in addition to their very high
price of scenarios la, lb, and lc, relegates them to be ranked
last. Therefore, scenario lb is fourth, scenario lc is fifth,
and scenario la is sixth, in order of their increasing cost.
V. EVALUATION OF THE ALTERNATIVE SCENARIOS BASED ON THEIR
SENSITIVITY TO THE ASSUMPTIONS OF THE ANALYSIS
The complexity of analyzing alternative high-altitude
standards for 1984 and later model year motor vehicles
necessitates the use of simplifying assumptions and
projections. No matter how carefully considered, however,
these estimates are subject to error and individual
interpretation. Since the choice of scenario 2 as the best
alternative control strategy may depend on these estimates, the
sensitivity of this choice to the underlying assumptions must
be explored.
The sensitivity evaluation will be conducted in two parts.
First, the critical assumptions concerning the technical
requirements and catastrophic failure rates of electronic
control systems in addition to the number of low-altitude
vehicles operating at high altitude will be reviewed to
determine if any other scenario might be ranked first instead
of scenario 2. Second, the merits of scenario 2 itself will be
evaluated by reviewing the assumptions concerning the
associated change in fuel economy, the fleet mix of feedback
and nonfeedback systems, the number of exemptions, the use of
low-altitude vehicles in high-altitude areas, and the level of
the emission standards.
-------
VII-11
A. Reexamination of the Control Scenario Ranking
For scenarios la, lb, and lc the data in Figures VII-1
through VII-3 can be used to show that even if a favorable case
is constructed with regard to the technical requirements and
catastrophic failure rates, these scenarios would still remain
very cost ineffective. The emissions consequence of reducing
the catastrophic failure rate for feedback systems used in the
analysis has already been discussed in detail. The potential
CO benefit of such a change, therefore, can be assumed to be as
great as the combined total for scenarios 2 and 3 (0.25 metric
tons per vehicle). since the effect of catastrophic failures
is not as significant for HC emissions, the reduction in this
pollutant is unchanged from that shown for scenario 1 in Figure
VII-2. Also, assume that the emission control requirements
were overestimated because of the limited amount of available
data. A favorable case would involve reducing the emission
control cost below the low estimate of $265 per vehicle for
scenario lb by a third to perhaps $175 per vehicle. This could
result from using the upper range of fuel economy benefits or
by not replacing nonfeedback systems with feedback systems.
Even these extremely favorable costs and emission benefits
would iield cost-effectiveness values in excess of $8,300 per
ton of HC and $350 per ton of CO. These values are still much
higher than those for scenario 2 (Figure VII-3). For scenarios
3a and 3b the same type of sensitivity analysis can be
performed with similar results.
The environmental benefits of implementing scenarios la,
lb, lc, 3a, and 3b increase if a significant number of vehicles
operating at high altitude are assumed to be low-altitude
vehicles. This would increase the gaseous emissions under the
base scenario since some vehicles would be emitting at
uncontrolled levels. By implementing continuous control
strategies, every vehicle would meet the appropriate standards
at high altitude regardless of where it was originally
purchased. This would increase the benefits of high-altitude
regulations under scenarios 1 and 3. A hypothetical "worst
case" was constructed in Chapter VII to evaluate the cost
effectiveness of such an assumption. The sensitivity analysis
in that chapter assumed that one out of every ten vehicles
operated at high altitude originated from a low-altitude area.
This made very little difference in the cost effectivenss of
scenario 1 but did make scenario 3 significantly more cost
effective. For example, the cost of reducing HC under scenario
3 decreased from a range of $95,000 to $325,000 per metic ton
to a range of $43,200 to $148,000 per metric ton. The cost of
reducing CO decreased from a range of $6,200 to $21,100 per
metric ton to a range of $3,400 to $11,500 per metric ton.
However, these improved cost-effectiveness values remain
substantially more expensive than other emission control
strategies. Therefore, scenario 2 remains the most reasonable
alternative.
-------
VII-12
B. Reexamination of the Merits of Scenario 2
The sensitivity analysis has thus far shown that scenario
2 is the most desirable alternative strategy analyzed in this
report. Now, scenario 2 itself will be evaluated.
Specifically, even the high cost estimates which exclude the
expected fuel economy benefit show it is reasonably cost
effective and that its cost to the consumer is not excessive
(about $15 per vehicle). However, how sensitive is this to the
assumptions of the analysis and would concerns about these
assumptions detract enough from the merits of scenario 2 to
remove it from further consideration? These questions are
discussed below.
The sensitivity analysis, contained in chapter IV and V,
shows that the alternative scenarios are especially sensitive
to five of the assumptions that are necessary to complete this
study. First, the analysis is very sensitive to changes in
fuel consumption which should result from the installation of
high-altitude emission control hardware, in fact, the net cost
of scenario 2 was found to be so sensitive to this, that the
lower limit of the possible fuel economy improvements (the
range was 0 to 4 percent for nonfeedback controlled vehicles)
was included in the cost-effectiveness values. Therefore, the
"worst case" (i.e., the lower limit), has already been
accounted for. If the upper limit of 4 percent improvement
were included in the cost-effectiveness analysis, the potential
net savings would be greater than that which already may be
possible. Because of the extreme sensitivity of the analysis
to projected changes in fuel economy, this factor must be
carefully reevaluated as additional information becomes
available.
Second, the cost effectiveness should be considerably
better if many manufacturers switched from feedback systems to
nonfeedback systems. such a change has already been made by
Ford and is accounted for in EPA's emission control estimates.
Ford originally intended to utilize feedback systems, but
switched to nonfeedback systems because of their lower selling
price. Whether other manufacturers will do this is unknown.
This analysis has generally assumed that manufacturers
currently using feedback systems will continue to do so.
However, these manufacturers have stated that the competitive
nature of the automotive market will force a continued
reevaluation of their commitment to feedback systems. Because
nonfeedback systems have no inherent ability to compensate for
the effects of altitude on vehicle emissions, the benefit of
controlling these systems in scenario 2 should be greater than
the benefits of controlling feedback systems. For this reason,
the introduction of more nonfeedback systems into the market
should improve the cost effectiveness of scenario 2.
-------
VII-13
Third, the desirability of scenario 2 is also very
dependent on the number of vehicles that could be exempted.
Because the exemption provision has such a pervasive effect on
the economic, energy, and social implications of any
high-altitude standard, it will be briefly discussed further.
In the future, the number of vehicles that may need to be
exempted from the proportional standards under the base
scenario and from the statutory standards under scenario 2 was
estimated in Chapter II to be about 5 and 15 percent,
respectively, of the current high-altitude fleet. These are
only rough estimates because of the dynamic nature of the motor
vehicle fleet composition. Furthermore, this study assumes
that the exemption criteria used in the 1984 high-altitude
regulations would be patterned after the interim high-altitude
standards where exempted vehicles may not be sold above 4,000
feet. This criterion precipitates one of the most significant
issues involved in scenario 2: what type of vehicles would be
available at high altitude under the base scenario but would be
unavailable under scenario 2?
The justification for granting exemptions from the interim
high-altitude standards (base scenario) was that exempted
vehicles generally would be low-powered vehicles that were
designed primarily for the low-altitude market. When these
vehicles are driven at high altitude, the lower air density
degrades their performance to such a degree that they would
only be sold in small numbers, hence, the impact of exemptions
would be minimal. From this, it can logically be assumed that
the vehicles which have somewhat more power than the exempted
vehicles must be good sellers, because they provide better fuel
economy. Under scenario 2 some of these high fuel economy
vehicles, which would be sold in the absence of such a
regulation, would become unavailable. Such vehicles could
represent 10 percent of the current market (15-5 percent).
This could have three consequences: 1) an adverse consumer
reaction may be generated, 2) it appears to be contrary to
national energy policy, and 3) there may be an aaverse economic
impact.
On a fleet-wide basis, the expected fuel economy penalty
from eliminating some of the more fuel efficient vehicles at
high altituoe would be offset by the greater fuel economy
benefit that is expected to accompany statutory standards.
This benefit occurs because in order to comply with the
standards, manufacturers must reduce excessively fuel-rich
engine operations. It could also be argued that the fuel
efficiency benefit of the exempted vehicles may be more
imagined than real. Those vehicles requiring exemptions
generally operate with power enrichment at high altitude. This
operation results in additional fuel being metered into the
combustion chamber to maximize power output. However, using
this extra fuel also decreases fuel economy, sometimes very
-------
VII-14
significantly. Because the effect of increasing altitude is to
reduce an engine's power output, a high fuel economy vehicle at
low altitude which rarely operates in power enrichment may
operate a significant amount of the time in this mode at high
altitude to compensate for the power loss that occurs at higher
elevations. This could, theoretically, make a relatively
higher powered vehicle more fuel efficient when operated at
high altitude, since enough engine power would be available to
prevent or minimize excursions into power enrichment.
Therefore, the potential fuel economy impact of granting
exemptions cannot be settled at this time. However, the
present analysis assumed no fuel economy penalty due to the
increased exemptions; any excursion from this assumption could
be in the direction of a fuel penalty ana added cost.
The potential for adverse consumer reaction is also
speculative. Further technical achievements may reduce the
number of exempted vehicles. The mass marketing of throttle
body injection with its more precise fuel metering and, hence,
better emissions control capability may reduce the need for
exemptions. Also, as better information becomes available, the
exemption criteria may be refined to resolve apparent problems
regarding the number and type of exempted vehicles. Finally,
other options involve waiving the high-altitude requirements
for some vehicles, thereby allowing them to be sold at higher
elevations, or requiring these vehicles to meet a less
stringent high-altitude standard.
Fourth, the incremental cost-effectiveness of scenario 2
relative to the other scenarios could be affected by
low-altitude vehicles being driven in high-altitude areas. in
Chapter V, the environmental consequences of a high-altitude
fleet comprised of 10 percent low-altitude vehicles were
discussed as part of the sensitivity analysis. This was
considered to be the upper limit of possible low-altitude
vehicle use at high altitude. As explained in that chapter,
emissions at high-altitude would be greater under the
fixed-point strategies (scenario 2 and the base scenario) if a
certain fraction of the vehicles were low-altitude vehicles, as
opposed to all high-altitude vehicles. This is because many
low-altitude vehicles cannot compensate adequately for the
effects of less dense air under these scenarios. However,
under a continuous strategy (scenarios 1 or 3), this trend
would not be true since low-altitude vehicles at higher
elevations would be able to adjust. Since the emission
reduction potentials of the various scenarios are affected by
the presence of low-altitude vehicles, cost effectiveness is
also affected. In Chapter VII it was determined that in the
"worst case," the cost effectiveness of scenario 2 would
increase for HC control from $580 per metric ton to $625 per
metric ton. Similarly, the cost of CO removal increases from
$30 per metric ton to $35 per metric ton.
-------
VII-15
The fifth, and final, assumption that will be reviewed was
not specifically analyzed in the previous chapters and concerns
the emission reductions that may be achievable if scenario 2
was implemented. The costs and benefits of this scenario are
based on the assumption that statutory standards would be
promulgated at high altitude. These standards were chosen as a
readily identifiable alternative to the proportional standards
of the interim (1982-83) high-altitude program and also to be
consistent with statutory requirements for 1984 and later. The
previous discussion in this section demonstrated that
high-altitude regulations, as with any requirement, must be
chosen to moderate or eliminate a complex mixture of
potentially adverse impacts (e.g., environmental, economic,
energy, and model availability). The most efficient standards,
therefore, may be at some level other than the statutory
standards, hhile it appears that more stringent control beyond
the proportional standards is feasible and cost effective
perhaps down to the levels of the statutory standards, less
control may provide the majority of the needed environmental
benefits in a more cost-effective manner. Only further study
can identify the optimum level of control.
The above discussion has served to identify some
potentially negative effects of implementing scenario 2. It
has also served to show that some of these effects can neither
be proven or disproven at this time. In addition, the exact
level of the emission standards in scenario 2 which ultimately
may be chosen will depend on a complex mixture of variables
representing the technical feasibility, total cost, and the
number of exemptions that may be required. The appropriate
level for high-altitude standards appears to be below the
proportional values and may be as low as the statutory values
contained in this analysis. As with any program that attempts
to improve public health and is in its early stages of
development, further study can be expected to define the exact
level of the standards and to resolve or mitigate potential
adverse effects. Therefore, it appears that the air quality
benefits of continuing the existing proportional standards at
high altitude may be improved upon if a control program
incorporating aspects similar to scenario 2 is adopted.
VI. COMPARISON OF SCENARIO 2 TO THE BASE SCENARIO
This section compares the future air quality benefits of
the base scenario and scenario 2, and examines whether the
incremental benefits of scenario 2 are great enough to merit
further evaluation with regard to the effects on LDGTs and
LDDs. This decision will be made by reviewing the benefits of
the two scenarios in relation to the future air quality needs
of high-altitude regions, as well as further considering of the
basic assumptions with which scenario 2 was developed.
-------
VII-16
Before discussing the air quality modeling results for the
two scenarios, a brief discussion of the expected effects of
incremental control strategies is useful. Any single
incremental strategy will usually yield much smaller air
quality benefits than those achieved with the original
controls. Most of the controls producing dramatic results have
already been implemented; future gains in air quality will
depend on a combination of incremental control strategies.
Each strategy may affect air quality slightly, but in
combination, they can significantly improve it. For example,
controlling LDTs as well as LDVs at high altitudes would
provide greater benefits. Also, in presenting the results of
air quality models, the absolute values of ambient pollutant
concentrations or violations of the air quality standards are
subject to greater error than are the relative differences
among the control strategies being analyzed.
Figures VII-4 and VII-5 show estimated impacts of scenario
2 and the base scenario on the air quality of major
high-altitude cities. Figure VII-4 shows that scenario 2 could
slightly reduce the maximum ozone concentrations beyond those
achieved by the base scenario (generally being about a 1
percent improvement for some of the years studied). Figure
VII-5 shows that in 1989 and 1990, one less violation of the
ozone NAAQS is projected to occur under scenario 2 for the
high-growth no-l/to case. With I/M, scenario 2 does not appear
to provide any additional reduction in NAAQS violations.
The differences in ambient CO concentrations resulting
from the two scenarios, however, are more significant than
those shown for ozone above. The benefits of implementing
scenario 2 are projected to be somewhat larger and occur with
greater frequency. Figure VII-4 shows that in essentially all
years, there is a 1 to 3 percent improvement under both growth
assumptions, with and without I/M.
The projected number of NAAQS violations for CO are shown
in Figure VII-5. Implementing scenario 2, without I/M, could
reduce the number of violations under the low-growth case by
one in 1986 and 1989. If higher growth rates without I/M are
assumed, scenario 2 may provide 1 to 3 fewer violations in 1987
through 1990. Under the high growth, with I/M case, two less
violations are shown in 1986 and one less in 1987.
To summarize the discussion thus far, the air quality
models show scenario 2 is projected to have a positive impact
on air quality in high-altitude regions, although the impact is
relatively small. This is not surprising, however, because
scenario 2 is an incremental control strategy and needed
improvements in air quality will only come by combining several
incremental controls (e.g., light-duty trucks, each having a
small benefit of its own).
-------
VII- 17
FIGURE VII-4
Average Percent Reduction in Expected
CO and Ozone Concentrations from 1979
Base Year in Selected High-Altitude Cities [a]
80
70
60
50
40
30
20
10
03
Low Growth
Scenario 2 and Base Scenario
- — T "^Scenario 2
Base Scenario
y
86 87 88 89 90
YEAR
93
95
O
3
Q
LU
O
DC
LU
Q.
80
70
60
50
40
30
20
10
03
High Growth
Scenario 2 and Base Scenario
Scenario 2 and Base Scenario
¦
86 87 88 89 90
YEAR
93
95
80
70
60
50
40
30
20
10
Scenario 2
Base Scenario
Scenario 2
.Base Scenario
CO
Low Growth
i i i
86 87 88 89 90
YEAR
i l
93
95
CJ
D
G
LU
QC
LU
o
DC
LU
Q.
80
70
60
50
40
30
20
10
Base Scenario
Scenario 2
Scenario 2
Base Scenario
CO
High Growth
i i
a.
86 87 88 89 90
l I
93
95
YEAR
[a] OZONE VALUES REFLECT WORST CASE HC/NOx RATIO
OF AMBIENT CONCENTRATIONS.
Key:
With l/M
Without l/M
-------
VII- 18
FIGURE VII-5
Number of CO and Ozone NAAQS
Violations in High-Altitude Cities [a]
5 r
03
Low Growth
Scenario 2 and Base Scenario
J.
Scenario 2 and
Base Scenario
86 87 88 89 90
YEAR
93
95
5 r
CO
z
O 4
t-
<
_I
2 3
o
cc
LU
CQ
03
High Growth
/ Base Scenario
Scenario 2 \
Scenario 2 and Base Scenario
a.
86 87 88 89 90
YEAR
93
95
45^
35
25
Scenario 2 and'
Base Scenario
CO
Low Growth
Base Scenario
¦ \ Scenario 2
86 87 88 89 90
I I
93
95
YEAR
cc
UJ
00
45
CO
o 35
<
_J
o
> 25
15
\
\\
\\
- n
u
\\
CO
High Growth
Base Scenario
Scenario 2
Base
Scenario^
Scenario 2
a.
86 87 88 89 90
YEAR
93
95
[a] OZONE VALUES REFLECT WORST CASE HC/NOx RATIO
Key:
With l/M
----- Without l/M
-------
VII-19
Before making a final decision on the merits of scenario
2, it remains to be shown whether even the small benefits of
this scenario are needed to improve the air quality of
high-altitude regions. To evaluate this question, the NAAQS
attainment dates and trends in high-altitude air quality will
be used. As shown in Figure VII-5, high-altitude cities appear
to require additional emission reductions in future years since
the NAAQSs for ozone and CO are never projected to be achieved
by the 1987 statutory deadline even when scenario 2 is
implemented with an inspection/maintenance program. Also, the
trend lines in Figure VII-4 show that by about 1990 to 1993,
improvements in ozone and CO pollution often may have generally
slowed and in some cases may have essentially stopped so that
air pollution levels might not continue to decline and could
possibly increase under some conditions. Specifically, ozone
may be a problem in some areas since compliance with the NAAQS
is projected to be achieved in all high-altitude cities only
under the low-growth, with-l/M strategy (Figure VII-5). The
NAAQS for CO is predicted to be attained between 1988 and 1990
under the low-growth cases, but for the high-growth cases,
attainment may be postponed until 1990 with I/M and between
1990 and 1993 without I/M (Figure VII-5). However, as alluded
to above, the CO trend lines in Figure VII-4 for the
high-growth cases generally show a reduction in the rate at
which additional improvements in emission concentrations for
subsequent years are achieved, especially in the base
scenario. Because of this and the uncertainty associated with
the projections, continued compliance with the NAAQS may not be
assured. Therefore, additional emission control beyond that
provided by the base scenario appears to be justified at this
time.
The discussion contained in Chapter V pointed out the
uncertainties involved with the computer models which produced
the air quality projections in this report. These
uncertainties may cause the projections to be better or worse
than those documented here. In particular, there are a number
of identifiable reasons why the analysis may underestimate the
number of violations of the NAAQS. To simplify the analysis,
EPA assumed that all 1984 and later model year vehicles
operated at high altitude meet the appropriate emission
standards in each scenario. While this is valid for the vast
majority of vehicle miles traveled (VMT), it is reasonable to
expect that some mileage would occur from unregulated
low-altitude vehicles operated by visitors or permanent
residents who had moved from low-altitude areas. Many of these
uncontrolled low-altitude vehicles will pollute significantly
more than high-altitude vehicles which are subject to
regulation. Also, the air quality models upon which this
analysis is based assumed an ambient temperature of 75°F. In
reality, the average temperature in high-altitude areas during
periods of unhealthy air is much lower. At lower ambient
temperatures the amount of pollution from motor vehicles
-------
VII-20
increases, especially CO emissions. This potential increase in
emissions is not accounted for in this analysis. Therefore,
the number of NAAQS violations in high-altitude areas could be
somewhat greater than shown in Figure VII-5.
In summary, the data contained in Figures VII-4 and VII-5
indicate that additional cost-effective forms of pollution
control may be neeaed to attain or assure compliance with the
NAAQS in high-altitude regions of the country. However, as
indicated in the previous paragraph and in Chapter V, the air
quality projections presented above could be subject to error
in either direction. Detailed analysis of the air quality of
those areas just above or below the NAAQS would be necessary to
confirm the need for additional controls such as those provided
by scenario 2.
VII. CONCLUSIONS
The analyses in this report indicate that implementing an
alternative program similar to scenario 2 will provide
additional emission control for high-altitude regions in a
cost-effective manner. For this reason, scenario 2 merits
further consideration regarding LDGTs and LDDs.
-------
Chapter VIII
Control of Light-Duty Trucks
I. INTRODUCTION
This chapter analyzes the effect of h igh-altitude
standards for light-duty gasoline-fueled trucks (LDGTs).
Specifically, imp lenien ting fixed-point statutory standards
(scenario 2) relative to fixed-point proportional standards
(base scenario) will be examined with regard to technology,
economic impact, enviromental impact, and cost effectiveness.
As describee in Chapter VII, scenario 2 was found to be a
viable alternative control strategy for LDGVs. The statutori
standards of this scenario for light-duty trucks will be
assumed to be implemented in 1984 as was the case for
light-outy vehicles.
It shoulo be noted that high-altitude LDT standards are
not mandated in section 206 of the Act as are the high-altitude
LDV requirements. If LDT standards at high altitude are
warranted, however, EPA may establish such standards under the
general rulemaking authority of section 202.
II. TECHNOLOGY ASSESSMENT
For light-duty gasoline-fueled vehicles (LDGVs) under
scenario 2, EPA projected that all vehicles v*ith nonfeedback
(open-loop) control would require three fixed-step aneroids and
a recalibration of engine-related control parameters. Also,
all LDGVs would require larger charcoal canisters for
evaporative HC emission control (Chapter III).
There are two basic similarities between these LDGVs ana
all LDGTs which lead to the conclusion that these technologies
should appli to LDGTs also: 1) both LDVs and LDTs certify
using the same test procedures, and 2) current emission control
technology on man^ LDGVs at low altitude, (i.e., nonfeedback
controlleo systems) is very similar to that needed for LDGls to
meet their more stringent 1984 low-altitude standards.[1]
Therefore, LDGTs will require a recalibration of engine control
parameters ana the addition of three fixed-step aneroios to
meet the high-altitude exhaust emission standards, since EPA
estimates that essentially all such vehicles will be equipped
with nonfeedback systems at low-altitude in 1984 and later
model ^ears. Similarly, complying with the statutory-
evaporative HC standard will require a 25 percent increase in
charcoal loading of the storage canister.
III. ECONOMIC IMPACT
As with LDGVs, the cost of implementing alternative high-
altitude standards tor LDGTs will be determined incrementally
to the base scenario (i.e., the fixed-point proportional
standard) (Table II-4, Chapter II). Under the base scenario,
LDGTs will have the same control hardware as shown for the base
scenario in Table VI-2 of the Economic Impact chapter for
-------
V1II-2
nonfeedback LDGVs. In that table, nonfeedback controlled
vehicles required engine recalibrations in addition to one
fixed-step aneroid.
The remainder of this section will follow the same outline
as the Economic Impact chapter for LDGVs. Thus, the cost to
manufacturers, the cost to users, the aggregate costs, and the
socioeconomic impact will be examined separately.
A. Cost to Manufacturers
The cost to manufacturers has two elements: variable
costs and fixed costs.
1. Variable Costs
The only variable cost resulting from 1984 high-altitude
exhaust emission standards is the cost of two additional
aneroids for all LDGTs. The estimated cost of these two
devices for LDGVs, based on discussions in Chapter IV, is
$14-18. The additional production of aneroids for LDGTs over
that of LDGVs lower costs, because of the economies of scale.
To be conservative, however, the same costs based on LDGV
production volumes will be used here for LDGTs.
The variable cost for the high-altitude evaporative
standards is also the same as that previously estimated for
LDGVs in Chapter V. Complying with the statutory evaporative
HC standard is estimated to increase the total hardware cost
for LDGTs by $2-3 to a total of $16-21 per vehicle.
2. Fixed Costs
Fixed costs include development and certification costs.
a. Development costs. Development costs for scenario 2
are estimated incrementally from the base scenario. Under the
base case, all 1984 LDGTs have to be redesigneo at low altitude
to comply with the more stringent HC and CO standards
implemented in that mooel year. All nonexempt LDTs sold at
high-altitude (it is assumed LDTs require the same percentage
exemption as LDVs), would also have to undergo development to
meet the new proportional standards which would accompany the
more stringent low-altitude standards.
Scenario 2 would have no effect on low-altitude
development efforts since high-altitude control technology
(aneroids) should not affect low-altitude engine calibrations
(a more complete discussion is presented in Chapter V),
Scenario 2, however, could affect beneficially high-altitude
development.
-------
VIII-3
In the preceding paragraph, it was assumed that LDTs will
require the same exemptions as LDVs (i.e., 5 percent for the
base scenario and 15 percent for scenario 2). The greater
number of exemptions in scenario 2 implies that fewer engine
families will undergo development at high altitude than in the
base scenario, based on the relationship between engine
families and vehicle sales outlined in Appendix III. Rather
than account for this potential savings, no development cost
increment ($0) will be assumed in this analysis to be
conservative.
b. Certification. The incremental certification costs
for LDGTs in 1984 is the additional cost over what is assumed
to take place under the base scenario. Under this scenario, it
is projected that all nonexempt engine families will have to be
recertified at high altitude due to the new proportional
standards beginning in 1S84. As previously discussed with
regard to development costs, the additional exempted engine
families in scenario 2 would reduce the certification costs in
relation to those already occurring in the base scenario. To
be conservative, this potential certification savings is not
accounted for in this analysis. Instead the incremental
certification cost is considered to be $0, as it was for
incremental development costs.
B. Cost to Users
The sticker price increase for users of high-altitude
vehicles is simply the control hardware cost since there is no
increment in development or certification associated with
scenario 2. This cost is about $16-21 per high-altitude LDGT.
The net vehicle cost of scenario 2 includes not only the
sticker price increase but also any increment in maintenance or
fuel economy. There should be no additional maintenance
requirements as a result of implementing the control hardware
described above. However, the addition of two aneroids could
result in a fuel economy benefit. As with LDGVs, the fuel
economy benefit is estimated to be about 2 percent (see Chapter
III for the derivation of this fuel economy benefit). For
LDTs, a 2 percent fuel economy benefit represents a cost
savings of about $85 per vehicle. This is based on an average
LDGT life of 120,000 miles accumulated over a period of 12
years,[2] a fuel economy estimated to be the proposed 1985 CAFE
standard of 21 mpg, an unleaded gasoline price of $1.30 per
gallon, ana discounting the resulting fuel savings (at 5
percent per year) to year of purchase. Again referring back to
the LDGV economic analysis contained in Chapter V, the
sensitivity of scenario 2 to the estimated fuel economy benefit
was so great that a cost range was used for the net vehicle
cost. This sensitivity is even more significant with regard to
LDGTs since the fuel economy benefit applies to all light
trucks sold at high altitude instead of only a portion of the
-------
V1II-4
fleet. Therefore, the net vehicle cost for LDGTs will be
estimated in a similar manner. The lower limit of the range is
found by including the fuel economy benefit and the upper limit
is found by excluding it. This results in a potential savings
of $69 or a potential cost of $21 per vehicle.
C. Aggregate Costs
The projected number of 1984-88 LDGTs sold in areas above
4,000 feet (excluding California) is approximately 409,000.
This figure was derived from nationwide sales estimates made by
Data Resources Corporation,[3] and the same estimates of
population distribution according to altitude used to determine
LDGV emissions in Chapter IV. These supporting data can be
found in Table VIII-1.
The aggregate costs of the high-altitude standards apply
to the first five years of this regulation or 1984-88.
However, the present value in 1984 will be used, so that the
cost can be compared on an equivalent basis to the aggregate
cost for LDGVs. Based on a 1984-88 high-altitude LDGT
production of 409,000, the 5-year aggregate cost of scenario 2
varies from a potential savings to about $7 million.
The costs for LDGTs are summarized in Table VIII-2.
D. Socioeconomic Impact
The price increase of a light-duty truck was estimated to
be a maximum of $21. Given the range of variability in the
analyses, this is comparable to the maximum sticker price
increase of $15 estimated for LDGVs under scenario 2 (see
Chapter V). The LDGV analysis concluded that such a sticker
price increase should not affect a dealer's sales or a
consumer's ability to purchase a vehicle. The same conclusion
should hold here for LDGTs, because both the absolute cost
increase and the percentage increase in the purchase price of
these vehicles is comparable to that estimated for LDGVs.
Thus, the sticker price increase for LDGTs should not
significantly affect sales.
There are no incremental capital costs associated with
implementing scenario 2. Manufacturers of LDGTs will already
incur such expenses under the base case because of the new-
proportional standards taking effect in 1984.
Thus, as similarly analyzed for LDVs, the increase in
costs due to this regulation should have little socioeconomic
impact, if any, on manufacturers, dealers, and users.
-------
Year
1984
1985
1986
1987
1988
VIII-5
Table VI1I-1
Light-Duty Truck Sales Projectionsl3,4]
Nationwide
LDT Sales[a]
(IP** trucks)
3.24
3.36
3.40
3.46
3.54
Nationwide
LDGT Salesta]
(10^ trucks)
2.93
2.98
2.93
2.89
2.92
LDGT Sales[b]
Above 4,000
Feet (trucks)
82,000
83,000
82,000
81,000
81,000
5-Year Sum
17.0
14.65
409,000
[a] Light-duty truck.
Light-duty gasoline-powered truck.
[b] These values were determined by assuming that the number of
new vehicle sales is directly proportional to population
distribution. Since 3.1 percent of the U.S. non-California
population resides above 4,000 feet, this is also the
fraction of nationwide non-California sales that is expected
to occur in high-altitude areas of the country.
-------
VIII-6
Table VIII-2
Summary of Costs for LDTs ($1981)
Best
Estimate Incremental Aggregate
Fuel Net Capital Costs to Costs[d]
Hardware Development Economy [ a] Total[b] Manufacturers[c] (millions)
$16-21 $0 -85 $-69 to 21 $0 up to $7
[a] Because of the uncertainty associated with this fuel economy
estimate, its use is speculative at this time.
[b] No maintenance or certification costs expected.
[c] Includes 15 percent cost of capital.
[d] Basec. on a production of 409,000 for high-altitude sale
between 1984-88. Present value is in 1984.
-------
VI1I-7
IV. ENVIRONMENTAL IMPACT
This section explores the effects of scenario 2 on LDGT
emissions and the resultant air quality. The same basic
methodology used in Chapter III to describe the environmental
impact of LDGV control will be followed below.
A. Total Emissions
1. Methodology
As was the case with the LDGV analysis, total lifetime
emissions will be determined from LDGTs sold over a 5-year
period. For reasons discussed earlier, this time increment is
1984-88. To determine these emissions, three factors were
multiplied together: 1) the number of miles the average LDGT
is driven in its lifetime, 2) emission factors (the amount of
pollutant emitted per truck per mile of travel), and 3) the
number of trucks sold above 4,000 feet (the initial control
altitude for scenario 2). These factors are discussed below.
For the purposes of this study, the average lifetime
mileage for light-duty trucks is 120,000 miles. [2] To
determine lifetime emissions, emission factors were determined
for trucks that were halfway through this average lifetime
(i.e., at 60,000 miles). It is important to remember that
scenario 2 applies only to those vehicles and trucks sold above
4,000 feet and that the high-altitude emission data available
were based on tests conducted at Denver (elevation 5,300
feet). Thus, the average lifetime emission rates in this study
for LDGTs sold above 4,000 feet were based on tests at 5,300
feet.
As mentioned earlier, 100 percent of the future
high-altitude LDGT fleet is expected to use open-loop emission
control technology. Table VIII-3 contains the LDGT emission
factors for the base case and scenario 2 baseo on this
technology. Table VIII-3 also contains evaporative HC emission
factors for LDGTs. Light-duty gasoline-fueled trucks and
vehicles have the same evaporative hydrocarbon emission
characteristics ano, thus, the same emission standard under the
1982 and 1983 interim high-altitude emission regulations.
Therefore, the evaporative HC emission factors developed in
Chapter IV for LDGVs under the base scenario and scenario 2
also apply in the case of LDGT evaporative HC emissions.
By combining the mileage, emission factor and sales data
described above, the total exhaust and evaporative emissions
from 1984-88 LDGTs in areas above 4,000 feet can be
determined. These are presented in Table VIII-4 for both the
base scenario and scenario 2.
-------
VIII-8
Table VIII-3
Emission Rates for 1984-88
Light-Duty Gasoline Trucks at 5,300 Feet[a]
Pollutant Base Scenario Scenario #2
HC Zero-Mile Emission 0.78 0.63
Level (g/mi)
Deterioration Rate 0.14 0.14
(g/mi/10,000
miles)
Average Lifetime 1.62 1.47
Emission level
(g/mi)
CO Zero-Mile Emission 9.85 7.13
Level (g/mi)
Deterioration Rate 1.35 1.35
(g/mi/10,000
miles
Average Lifetime 17.95 15.23
Emission level
(g/mi)
NOx Zero-Mile Emission 1.26 1.26
Level (g/mi)
Deterioration Rate 0.04 0.04
(g/mi/10,000
miles)
Emission Level 1.5 1.5
(g/mi)
(g/mi)
Evap. Zero-Mile Emission 0.13 0.10
HC Level (g/mi)
Deterioration Rate 0 0
(g/mi/10,000
miles)
Average Lifetime 0.13 0.10
Emission Level
(g/mi)
[a] Emission Rate = Zero-Mile Level + (cummulative
Mileage/10,000), (deterioration rate); cumulative mileage
equals 60,000 miles, one-half of lifetime.
-------
Pollutant
HC
CO
NOx
Evap. HC
Total HC
VIII-9
Table VIII-4
Lifetime Emissions from 1984-68 LDGTs
Sold Above 4,000 Feet
(lp3 metric tons)
Base
Scenar io
79.5
880.9
73.6
6.38
85.88
Scenario 2
72.1
747.5
73.6
4.91
77.01
Reductions
(base scenario
minus scenario 2)
7.4
133.4
0
1.47
8.87
-------
VIII-10
2 . Discussion of Results
From Table VIII-4/ it can be seen that scenario 2 is
predicted to reduce total HC emissions from 1984-88 LDGTs by
approximately 8/870 metric tons, compared to the base
scenario, similarly, CO emissions are predicted to decrease by
roughly 133,400 metric tons. no NOx reductions would be
realized since the high-altitude NOx standards under scenario 2
would be the same as under the base scenario.
Referring back to chapter iv, one sees that scenario 2 is
shown to lower HC and CO emissions from LDGVs sold above 4,000
feet by approximately 13,800 metric tons and 331,000 metric
tons, respectively. Thus, combining the effects of scenario 2
on LDGVs and LDGTs yields total 5-year HC reductions of roughly
22,670 metric tons and total CO reductions of approximately
468,400 metric tons. as can be seen, roughly a third of the
total HC and CO emission reductions result from LDGT control.
The consequences of controlling LDGT emissions on air quality
in high-altitude areas are discussed in the following section.
B. Air Quality
The air quality discussion in this chapter is an extension
of that found in Chapter IV for LDV control; both use the same
models, study sites, and growth rates. The only difference is
the addition of emission controls on LDTs (both gasoline and
diesel) to the base scenario and scenario 2. The LDV emission
factors are unchanged. Due to the large number of composite
emission factors (average emission factors for a given vehicle
category in a given year), they are included in reference 4.
Tables VIII-5 through VIII-8 present the results of the
air quality analysis in two ways: 1) percent reductions in
pollutant concentrations from a baseline year (1979), and 2)
changes in the projected number of NAAQS violations. To
clarify the air quality consequences of controlling LDTs at
high altitude, the tables also include the effects of LDV
control only. The inclusion of this information allows LDT
standards to be evaluated by determining: 1) the air quality
differences between the base scenario and scenario 2 with and
without truck control, and 2) the resulting air quality change
when more stringent truck control is included with LDV
standards in scenario 2. Both of these methods of evaluation
are discussed below.
Tables VIII-5 and VIII-6 show that scenario 2 with truck
control may result in lower ambient CO concentrations and
slightly fewer violations of the CO NAAQS than under the base
scenario with truck control. These benefits are first apparent
in 1986 where one less violation is expected for the high
-------
Table VIII-5
Averaye Percent Reduction in Expected
Second Highest 8-Hour CO Concentrations from 1979
Base Year in Six lli^h-Altituue Cities
(Low and Iligh Growth) [aj
Year
1986
199U
1995
With I/M
W/0
I/M
Witli I/M
W/O
I/M
With I/M
W/O
I/M
Scenar io
Low Hiyh
Low
High
Low High
LiOW
ii iyh
Low iliyil
Low
11 iyh
base -
55 49
46
38
12 6 6
64
56
79 72
73
u4
With Truck
Control
Base - 55 48 45 37 71 65 63 54 78 7U 72 o2
Without
Truck
Control
#2 - With 56 5U 47 39 73 u7 66 57 81 75 76 67
Truck Con-
trol
#2 - With- 55 49 46 38 72 65 64 56 8U 73 74 o5
out Truck
Control
[a] The cities examined are Denver, Colorado Springs, Ft. Collins, Greeiey,
Albuquerque, and salt Lake City.
-------
Table VIII-6
out Truck
Control
Nuiaber of Violations of CO WAA^S
in Six High-Altitude Cities (low and high growth) LaJ
iy86
Year
lyyO
iyy5
Witu
I/M
W/0
I/M
Witn
I/M
w/o
1/14
With
x/fri
W/O
1/L'l
scenar io
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Base -
With Truck
Control
7
16
20
43
0
0
0
3
0
U
0
0
Base -
Without
Truck
Control
7
18
21
43
U
0
0
4
U
U
U
U
#2 - With
Truck Con-
trol
5
i5
18
3y
0
0
U
2
U
U
0
0
#2 - With-
7
iG
20
43
0
0
0
3
U
U
0
0
[a ] The cities examined are Denver, Colorado Springs, Ft. Collins, Greeley,
Albuquerque, and Salt Lake City.
-------
Table VI1I-7
Averaye Percent reduction ia Expected Maximum
1-llour Ozone Concentrations froio 1979 base Year
in Denver and salt Lake City (low and hiyh yrowthHaJ
Year
1986 I5¥D 1*95
With I/H W/O I/H With I/H W/0 I/fl With I/H W/O I/M
Scenar io Low 11 iyh Low High Low Hiyh Low liiyh Low Hiyh low Uiyn
Dase - 24-32 22-28 21-27 18-24 29-38 2b-33 25-34 22-29 28-37 24-31 20-35 21-28
With Truck
Control
Base - 24-32 22-2U 21-27 18-24 29-37 25-33 25-33 22-28 28-36 24-31 20-34 21-2U
Without
Truck
Control
«2 - With 24-33 22-29 21-27 18-24 29-38 25-33 26-34 22-29 29-37 24-32 20-35 22-29
Truck Con-
trol
42 - With- 24-32 22-29 21-27 18-24 29-37 25-33 25-34 22-29 2U-37 24-31 2o-35 2i-28
out Truck
Control
[a) Note that a ranye of values are reported. These reflect two different ratios ot riC/NOx
ambient concentrations, as discussed in Chapter IV. Values from tne nx^.ier ratio ure iiateu
f irst.
-------
Table VIII-8
Nuuber of Violations of Oioone NAA^S
iu Denver and Salt Lake City (low and hi^n growtn)[d]
Year
T9ul> I¥yo T5W
With I/M
W/O
I/M
With I/M
W/O
I/M
With I/M
W/O
I/M
Scenar io
Low
Iliyh
Low
High
Low
High
Low
High
Low
iii^ii
Low
liic|h
Base -
With Truck
Control
0-1
0-1
0-2
1-2
0-0
0-1
0-1
0-2
U-l
u-l
U-l
1-2
Base -
Without
Truck
Control
0-1
0-1
0-2
1-2
0-0
0-1
0-1
0-2
0-1
0-1
U-l
1-2
#2 - With
Truck Con-
trol
0-1
0-1
0-2
1-2
0-0
0-1
0-1
0-1
0-1
0-1
U-l
1-2
%2 - With-
out Truck
Control
0-1
0-1
0-2
1-2
U-0
0-1
0-1
U-l
0-1
0-i
u-l
1-2
[a] Note that a ran^e of values are reported. These reflect two uifrerent ratios
of llC/NOx ambient concentrations, as discussed in Chapter XV. Values from cue
lower ratio are listed first.
-------
VIII-15
growth with I/M case and two less violations are expected for
the low-growth with I/M case (Table VIII-6). For 1990 and
beyond, no violations are expected due to the combined effects
of more stringent low-altitude standards and I/M. Chapter IV
showed that without truck control the number of violations
under scenario 2 and the base scenario was the same until 1988
with I/M. Thus, including light trucks under scenario 2
appears to yield fewer violations two years sooner than would
otherwise be possible.
Without I/M, the CO benefits of controlling trucks under
scenario 2 first appear in 1986, when four less violations are
expected with high growth and two less violations are expected
with low growth. Similarily, without truck control, the total
number of violations appear greater for scenario 2 in this year
under either the low- or high-growth cases, in 1990, one less
violation is projected for the high-growth without-l/M case
under scenario 2 than under the base scenario when truck
control is considered. Air quality benefits without LDT
control also appear in this year but again the total number of
violations is projected to be greater than in the with-truck
control option. Looking only at scenario 2, adding truck
control should result in four fewer violations during 1986 and
one less in 1990 under the high-growth case and two less
violations in 1986 under the low-growth case. Thus, adding
truck control to scenario 2 resulted in a larger projected air
quality benefit than when only light-duty vehicles were
controlled. This was true whether or not I/M was present.
With regard to ozone air quality, Table VIII-7 shows that
adding LDT control to scenario 2 could lower ambient ozone
concentrations below what the base scenario would provide.
With I/M, in 1986 and 1995 the additional reduction could be 1
percent for the low- and high-growth cases. Without I/M, the
reduction is 1 percent more in 1990 for scenario 2 with truck
control under the low-growth case than for the base scenario.
Adding more stringent truck control to scenario 2 when I/M is
assumed could reduce ozone concentrations by 1 percent in each
year studied under the low-growth case and by 1 percent in 1995
under the high-growth case. When no I/M program is assumed,
implementing more stringent truck control in this scenario
should result in a benefit during 1990 of up to 1 percent under
the low-growth case. In 1995, the benefit could be 1
additional percent under high growth. It can be seen from
Table VIII-8 under both growth cases, that adding truck control
did not further reduce the number of NAAQS violations occuring
under scenario 2 compared to the base scenario in the with-I/M
situations. Without I/M, there could be one less violation in
1990 under high growth. Since, however, there appears to be no
difference in ozone NAAQS violations when more stringent LDT
standards are added to scenario 2, this benefit must be
attributed to LDV control.
-------
VIII-16
C. summary of Air Quality Information
The analysis in this section predicted that adding LDGT
control to scenario 2 would lead to additional reductions in HC
emissions beyond those of the base scenario by approximately
8,870 metric tons from the 1S84-88 LDGTs sold above 4,000
feet. It would also lower CO emissions from these trucks by
133,400 metric tons more than the base scenario over their
lifetime.
Fewer violations of the CO NAAQS are projected when LDT
control is added to scenario 2 under both low and high growth,
but because of I/to, no violations under either the base
scenario or scenario 2 are projected beyond 1990. Vvithout I/H,
adding truck control could also result in fewer violations of
the CO NAAQS than would be expected under scenario 2 without
truck control. This is true under both low and hicjh-growth
rates. Attainment of the CO standard is not projected until
after 1S90 when high growth and no I/to are assumed. Adding LDT
control to scenario 2 ciid not result in fewer violations of the
NAAQS for ozone with or without I/M. Thus, any cost-effective
HC control strategy should be seriously considered as a partial
means of attainment.
As was mentioned earlier concerning the air quality
projections contained in Chapter IV, the air quality
projections made here must be used with some caution. touch of
the input data to the model, growth rates for vehicle-miles
traveled (VtoT), VtoT breakdown by vehicle class, and average
speed, are based on national averages and not local data.
Also, relatively simple models have been used, modified
rollback and EKKA. These models require relatively simple oata
which were easily available for all the areas of concern. Air
quality projections using input data specific to individual
locations and the use of more sophisticated iriooels, as is often
done by local or state agencies, could result in more accurate
projections. Also, as has been mentioned earlier, the emission
factors projected for future years are not firm, especially for
those equipped with feedback controls. Changes in these
factors in the future could go in either direction. Thus, the
number of violations of the NAAQS projected in any given year
must be seen as an estimate and could actually occur a number
of years earlier or later. Since the need for statutory
control is, unfortunately, based on whether or not
high-altitude areas achieve the NAAQS, a conclusive
determination of the need for statutory control over
proportional control cannot be made at this time.
V. COST EFFECTIVENESS
Results from the cost and total emission sections of this
chapter will be combined to determine the cost effectiveness of
adding LDGT control to scenario 2. The same procedure outlined
-------
VIII-17
in Chapter V to determine the cost effectiveness of LDGV
control is followed in this section. The results of this
analysis are shown in Table VIII-9 along with the incremental
cost effectiveness of LDGV control for scenario 2 as determined
in Chapter V. The incremental cost effectiveness of combined
LDGV and LDGT control is also presented in Table VIII-9.
As this table shows, adding control of light-duty
gasoline-fueled trucks to scenario 2 makes it a slightly more
cost-effective approach. Under "worst case" assumptions, the
cost of removing HC is reduced from $575 per metric ton for
LDGVs only to $535 per metric ton for both classes of
vehicles. The cost effectiveness of CO removal is unchanged
under worst case assumptions.
Table VIII-10 lists the cost effectiveness of other
control strategies already adopted. (This same table also
appears in Chapter VI.) From this table, it can be seen that
when LDGT control is added to scenario 2, it becomes more cost
effective than three of the other existing HC strategies and
two of the other existing CO strategies.
-------
Table VIII-9
Incremental Cost Effectiveness of Scenario 2
Emission Incremental
,Reductions Cost Effectiveness
Cost[a] (10~3 metric tons) (dollars/metric ton)
Vehicle
(dollars
per vehicle)
per
vehicle
HC
CO
Type
Low
High
HC
CO
Low High
Low High
LDGV
-11
15
13 .0
239.9
neg. 575
neg. 30
LDGT
-69
21
21.7
326 .2
neg. 485
neg. 30
LDGV and
-24
16
15.0
259 .7
neg. 535
neg. 30
LDGT[b]
[a] 1981 dollars discounted to 1984. The low estimates are based on the
inclusion of tentative fuel-economy benefits.
[b] Combined LDGV and LDGT costs were obtained by weighting the individual
LDGV and LDGT costs according to sales above 4,000 feet.
-------
VIII-19
Table VIII-10
Cost Effectiveness Comparison With
Other Emission Control Strategies
(1981 dollar per metric ton)
Control Program
LDV Statutory[6]
Standards
LDV I/M7]
LDT 1984[1]
Standards
HDE 1984
Standards!8][b]
(gasoline)
(oiesel)
Baseline
Emissions[a]
HC 0.9
CO 15
HC 1.7
CO 18
HC 1.5
CO 25
HC 1.5
CO 25
Emissions
After Control
Strategy
Implemented
HC 0.41
CO 3.4
HC 0.8
CO 10
HC 1.3
CO 15.5
HC 1.3
CO 15.5
Cost
Effectiveness
HC
734
640
195
305
325
CO
67
58
15
10
Motorcycle
Neg.
Standards!9]
HDG Lvap.[10][d]
Interim 1982-83
HA Standards!11]
HC 9
CO 34.67
HC 1.8
HC 1.47 (cars)
4.19 (trucks
CO 16.23 (cars)
73.02 (trucks)
HC 8-22.5[c] 582
CO 27.4
HC 0.17 200
HC 1.33 (cars) 393
3.78 (trucks)
CO 13.21 (cars)
55.65 (trucks)
12
La] Emission levels are in g/nii except for the HDE 1984 standards
which are in grams per brake horsepower-hour.
lb] The baseline and after-control strategy emission values were
based on different test procedures (see ref. [9]).
[c] Sliding scale based on engine oisplacement (cubic
centimeters).
[d] The evaporative standard is in terms of g/test and converted
to g/mi here to facilitate comparison.
-------
VIII-20
References
1. "Regulatory Analysis and Environmental Impact of Final
Emission Regulations for 1984 and Later Model Year Light-Duty
Trucks," U.S. EPA, OANR, OMS, ECTD, Kay 1980.
2. "Average Lifetime Periods for Light-Duty Trucks and
Heavy-Duty Vehicles," U.S. EPA, OANR, OMS, ECTD, Passavant, Glenn
W., November 1979.
3. Data Resources, Long-Term Review, Fall 1980.
4. "Air Quality Analysis for the 1S84 High-Altitude Report
to Congress," U.S. EPA, March 1982.
5. "Interagency Task Force on Motor Vehicle Goals Beyond
1980," U.S. EPA, Department of Transportation, Prepared by the
Panel on Automotive Manufacturing and Maintenance, March 1976.
6. "Update on Cost Effectiveness of Inspection and
Maintenance," U.S. EPA, OANR, OMS, ECTD, IMS, EPA-AA-IMS/81-9,
April 1981.
7. "Regulatory Analysis and Environmental Impact of Final
Emission Regulations for 1984 and Later Model Year Heavy-Duty
Engines," U.S. EPA, OANR, OMS, ECTD, SDSB, December 1979.
8. "Environmental and Economic Impact Statement - Exhaust
and Crankcase Regulations for the 1978 and Later Model Year
Motorcycles," U.S. EPA, OANR, OMS, ECTD, SDSB, 1976.
9. "Regulatory Analysis for the Proposed Evaporative
Emission Regulations for Heavy-Duty Vehicles," U.S. EPA, OANR,
OMS, ECTD, SDSB, April 1980.
10. "Environmental and Economic Impact Statement for the
1982 and 1983 Model Year High-Altitude Motor Vehicle Emission
Standards," U.S. EPA, OANR, OMS, ECTD, SDSB, October 1980.
-------
Chapter IX
Control of Light-Duty Diesel Gaseous Emissions
I . INTRODUCTION
This chapter addresses the effects of high-altitude
gaseous emission standards on light-duty diesel-powered
vehicles and trucks (LDDVs and LDDTs). As was the case with
the discussion of light-duty gasoline-fueled trucks in the
previous chapter, this analysis focuses on comparisons between
scenario 2 and the base scenario. This format will be followed
since scenario 2 was found to be the most cost-effective of the
control strategies which were analyzed in this report as
alternatives to the base scenario for further reducing
emissions from light-duti gasoline-fueled vehicles in
high-altitude areas (Chapter VII).
To remain consistent with the previous analyses, the years
being studied in this chapter with regard to costs and total
emissions are 1984-86 for both LDDVs and LDDTs. In reality, it
may be desirable to delay additional high-altitude control one
year beyond the first year of the more stringent high-altitude
standards for light-duty gaseous vehicles (LDGVs), because the
I.0 g/mi NOx standard ma^ be waived up to 1.5 g/mi for LDDVs
until 1985 (as allowed in section 202(b)(6)(B) of the Act). In
fact, such NOx waivers have already been granted for many 1984
model year LDDVs. Delaying the implementation date of the more
stringent high-altitude standards for one year would prevent
manufacturers from having to design, build, and certify
vehicles to meet high-altitude standards that would be in
existence for onli one year. However, such a delay would
require amending the Act. In any event, for the purposes of
this analysis assuming that the LDDV statutory standards
(including 1.0 g/mi NOx) begin in 1984 instead of 1985 is
insignif icant.
II. TECHNOLOGY ASSESSMENT
Veri little actual data exist with which to quantify the
types of technology which may be needed by LDDVs and LDDTs to
comply with the statutory standards of scenario 2. Also,
unlike gasoline engines, the effect of decreasing air density
cannot be compensated for completely in diesel engines.
Gasoline engines burn a homogeneous air/fuel mixture of fairly
constant proportions. When air density decreases, which
reduces the mass of air in the cylinder, the air/fuel ratio can
be retained at its optimum value by either opening the throttle
to allow more air, or by metering less fuel. This works well
until the power requested of the engine approaches its maximum
and power enrichment sets in.
A diesel engine burns a heterogeneous air/fuel mixture
with no throttle to control the air flow. Thus, at high
-------
IX-2
altitude, less air will enter the combustion chamber at all
times unless the engine is turbocharged. Generally, this
reduction in cylinder air will not affect combustion
dramatically, because the overall air/fuel mixture is very lean
(excess air). However, there will be definite effects near
full power. Full power is usually limited by smoke level,
which is an indication that some of the fuel is not burning due
to a lack of oxygen. If the vehicle is operated at full power
at high altitude, even less air is available, and HC and CO
emissions will increase dramatically under full-power operation.
The easiest solution to this "full-power" problem is to
limit the maximum fuel flow at high altitude relative to that
at low altitude. Limiting the maximum fuel rate can be
accomplished with a simple manual adjustment and this technique
is expected to be used, to varying degrees, by most oiesel
manufacturers to meet the current interim (1982-83)
high-altitude standards. Reducing the maximum fuel flow keeps
the engine out of excessively rich operating modes. It also
reduces the maximum power of the engine. Since most
diesel-powered vehicles are not overpowered, this may be a
serious drawback if the maximum fuel rate must be significantly
altered, especially in hilly regions. While this fact would
tend to reduce sales of these vehicles at high altitude, the
majority of the effect is there even without high-altitude
emission control since the additional fuel does not increase
engine power at high altitude to the same extent as it does at
low altitude.
The above discussion applies most directly to situations
where fairly high NOx levels are allowed, and hence, no exhaust
gas recirculation (EGR) is needed to control NOx emissions. As
EGR is added to the combustion chamber for NOx control, which
will generally be required under the 1.0 g/mi standard, the
effective air/fuel ratio (i.e., the oxygen/fuel ratio)
decreases. Theoretically, the adverse effect on HC and CO
emissions of further reducing the amount of combustion air by
moving to high altitude may be exacerbated by the use of EGR.
This could make controlling these emissions to the requisite
levels in conjunction with the 1.0 g/mi InOx standard more
difficult than is currently the case for 1982 and 1983.
Table IX-1 shows the available data regarding the effect
of altitude on the emissions from diesels both with and without
EGR. It must be noted that because of the paucity of
information, firm conclusions as to the effect of EGR on
high-altitude emissions and the requisite control technology
are difficult to make. Based on the available data, however,
it appears that EGR may not necessarily exacerbate KC and CO
emission increases at higher elevation. Therefore, the control
technique that is currently used to comply with the current
1982-83 HC and CO proportional standards (i.e., limiting the
-------
Taole IX-1
Average FTP
Powered
Emission
Passenger
Levels of Diesel-
Cars (no EGR)
In-Us
;e Vehicles
11]
Test
Vehicle Site
HC
(y/ui)
CO
(g/Mi)
NOx
(y/mi)
F .E .
(mi/^al)
Par t.
(y/mi)
1980 Volkswagen Hiyh
90 CID Low
.57
.22
1.88
.78
1. uy
1.05
3 J. 3
39 .b
.40
.26
1979 Olds Hiyh
260 CID Low
1.39
. 68
2.21
1.50
1.29
1.43
20 .5
22.9
.96
.78
1979 Olds High
260 CID Low
.97
.39
2 .57
1.49
1.56
1.67
20.6
24.2
1.82
1.13
1974 Peugeot High
129 CID Low
6.74
3 .86
8.88
3.83
.98
.93
21.0
25.2
2.43
.897
1977 Mercedes High
147 CID Low
.65
.25
1.04
.67
1.51
1.29
25.8
31.8
.47
. 3o
Average (all vehicles) High
Low
Increase
2 .06
1.08
91%
3 .32
1.65
101%
1.28
1.27
1%
23.6
27.6
-14%
1.22
.69
7 8 %
Average (w/o Peuyeot) High
Low
Increase
.89
.38
134%
l.ya
l. li
74%
1.36
1.36
0%
24.3
28.2
-14%
.92
.03
4b%
Experimental Vehicles with
Anaioy EGR
Control l z'j
Test
Vehicle Site
HC
(y/iai)
CO
(y/ui)
NOX
(y/ini)
Part.
(y/i.ii)
#04530, GM High
5.7 L, 4000 # IW Low
.60
.34
2 .3
1.4
.83
1.18
1.U4
.46
#04531 / GM High
5.7 L, 4000 # IW Low
.44
.22
.98
1.04
1.21
1.48
.54
.34
-------
IX-4
maximum fuel flow) should also be sufficient to comply with
similar HC and CO standards in the base scenario.
Of course the vehicles equipped with EGR systems in Table
IX-1 do not attain the 1.0 g/mi NOx standard at low altitude,
so it is possible that if the rate of EGR was increased to meet
the standard/ HC and CO emissions could indeed be negatively
affected. However, to attain the 1.0 g/mi NOx standard will
require more than just simply increasing the EGR rate.
Advanced electronic (analog) units are expected to be used.
These more sophisticated systems should provide manufacturers
with the flexibility to tailor the rate of EGR in such a way as
to mitigate any negative effect on HC and CO at high altitude.
Also, if limiting the maximum fuel flow in conjunction with
more advanced analog EGR systems is insufficient to attain the
proportional standards of the base scenario, other control
techniques can be easily used, such as manually readjusting the
fuel injection timing. Therefore, the proportional standards
of the base scenario should be readily attainable with 110
additional hardware.
Complying with the statutory standards of scenario 2 will
be more difficult. Although the two previously described
control techniques also may be sufficient to attain these more
stringent standards, it is likely that additional emission
control will be needed. This emission reduction may be
achieved by modifying the EGR system on high-altitude
vehicles. Returning to Table IX-1, it is clear that even
though NOx emissions naturally decrease with increasing
elevation, this effect is even greater for EGR equipped
vehicles. Manufacturers can take advantage of this NOx
phenonmenon to control HC and CO levels by reducing the rate of
EGR at high altitude and allowing NOx to increase slightly back
to the level of the NOx standard. This would ease any negative
effect of EGR on HC and CO emissions and make these pollutants
easier to control via limiting the maximum fuel flow and
resetting injection timing.
Nevertheless, it is possible that these techniques will be
insufficient in controlling emissions from worst case
vehicles. In these few instances, it may be necessary to
employ more sophisticated digital electronic EGR systems at
high altitude. However, the potential need for such complex
systems would be eliminated entirely if such vehicles were
eligible for exemption from the high-altitude requirements.
Therefore, to comply v»ith scenario 2, high-altituae LDDVs are
expected to require manual adjustments to limit the maximum
fuel rate ana recalibrate fuel injection timing, in addition to
using a differently calibrated electronic control module for
EGR.
For high-altitude LDDTs, the requisite emission controls
are essentially analogous to those for LDDVs. The only
-------
IX-5
important difference is that the low-altitude NOx standard for
light trucks (2.3 g/mi) is considered less stringent than the
NOx standard for passenger cars (1.0 g/mi), so many LDDTs will
not be equipped with EGR systems, while those that are will use
simple non-electronic units. For those LDDTs without EGR, the
high-altitude standards should generally be somewhat easier to
achieve since the increase in HC and CO will not be exacerbated
by this NOx control technique. For the purposes of this study,
however, EPA will assume that all LDDTs are equipped with
non-electronic EGR systems to be conservative and to simplify
the analysis. Therefore, to comply with the base scenario,
high-alititude LDDTs will require a manual adjustment to limit
the maximum fuel rate and recalibrate fuel injection timing, if
necesary. To comply with scenario 2, these vehicles are
expected to require the above two adjustments, in addition to
changing the EGR valve.
III. ECONOMIC IMPACT
As in the previous chapters of this report, the cost of
implementing the more stringent high-altitude standards of
scenario 2 will be determined incrementally to the cost of
continuing the proportional standards under the base scenario.
The only difference in hardware between the two scenarios
is that LDDVs and LDDTs are expected to require differently
calibrated electronic EGR modules and EGR valves, respectively,
to comply with scenario 2. Since in each case this is a
replacement of an existing unit, there is no incremental
hardware cost associated with the change, as discussed more
fully in Chapter V regarding similar modifications to LDDVs.
Although there are no incremental hardware costs, all
diesel vehicles sold in high-altitude areas will require
oevelopment to recalibrate the fuel-metering device, injection
timing, and optimize the EGR control. This may entail
additional development and certification costs than would occur
under the base scenario in some cases. For LDDVs, eight
non-California engine families were sold in 1980. It is
assumed that twice this amount will be sold in 1984 due to the
dramatic increase expected in future diesel sales. Since 15
percent of the engine families are assumed to be exempt under
scenario 2, a total of 14 engine families should require
development testing. The number of development tests estimated
here will be the same as that estimated for the 1982 and 1983
interim standards,[6] or 20 development tests per engine
family. The development costs for diesel-powered vehicles
should also be the same as previously estimated for the
gasoline-fueled LDVs and LDTs, or $500 per test. The total
development cost is then $140K for LDDVs and should occur one
year prior to the implementation date of this regulation, or
1984. As discussed earlier for LDGVs in Chapter V, there is no
difference between the certification requirements of the base
-------
IX-6
scenario and scenario 2. Therefore, no additional
certification costs for LDDVs will be incurred by implementing
scenario 2.
For LDDTs, all engine families will undergo development
testing under the base scenario in 1984 to comply with the more
stringent proportional standards taking effect in that year.
Therefore, there is no increment in either development or
certification for LDDTs under scenario 2.
The development costs for LDDVs can be amortized over each
affected diesel vehicle. First, the diesel vehicle production
at high altitude must be determined. The number of 1984-88
LDDVs sold in areas above 4,000 feet (excluding California)
have already been projected to be approximately 221,000. These
figures were derived from nationwide sales estimates made by
Data Resources Corporation, projections of diesel production
taken from the recently promulgated standards for control of
particulate emissions from light-duty diesels[7], and the same
estimates of population distribution used to determine LDGV
emissions in Chapter III. These supporting data are contained
in Table IX-2. If the development cost were recovered over the
LDDVs projected to be sold at high altitude from 1984 to 1988,
and if a 10 percent discount rate is applied, the cost per
vehicle is roughly $1. Thus, meeting scenario 2 will increase
the purchase price of the average high-altitude LDDV by $1 over
that which would occur under the base scenario. For LDDTs,
there will be no increase in purchase price.
The aggregate cost of adding light-duty diesel control to
scenario 2 can now be determined. Based on the above-mentioned
cost per LDDV, the production volume listed in Table IX-2, and
a 10 percent discount rate, the 5-year aggregate cost is
approximately $166,000 (1981 dollars discounted to 1984).
IV. ENVIRONMENTAL IMPACT
In this section, the effects of scenario 2 on emissions
with regard to light-auty diesel-powered vehicles ana trucks
are explored. The same basic methodology used in Chapters V
and IX to estimate the environmental impact of LDGV and LDGT
control are followed below. Since the air quality modeling
results in those chapters included diesel control, this topic
will not be readdtessed here. Only the effect on emissions is
presenteo.
A. Methodology
As mentioned earlier in Chapters IV and VIII, total
lifetime emissions will be determined from LDDVs and LDDTs sold
from 1984 to 1988. To determine these emissions, three factors
are multiplied together: 1) the number of miles the average
LDDV or LDDT is expected to be driven in its lifetime, 2)
-------
IX-7
Table IX-2
Light-Duty Diesel Sales Projections
(1Q3 Vehicles or Trucks) l6,7J
LDDV Sales[a]
Nationwide Above
Year LDDV Sales 4,000 Feet
1984 1,100 31
1985 1,300 36
1986 1,600 45
1987 1,900 53
1988 2,000 56
5-Year Sum 7,900 221
T These values were ueteriiiined by assuming that the nuuoer
of new vehicle sales is directly proportional to
population distribution. Since 3.1 percent of the U.S.
non-California population resides above 4,0U0 feet, this
is also the fraction nationwide non-California sales that
is expected to occur in nigh-altitude areas of the country.
-------
1X-8
emission factors (the amount of pollutant emitted per vehicle
or truck per mile of travel), and 3) the number of
diesel-fueled vehicles and trucks sold above 4,000 feet, the
initial control altitude for scenario 2 and the base scenario.
The average lifetime mileage for light-duty vehicles is
assumed to be 100,000 miles and that of light-dut> trucks,
120,000 miles.[6,8] To estimate lifetime emissions, emission
factors were determined for LDDVs and LDDTs that were halfway
through their average lifetimes (i.e., at 50,000 and 60,000
miles, respectively). Since the same methodology was followed
to determine emission factors for LDDVs and LDDTs as was used
to determine those for LDGVs, the discussion of this topic in
Chapter III is equally applicable here ana will, therefore, not
be repeated. It is important to remember, however, that
scenario 2 applies only to those vehicles and trucks sold above
4,000 feet and that the high-altitude emission data available
were based on tests conducted at Denver (elevation 5,300
feet). Thus, the average lifetime emission rates in this stud>
for LDDVs and LDDTs sold above 4,000 feet were based on tests
at 5,300 feet. These are presented in Table IX-3.
By combining the mileage, emission factor and sales data
previously presented in Table IX-2, the total exhaust emissions
from 1984-88 LDDVs and LDDTs in areas above 4,000 feet can be
determined. These are presented in Table IX-4 for both the base
scenario and scenario 2.
B. Discussion of Results
Table IX-4 shows that scenario 2 would reduce HC emissions
from 1985-89 LDDVs by approximately 3,320 metric tons compared
to the base scenario and lower hC emissions from LDDTs sold in
the same ^ears by 1,190 metric tons. Similarly, CO emissions
would decrease by roughly 21,000 and 6,260 metric tons from
LDDVs and LDDTs, respectively. No NOx reductions would be
realized since the high-altitude NOx standards under scenario 2
would be the same as under the base scenario.
Referring back to Chapter IV, one sees that scenario 2
would lower KC and CO emissions from LDGVs sola above 4,000
feet by approximately 17,940 metric tons and 331,000 metric
tons, respectively, for the 5-year period studied. In Chapter
VIII, the effects of scenario 2 on LDGTs were combined with
those from LDGVs to yield total 5-year HC reductions of roughly
22,670 metric tons and total CO reductions of approximately
468,400 metric tons from these sources. When benefits from
adding LDDV and LDDT control to scenario 2 are included, total
additional 5-year HC emission reductions beyond those provided
by the base scenario are roughly 27,180 metric tons while CO
emissions would be further reduced by approximately 495,700
metric tons. Thus, approximately 17 percent of the total HC
emission reductions and 6 percent of the total CO emission
-------
IX-9
Table IX-3
Emission Rates for 1985-89 Light-Duty
Diesel-Fueled Vehicles and Trucks at 5/300 Feet[a]
Base Scenario Scenario #2
Pollutant LDDVlbJ LDDTLbJ LDDV LDDT
HC Zero-Mile Emission 0.54 0.76 0.39 0.61
Level (y/mi)
Deterioration Rate 0.03 0.06 0.03 0.06
(y/mi/10,000 miles)
Averaye Lifetime Emis- 0.69 1.12 0.54 0.97
sion Level (y/mi)
CO Zero-Mile Emission 2.22 2.77 1.27 1.9b
Level (g/mi)
Deterioration Rate 0.05 0.09 0.05 0.09
(y/mi/10,000 miles)
Averaye Lifetime Emis- 2.47 3.31 1.52 2.52
sion Level (y/mi)
NOx Zero-Mile Emission 0.75 1.89 0.75 l.d9
Level (y/mi)
Deterioration Rate 0.05 0.07 0.05 0.07
(g/mi/10,000 miles)
Averaye Lifetime 1.0 2.31 1.0 2.31
emission Level
(g/mi)
I a J Emission Rate = ZML + (M)(DR)
Where:
ZML = Zero-Mile Level
M = Cumulative Mileage/10,000 (Cumulative mileage = 50,000
miles for LDDVs and 60,000 miles for LDDTs)
DR = Deterioration Rate
[b] Liyht-Duty Diesel Vehicle
Light-Duty Diesel Truck
-------
IX-10
Table IX-4
Lifetime Emissions from 1985-39
LDDVs and LDDTs Sold Above 4,U0U Feet
(103 metric tons)
Reductions
(base case
Base Case Scenario 2 minus scenario 2)
Pollutant
LDDV
LDDT
LDDV
LDDT
LDDV
LDDT
HC
15.25
8.87
11.93
7.68
3 .32
1.19
CO
54.59
26.22
33.59
19.96
21.00
6 .26
NOX
22.1
18.30
22.1
18.30
0
0
-------
IX-11
reductions result from LDD control. This shows that
controlling light-duty diesel motor vehicles can produce
significant reductions in HC emissions.
V. COST EFFECTIVENESS
Results from the economic ana environmental impact
sections of this chapter were combined to determine the cost
effectiveness of adding LDDV ana LDDT- control to scenario 2.
The same procedure outlined in Chapter V to determine the cost
effectiveness of LDGV control was followed in this section.
The results of this analysis can be found in Table IX-5 along
with the incremental cost effectiveness of LDGV and LDGT
control for scenario 2 as determined in Chapters V and VIII.
The incremental cost effectiveness of combined LDV and LDT
control is also presented in Table IX-5.
As shown in Table IX-5 controlling LDDVs and LDDTs to the
levels associated with scenario 2 is more cost effective than
controlling LDDVs and LDDTs to the same levels. It is not
surprising then, that adding the control of light-duty
diesel-fueled motor vehicles to scenario 2 makes it a more
cost-effective approach than if only light-duty gasoline-fueled
motor vehicles were controlled more stringently. For the worst
case analyzed, the cost of removing HC is reduced from $535 per
metric ton to $465 per metric ton. The cost effectiveness of
CO removal for the worst case analyzed remains unchanged at $30
per metric ton.
Table IX-6 lists the cost effectiveness of other control
strategies already adopted. (This same table also appears as
Table VI-3 of Chapter VI.) From this table it can be seen that
when LDDV and LDDT controls are viewed either spearately or in
conjunction with LDDVs and LDGTs, the cost effectiveness of
scenario 2 is among the most cost efficient approaches to
reducing emissions of HC and CO.
-------
IX-12
Table IX-5
Incremental cost Effectiveness of Scenario 2
Costs[a]
(dollars per
Emission Reductions
(10^ metric tons
Incremental
Cost Effectiveness
Vehicle
vehicle )
per
vehicle)
HC
CO
Type
Low
High
HC
CO
LOW
High
LOW
High
LDGV
-11
15
13.0
239.9
neg
. 575
neg
30
LDGT
-69
21
21.7
326.2
neg
. 485
neg
30
LDGV and
LDGT
-24
16
15.0
259 .7
neg
. 535
neg
30
LDDV
1
15.0
95.0
35
5
LDDT
0
18.0
94.8
0
0
All Light-
-21
14
15.1
236.9
neg
. 465
neg
30
Duty Vehi-
cles and
Trucks
[a] 1981 dollars discounted to the year of vehicle purchase. The
low estimates for gasoline-fueled vehicles are based on the
inclusion of tentative fuel-economy benefits.
-------
IX-13
Table IX-6
Cost-Effectiveness Comparison With
Other Emission Control Strategies
(1981 dollars per metric ton)
Emissions
After
Control
Cost
Baseline
Strategy
Effectiveness
Control Program
Emissions[a]
implemented
HC
CO
LDV Statutory
HC
0.9
HC
0.41
734
67
Standards[10]
CO
15
CO
3.4
LDV I/Mtll]
—
—
640
58
LDT 1984
HC
1.7
HC
0.8
195
15
Standards[12]
CO
18
CO
10
HDE 1984
Standards
(gasoline)
HC
1.5
HC
1.3
305
10
CO
25
CO
15.5
(diesel)
HC
1.5
HC
1.3
325
—
CO
25
CO
15.5
Motorcycle
HC
9
HC
8-22.5[c]
582
Neg.
Standaras[14]
CO
34.67
CO
27.4
HDG Evap.[15][d]
HC
1.8
HC
0.17
200
—
Interim 1982-63
HC 1.
47 (cars)
HC 1.
33 (cars)
393
12
HA Standaras[9]
4..
19 (trucks)
3.
78 (trucks)
CO 16
.23 (cars)
CO 13
i.21 (cars)
73
.02 (trucks)
55
i.65 (trucks)
[a] Emission levels are in grams per mile except for the HDE
1984 standards which are in grams per brake
horsepower-hour.
[b] The baseline and after-control strategy emission values
were based on different test procedures (see reference 13).
[c] Sliding scale based on engine displacement (cubic
centimeters).
[d] The evaporative standard is in terms of g/test and
converted to g/mi here to facilitate comparison.
-------
IX-14
References
1. "Testing of Five Diesel-Powered Passenger Cars at High
and Low Altitude," U.S. EPA, OANR, OMS, ECTD, TEB, Thomas Tupaj,
October 1980.
2. "General Motors Closing Comments to the U.S. EPA Public
Hearing on the 1984 High-Altitude Requirements of the 1977 Clean
Air Act Amendments," April 22, 1980.
3. "General Motors Comments to the EPA Hearing on' the 1984
High-Altitude Emission Control Requirements," March 3, 1980.
4. "Cost Estimations for Emission Control Related
Components/Systems and Cost Methodology Description," U.S. EPA,
OANR, OMS, ECTD, Lindgren, Leroy H., EPA-460/3-78-00, March, 1978.
5. "Closed-Loop Control, Three-Way Catalyst Automotive
Emission System Costs and Issues," U.S. Department of
Transportation, National Highway Traffic Safety Administration,
Contract No. DOT-HS-8-01912, September 1979.
6. "Average Lifetime Periods for Light-Duty Trucks and
Heavy-Duty Vehicles," U.S. EPA, OANR, OMS, ECTD, SDSB, Glenn W.
Passavant, November 1979.
7. Data Resources, Long-Term Review, Fall 1980.
8. "Percentage of VMT above 1,800 and 4,000 Feet," Memo
from Daniel P. Heiser, U.S. EPA, OANR, OMS, ECTD, SDSB, EPA to
Richard Wilcox, U.S. EPA, OANR, OMS, ECTD, SDSB, February 1981.
9. "Final Regulatory Analysis, Environmental and Economic
Impact Statement for the 1982 and 1983 Model Year High-Altitude
Motor Vehicle Emission Standards," U.S. EPA, OANR, OMS, ECTD,
SDSB, October 1980.
10. "Interagency Task Force on Motor Vehicle Goals Beyond
1980," U.S. EPA, Department of Transportation, Prepared by the
Panel of Automotive Manufacturing and Maintenance, March 1976.
11. "Update on Cost Effectiveness of Inspection and
Maintenance," U.S. EPA, OANR, OMS, ECTD, I/M Staff,
EPA-AA-IMS/81-9, April 1981.
12. "Regulatory Analysis and Environmental Impact of Final
Emission Regulations for 1984 and Later Model Year Light-Duty
Trucks," U.S. EPA, OANR, OMS, ECTD, SDSB, May 20, 1980.
13. "Regulatory Analysis and Environmental Impact of Final
Emission Regulations for 1984 and Later Moael Year heavy-Duty
Engines," U.S. EPA, OANR, OMS, ECTD, SDSB, December 1979.
-------
IX-15
References (cont'd)
14. "Environmental and Economic Impact Statement - Exhaust
and Crankcase Regulations for the 1978 and Later Model Year
Motorcycles," U.S. EPA, OANR, OMS, ECTD, SDSB, 1976.
15. "Regulatory Analysis for the Proposed Evaporative
Emission Regulations for Heavy-Duty Vehicles," U.S. EPA, OANR,
OMS, ECTD, SDSB, April 1980.
-------
Chapter X
Light-Duty Diesel Particulate Euissions
I. INTRODUCTION
As mentioned earlier, section 206(f) of the Clean Air Act
requires that the liyht-duty vehicle emission control
provisions of section 202 apply in high-altitude areas
beginning with the 1984 model year. In addition to gaseous
pollutants, EPA has also established particulate standards for
diesel-powered vehicles under section 202. Since particulate
emissions are addressed by section 206(f) of the Act, they have
been included within the scope of this report also.
Previous chapters discussed various aspects of controlling
gaseous emissions from light-duty vehicles and trucks at high
altitude. Much of those analyses compared the various control
strategies outlined in Chapter II to a base scenario, which
called for a continuation of the interim 1982-83 high-altitude
standards. Because particulate emissions from diesel-powered
light-duty vehicles and trucks (LDDVs and LDDTs) are not
controlled by the interim program, they are being addressed
separately below.
This chapter specifically analyzes tiie technology expected
to be needed to reduce particulate emissions at hi^h altitude.
Unlike the previous analyses that evaluated the control or
gaseous emissions, this analysis will not specifically include
estimates of the cost of control. The available emissions data
are too limited to project the necessary control technology
with the confidence needed to justify assigning a definite cost
to the technology. In fact, the data are so limited that it is
not possible to definitively determine what the actual
¦proportional" diesel particulate stanuard should be. All that
can be done at this time is to estimate the proportional level
and roughly identify what technology would likely be needed to
meet both the proportional and low-altitude levels at high
altitude. Some judgment will then be made concerning the cost
which is associated with the two levels. Likewise, the lack of
emissions data also prevents the accurate determination of the
overall emissions reduction and air quality benefit resulting
from control. Thus, these calculations will also not be
performed.
II. CONTROLLING PARTICULATE EMISSIONS FROM HIGH-ALTITUDE
LIGHT-DUTY DIESELS
The amount of data available on the effect of high
altitude on diesel particulate emissions is extremely limited.
That which is available is shown in Table X-l. Even within
this small data set, the information from one vehicle (Peugeot)
is not very useful because its HC emissions are extremely hiyh
for a diesel and it is likely that some malfunction was present.
-------
Table X-l
Average FTP Emission Levels of Diesel-Powered Passenger Cars
In-Use Vehicles without EGRj.1]
Test
HC
CO
NOx
F.E.
Part.
Vehicle
Site
(g/mi)
(q/mi)
(g/mi)
(mi/gal)
(g/iai)
1980 Volkswagen
High
.57
1.88
1.09
35.3
.40
90 CID
Low
.22
.78
1.05
39.5
.26
1979 Olds
Higli
1.39
2.21
1.29
20.5
.ya
260 CID
Low
.68
1.50
1.43
22.9
.78
1979 Olds
High
.97
2.57
1.56
2U .6
1.82
260 CID
Low
.39
1.49
1.67
24.2
1.13
1974 Peugeot
High
0 .74
8.88
.98
21. U
2 .43
129 CID
Low
3.86
3.83
.93
25.2
.90
1977 Mercedes
High
.65
1.04
1.51
25.8
.47
147 CID
Low
.25
.67
1.29
31.8
.36
Odoineter-52206
Average (all vehicles)
High
2 .06
3.32
1.28
23.6
1.22
Low
1.08
1.65
1.27
27 .6
.69
Increase
91%
101%
1%
-14%
78%
Average (without Peugeot)
High
.89
1.93
1.36
24.3
.92
Low
.38
1.11
1.36
28.2
. 6 3
Increase
134%
74%
0%
-14%
46%
Experimental
Vehicles
with Analog
EGR[2]
Test
HC
CO
NOx
Part.
Vehicle
Site
(q/mi)
(q/mi)
(g/mi)
(g/mi)
GM
High
.60
2.3
.83
1.04
5.7 L, 4000 * IW
Low
.34
1.4
1.18
.46
GM
High
.44
.98
1.21
.54
5.7 L, 4000 * IW
Low
.22
1.04
1.48
.34
-------
X-3
Ignoring this vehicle, the first set of vehicles shows
that particulate emissions increase by an average of 46 percent
when tested at high altitude. Since these vehicles represent
as near an uncontrolled baseline as is available, the best
estimate of a proportional diesel particulate standard would be
roughly 46 percent (an even 50 percent will be used hereafter)
higher than the low-altitude standard. This, of course, is
only a rough estimate due to the small number of vehicles
tested. Thus, in 1985, the proportional diesel particulate
standard for light-duty vehicles would be 0.3 g/mi relative to
the 0.2 g/mi low-altitude standard and would be 0.39 g/mi for
light-duty trucks relative to the low-altitude standard of 0.26
g/mi.
Now that the proportional standard has been estimated, the
next step is to estimate what technology will be needed to meet
this level and the low-altitude standard at high altitude.
Unfortunately, even less data are available here than were
available earlier. What is available is shown in the lower
half of Table X-l.
The two vehicles shown in the lower half of Table X-l are
General Motors' diesels equipped with analog EGR to reouce NOx
emissions from an uncontrolled level of around 1.7-2.0 g/mi.
As can be seen, the effect of altitude on particulate emissions
increases as NOx is controlled. This is not unexpected since
increased levels of EGR reduce the amount of oxygen available
for combustion. When moved to high altitude, these vehicles
are already operating richer than normal and the effect of
altitude is simply to move further in the same direction. With
diesels, the effect of a given shift in the air/fuel ratio has
an ever increasing effect on smoke and particulate emissions as
the air/fuel ratio is lowered. Thus, a greater effect occurs
with higher levels of EGR.
While not unexpected, this effect is important since by
1985 all light-duty diesels will be required to meet a 1.0 g/mi
NOx standard and most, if not all, vehicles will be equipped
with EGR to reach this level. This means that if not
controlled at all, the particulate emissions of these vehicles
at high altitude will likely be more than 50 percent higher
than their low-altitude levels and will exceed a proportional
standard.
At the same time, however, hydrocarbon emissions will be
reduced to meet at least the 0.57 g/mi proportional standard
regardless of whether or not particulate emissions are
controlled. As shown in Table X-l, all but one vehicle will
need some adjustment at high altitude to meet the 0.57 g/mi
standara. Any adjustment for hydrocarbon emissions, such as
limiting maximum fuel flow or adjusting injection timing,
should also reduce particulate emissions. This is likely to
bring particulate levels within the proportional levels. This
-------
X-4
must remain only a projection, however, since no high-altitude
particulate levels are yet available from 1982 Uieseis
certifying to the interim gaseous emission standards.
Even if the hydrocarbon controls were not totally
sufficient, however, the E.GR systems which will be on diesels
by 1985 snould be able to be adjusted to ensure tnat a
proportional particulate standard can be met. For example, the
data in the lower half of Table X-l show essentially the same
vehicle at two NOx levels (two different EGR rates). The first
vehicle with a low NOx level emits 1.04 g/mi particulate at
high altitude; a 126 percent increase over low altitude. Note,
however, that the second vehicle having a lower EGR rate
achieves essentially the same NOx level at hiyn altitude that
the first one achieved at low altitude. in other words, the
effect of altitude compensated for the reduction in the EGR
rate. However, the high-altitude particulate emissions in this
case are only 0.54 g/mi, which is only a 17 percent increase
over the original 0.46 g/mi level of the first vehicle. This
is well below the 50 percent allowance discussed above and
occurred before the application of any obvious control to the
maximum fuel rate to prevent locally rich combustion (though
this might also increase NOx emissions somewhat). Thus, with
simple controls such as adjustment to the present analog EGR
system and the maximum fuel rate, a proportional particulate
standard should be achievable.
These controls should be very inexpensive since no new
hardware should be required. The primary cost will be
associated with having to develop special high-altituue
calibrations, but this is already accounted for in tne
compliance costs for the proportional gaseous emission
standards. Therefore, the cost of adding a proportional
particulate standard to the proportional gaseous emission
standards should be very small or negli^ibxe.
It should be mentioned that this report has not explicitly
considered the effect of trap-oxidizer technology on
high-altitude particulate emissions. Trap-oxidizers are
expected to be used by most venicles to meet the 19d5
particulate standard.[1] No data are available on the effect
of high altitude on the particulate emissions of a
trap-equipped vehicle. However, on the basis of all the
information available on the effects of trap-oxidizers on
particulate emissions, these traps are generally proportional
control devices. That is, they reduce emissions by a constant
proportion regardless of the absolute emission levels entering,
within reasonable limits. Thus, the above argument should hold
for vehicles with or without traps.
There is one aspect of a trap-oxidizer for which the
argument may not hold rigorously. While the trap is a
proportional control device, its size is somewhat dependent on
-------
X-5
the absolute amount of particulate matter entering it per
mile. At least some trap-oxidizers are expected only to
burn-off the trapped particulate periodically and the number of
miles these vehicles can be driven between burn-off will depend
on the size of the trap. At high altitude, then, if more
particulate were entering the trap due to higher en^ine-out
emissions, the number of miles before burn-off became necessary
would be reduced. If the burn-off cycle is controlled
"intelligently" via electronic or mechanical controls, then the
frequency of burn-off could be simply increased and the problem
solved. This should be relatively inexpensive and could easily
be performed when the other engine parameters are adjusted for
high altitude.
Even if burn-off is not controlled intelligently, there
may be another factor operating at aiyh altitudes which woulu
automatically increase the frequency of burn-off. The lower
air density at high altitude will decrease the air/fuel ratio
at any given power level of a diesel engine. The result is
that both the comoustion and exhaust temperatures will increase
at high altitude at any given load. Since sufficient exhaust
temperature is the key to uurn-off, the trap-oxiuizer should
burn-off at lower loads at high altitude than at low axtituae,
increasing the frequency of burn-off. Given that the increase
in engine-out particulate emissions should only be on the order
of 15-20 percent (as estimated above), tnis frequency of
burn-off should only have to increase by the same amount, which
is fairly small. While only these general arguments can be
presented at this time, it would appear that the trap size
should not need to be increased on high-altitude venicles and
that proportional control of particulate emissions should be
inexpensive even for vehicles equipped with trap-oxiuizers.
It should be mentioned at this point that light-duty
trucks have not been specifically addressed. The reason for
this is that no hard data exist on the effect of hiq'n altitude
on particulate emissions from light-duty trucks. From the
low-altitude particulate data available 011 these trucks, [3] and
the fact that the engine used is essentially the same as those
for li^ht-duty vehicles, it is expected that the same
conclusions will hold for light-duty trucks as did for
light-duty vehicles. In the case of either diesel vehicles or
trucks, more data are needed uefore any definitive conclusions
can be drawn concerning the feasibility of a hiyh-altitude
particulate standard. However, at tnis time, Er>A expects that
a proportional standard should be inexpensive to achieve for
either vehicle class.
This brings up the question of the feasibility of meeting
the low-altitude particulate standards at high altitude. As
described above, using the little data that are available, the
particulate emission increase after simple controls are applied
may be only 15-20 percent higher than low-altitude levels.
-------
ii-6
With the addition of sophisticated analog EGR systems to
achieve the statutory gaseous emission standards, it is
possible that particulate emissions could be reduced 15-2U
percent to achieve the low-altitude levels at high altitude
with no additional cost. Unfortunately, there are no data to
support or reject this projection.
At this time, the best that can be said is that any
additional controls that are added to reduce HC and CO
emissions down to the statutory standards should aiso further
reduce particulate emissions. In any case, based on the
General Motors' data in Table X-l, the high-altitude
particulate level achievable by diesels should be fairly close
to the low-altitude standards since even simple EGR systems
result in hiyh-altitude particulate levels only 15-20 percent
higher than at low altitude. At this time, the best approach
appears to be to evaluate the effect of yaseous emission
control on particulate emissions as more data on the former
become available. The achievable particulate level could then
be simply defined as the level resulting from the control of
gaseous emissions to their appropriate level. Of course, the
effect of this control technology on particulate emissions
should be considered when calculating the cost effectiveness of
control.
III. SUMMARY
It would appear that a high-altitude proportional
particulate standard, when definitively determined, would be
inexpensive and achievable. In addition, further reductions
from proportional levels may be achievable from the application
of gaseous emission controls at little or no extra cost.
However, it is not possible to determine if the current
low-altitude particulate standards would be fully achievable at
hiyh altitude. Due to the severe limitations of the existing
data base, the only firm conclusions^ which can be made
concerning control beyond proportional levels G.s\ that: 1) the
reductions in particulate emissions resulting from the
application of gaseous emission controls should be considered
in determining the effectiveness of these controls; and 2) the
maximum level of particulate control that is feasible may be
the particulate level resulting from the application of these
gaseous emission controls.
-------
X-7
References
1. "Testing of Five Diesel-Powered Passenger Cars at High
and LOW Altitude," U.S. EPA, OANR, OMS, TEB, T. Tupaj, October
10'6Q.
2. General Motors Closing Comments to the U.S. EPA Public
Hearing on the 1984 High-Altitude Requirements of the 1977 Clean
Air Act Amendments, April 22, 193U.
3. "Regulatory Analysis of the Light-Duty Diesel
Particulate Regulations for 1982 and Later Model Year Li^ht-Duty
Diesel Vehicles," U.S. EPA, OAWR, OMS, ECTD, SDSB, February 2U,
1980.
-------
Chapter XI
High-Altitude Standards as a Consequence
of Revised Low-Altitude standards
I. INTRODUCTION
The previous chapters in this report analyzed various
alternative high-altitude standards which were based on the
current low-altitude gaseous standards of 0.41 g/mi HC/ 3.4
g/mi CO and 1.0 g/mi NOx for light-duty vehicles (LDVs).
Recently, however, the debate concerning amendments to the
Clean Air Act (CAA) has included the possibility of revising
certain of the statutory low-altitude emission standards for
LDVs upward from the current levels. Although such a revision
remains speculative at this time, the effect of less stringent
low-altitude standards on potential high-altitude standards
will be evaluated in this chapter to maximize the value of this
report regardless of the level of the low-altitude standards.
This chapter discusses the effects of revised low-altitude
standards on both gasoline-fueled LDVs (LDGVs) and
diesel-powered LDVs (LDDVs). Like previous chapters, the
standards affecting LDVs are assumed to be effective beginning
in the 1984 model year.
The effect of revised gaseous emission standards on the
technical feasibility of complying with the requisite diesel
particulate standards for LDDVs is not specifically addressed
in this chapter. However, it should be pointed out that by
revising the NOx standard upward from 1.0 g/mi, diesel
particulate standards may be less costly to attain at
high-altitude than estimated in Chapter X. The principle
control strategy for reducing NOx emissions from LDDVs is the
use of exhaust gas recirculation (EGR). Unfortunately this
type of NOx control method, besides generally increasing
particulate emissions, also appears to increase the sensitivity
of particulate emissions to altitude, making attainment of the
high-altitude particulate standard more difficult. Therefore,
by revising the NOx standard upward, less EGR will be required
and the accompanying lower particulate emissions along with the
decreased altitude sensitivity should result in easier and less
costly control.
II. DEFINITION OF CONTROL SCENARIOS
The high-altitude LDV control scenarios analyzed in this
chapter depend on what the revised statutory low-altitude
standards may be. Since the exact levels of any future
low-altitude standards have not been determined, an assumption
regarding possible standards is necessary. This analysis
assumes exhaust emission levels of the revised statutory
low-altitude standards are 0.41 g/mi HC, 7.0 g/mi CO, and 1.5
-------
XI-2
to 2.0 g/mi NOx (Table XI-1). A range of possible NOx
requirerr.ents is used to indicate the level of control that
appears to be most frequently discussed. no significant
difference in control technology at either low or high altitude
is expected within this range of NOx control. The evaporative
HC emission standards at low altitude are assumed to remain
unchanged at 2.0 gram/test (Table XI-1). Now that the
statutory low-altitude standards are known, the high-altitude
control scenarios can be defined.
As in previous chapters, there are many possible
strategies for controlling motor vehicle emissions in
high-altitude areas of the country. These strategies consist
of a combination of variables including: 1) the basic
philosophy of high-altitude emission control, 2) the allowable
emission levels for each standard, and 3) the availability of
exemptions for certain low-power vehicles. in order to limit
the analysis in this chapter to onl^ those possible scenarios
that are the most reasonable, each of these variables is
discussed below based on the conclusions of previous chapters
as they relate to the possibility of "revised" high-altitude
standards. A more complete general discussion of these
variables can be found in Chapter II.
A. The Philosophy of High-Altitude Control
This report analyzes two basic philosophical options. The
first option specifically requires that motor vehicles
continuously meet the appropriate emission standards at all
elevations without modification. Standards of this type are
referred to as "continuous" standards throughout this
document. The second option requires that all vehicles sold at
high altitude (above 4,000 feet) must be capable of meeting
high-altitude standards either automatically or after
modification when tested at 5,300 feet. Standards of this type
are referred to as "fixed-point" standards throughout this
document.
This analysis is restricted to considering only
fixed-point standards. Chapters II through VII shewed that
the continuous strategies analyzed in this report were
unreasonably burdensome by seriously restricting model
availability at high altitude, requiring expensive and
complicated emission control technology on some vehicles, and
by specifically controlling emissions at elevations above
approximately 6,000 feet, which are not expected to have an air
quality problem. Also, they were found not to be cost
effective. The primary cause of the poor cost effectiveness
was that while every vehicle in the nation had to be equipped
with hardware to ensure high-altitude emission control, these
costs were not offset by the emission reduction that occurred
only for those vehicles operating at high altitude (about 3 to
4 percent of all LDVs).
-------
XI-3
Table XI-1
Revised statutory and Revised Proportional
Emission Standards for Light-Duty Vehicles
Gaseous Standards
Revised
Year of
KC
CO
NOx Evap.HC[a]
Standard
Vehicle
Implementation
(g/mi)
(g/mi)
(g/mi) (g/test)
Statutory
LDGV[b]
1S84
0.41
7.0
1.5-2.0 2.0
LDDV[c]
1S84
0.41
7.0
1.5-2.0 N/A
Proportional
LDGV
1S84
0.57
11
1.5-2.0 2.6
LDDV
1S84
0.57
11
1.5-2.0 N/A
[a] Evaporative emission standards are not applicable (N/A) to
aiesel-powered vehicles. The low volatility of diesel
fuel produces few evaporative emissions.
[b] Light-duty c,asoline-fueled vehicle.
[c] Light-duty diesel-powered vehicle.
-------
XI-4
These findings should remain valid even if the
low-altitude standards are revised upward to the levels assumed
in this report. For example, the control strategies which
require compliance with statutory standards up to an elevation
of 10,200 feet would continue to be very difficult to achieve
in a cost-effective manner. Achieving such standards would
probably entail either exempting the majority of the fleet from
the certification requirements, in which case model
availability at high altitude would be drastically reduced, or
would entail using alternatives such as turbocharging the
engines of many if not most of the vehicles in the entire
nation, in which case the cost would be prohibitive. Even
compliance with continuous control strategies up to 6,000 feet
would remain cost ineffective because control hardware would be
required on all vehicles in the nation, while the benefit in
the form of reduced emissions would be realized only from those
vehicles operated at higher elevations. Therefore, fixed-point
standards appear to be a reasonable approach to solving
high-altitude air pollution problems.
B. The Levels of the standards
This report considers two levels of high-altitude emission
standards. The first level is referred to as "statutory"
standards and requires that the same numerical standards be met
at both low and high altitudes. The statutory standards are
assumed in this chapter to be 0.41 g/mi HC, 7.0 g/mi CO,
1.5-2.0 g/mi NOx, and 2.0 g/test evaporative HC (Table XI-1).
The second level is referred to as "proportional"
high-altitude standards. These "proportional" reduction
standards are considered to be equally as stringent as the
low-altitude standards at their respective altitudes, although
the high-altitude standards would have a higher numerical value
for HC and CO since motor vehicles naturally emit more of these
pollutants at higher elevations. For NOx, which normally
decreases with increased altitude, the numerical value of
high-altitude standard is the same as the low-altitude
standard. (Chapter III contains a more detailed discussion of
these natural phenomena.) This is consistent with the
Congressional mandate in section 202(f) of the Act which
specifically forbids a standard at high altitude from being
numerically more stringent than the corresponding low-altitude
standard.
The proportional high-altitude emission standards at an
elevation of 5,300 feet that correspond to the assumed
low-altitude standards of this chapter are 0.57 g/mi HC, 16
g/mi CO, 1.5-2.0 g/mi NOx, and 2.6 g/test evaporative HC.
However, as discussed in more detail in the control technology
section which follows, there appears to be little or no
difference in the requisite control hardware and cost of
compling with either a 16 g/mi or an 11 g/mi CO standard at
-------
XI-5
high altitude. This means that the more stringent 11 g/mi CO
standard would be a more efficient and cost effective
requirement than the 16 g/mi CO standard, since a greater
benefit (i.e./ fewer emissions) will result from spending about
the same amount of money. Put another way, this means that
high-altitude consumers would receive a greater return on their
investment in emission control hardware which is made when a
new vehicle is purchased. For these reasons, EPA believes that
11 g/mi CO is a more appropriate level of control when options
to implementing statutory standards at high altitude are
discussed, as is done in this report. Therefore, when
proportional high-altitude requirements are analyzed in this
chapter, the levels will be 0.57 g/mi HC, 11 g/mi CO, 1.5-2.0
g/mi NOx, and 2.6 g/test evaporative HC (Table XI-1).
C. High-Altitude Exemptions
Modifying some motor vehicles to comply with the
high-altitude exhaust emission standards shown in Table XI-1
may be technically difficult. (Chapter III contains a more
detailed explanation of this difficulty.) These vehicles
generally should be low-power, high fuel economy cars that are
designed principally for the low-altitude market. Although
these vehicles perform acceptably at lower elevations, they may
have extremely poor performance when operated at high altitude
because the less dense air at higher elevations reduces the
engine's power output. Such poor performers either would not
be sold or would be sold in small numbers at high altitude,
even in the absence of high-altitude regulations.
In the earlier chapters of this report, exemptions to the
high-altitude certification requirements were found to be an
effective way to rectuce the overall cost of the stanaards or,
in some cases, were found to prevent potentially negative
effects on model availability throughout the nation. In
Chapter II, the volume of exemptions was estimated based on
worst case assumptions for fixed-point statutory standards and
fixed-point proportional standards corresponding to the current
low-altitude emission standards of 0.41 g/mi HC, 3.4 g/mi CO,
and 1.0 g/mi NOx. In this chapter, the low-altitude standards
have been revised upward but the corresponding statutory and
proportional high-altitude levels are still considered to be as
stringent relative to the revised standards as the
high-altitude standards considered in the preceding chapters of
this report. Since the relative difficulty of complying with
the standards is regarded as being the same, it is assumed that
the need for exemptions will not change significantly.
Therefore, the previously estimated volume of exemptions will
be used here as shown below:
1. Five (5) percent of the fleet for high-altitude
control scenarios with fixed-point proportional standards; and
-------
XI-6
2. Fifteen (15) percent of the fleet for high-altitude
control scenarios with fixed-point statutory standards.
As in the previous chapters, this analysis includes
exemptions to illustrate the effect of including such
provisions in future high-altitude regulations. As additional
data become available it may be preferable to implement other
schemes such as waivers that would allow vehicles certified to
only low-altitude standards to be sold in high-altitude areas.
D. Selecting the High-Altitude Control Scenarios
The preceding discussion identified two control scenarios
for analysis in this chapter. The scenarios are:
1. Fixed-Point Statutory Standards
The numerical values of these standards are the same as
the revised low-altitude standards which are assumed in this
analysis, 0.41 g/mi HC, 7.0 g/mi CO, 1.5-2.0 g/mi NOx, and 2.0
g/test evaporative hC. Vehicles sold above 4,000 feet
beginning in the 1984 model year must comply with these levels
when tested at an elevation of 5,300 feet. Exemptions for
approximately 15 percent of the existing fleet will be
available.
2. Fixed-Point Proportional Standards
The numerical values of these standards are 0.57 g/mi HC,
11 g/mi CO, 1.5-2.0 g/mi NOx, and 2.6 g/test evaporative HC.
Although the 11 g/test CO high-altitude standard is not truely
proportional to the 7.0 g/mi CO low-altitude standard, the
technical difficulty of meeting it is considered to be
essentially equivalent to complying with a 16 CO high-altitude
standard which is the actual proportional value. (This is
explained in greater detail below.) Because of this equivalent
difficulty, the 11 g/mi CO standard is referred to as being the
proportional requirement in this chapter.
In this scenario, vehicles sold above 4,000 feet beginning
in the 1S84 model year must comply with the proportional levels
when tested at an elevation of 5,300 feet. Exemptions for
approximately 5 percent of the existing fleet will be
available.
One further remark concerning the control scenarios is
necessary before beginning the analysis. The primary purpose
of this report is to respond to the requirements of section
206(f)(2) of the Act. This section provides certain guidelines
which are useful in defining the analytical methodology in this
chapter of the report. First, the Act basically requires a
review of the economic impact and technical feasibility of
statutory standards of high altitude. Second, the technical
-------
XI-7
feasibility and air quality consequences of proportional
standards are to be evaluated. In the preceding chapters, the
methodology used to respond to the guidelines of section
206(f)(2) was the evaluation of all alternative scenarios by
comparison to a continuation of the fixed-point proportional
standards contained in the 1982 and 1983 high-altitude
regulations. This approach was used primarily to avoid useless
repetition of the detailed analyses which supported the
existing high-altitude standards.[1,2,3,4] Since compliance
with the proportional standards in this chapter is expected to
be similar to compliance with the current proportional
standards, the same analytical methodology will be used here to
remain consistent with the previous analyses in this report.
Therefore, implementing revised statutory standards (0.41 g/mi
HC, 7.0 g/mi CO, and 1.5-2.0 g/mi NOx and 2.0 g/test
evaporative HC) will be analyzed with regard to the incremental
costs and benefits that will accrue beyond those of continuing
a program with revised proportional standards (0.57 g/mi HC, 11
g/mi CO, 1.5-2.0 g/mi NOx and 2.6 g/test evaporative HC). To
provide additional continuity with previous chapters, the
fixed-point proportional standards will be referred to as the
"revised" base scenario and the fixed-point statutory standards
will be referred to as the "revised" scenario 2 throughout the
remainder of this analysis.
III. TECHNOLOGY ASSESSMENT
High-altitude exhaust emission control technology depends
upon the control hardware that will be used at low altitude.
Therefore, before the requisite high-altitude control hardware
can be estimated the low-altitude requirements must be
defined.
A. Low-Altitude Control Hardware
The control technology which may ultimately be used for
LDGVs to comply with the assumed low-altitude standards of 0.41
g/mi HC, 7.0 g/mi CO, and 1.5-2.0 g/mi NOx are estimated in an
EPA draft document entitled, "Motor Vehicle Emission Standards
for Carbon Monoxide and Nitrogen Oxides."[5] In that report,
EPA estimated that 100 percent of the LDGVs may be equipped
with "open-loop" (nonfeedback) systems under the revised
low-altitude standards. These systems do not require the more
sophisticated electronic microprocessor control of closed-loop
(feedback) systems which are often currently used to meet the
raore stringent 3.4 g/mi CO and 1.0 g/mi NOx standards.
Open-loop systems do not monitor the fuel/air ratio entering
the engine's combustion chambers as do closed-loop systems, but
instead use essentially fixed calibrations to control fuel
metering. Also, about 60 percent of the fleet is expected to
be equipped with air pumps and about 40 percent of the fleet
will use a less costly pulse air injection system. This
-------
XI-8
coiripares to current systems which almost always utilize an air
pump in conjunction with an oxidation-reduction catalyst or an
oxidation catalyst.
For LDDVs, the revised standards should be achievable
without the addition of any new emission control hardware.
More specifically, EPA expects that little or no EGR will be
required to achieve the revised NOx standards since little or
no EGR was required to meet similar Federal emission standards
in the 1980 model year (2.0 g/mi NOx), or the 1S81-82 model
years (waivers to 1.5 g/mi NOx were granted for some diesels).
B. Revised Ease Scenario Control Hardware
This section assesses the emission control technology that
will be required to make a low-altitude vehicle comply with
high-altitude proportional standards. This assessment will be
conducted in two parts. First, the requisite control hardware
that may be necessary to comply with the revised proportional
standards of 0.57 g/mi hC, 16 g/mi CO, and 1.5-2.0 g/mi NOx
will be identified. Second, the requisite hardware for
compliance with revised proportional standards of 0.57 g/mi HC,
11 g/mi CO and 1.5-2.0 g/mi NOx will be identified. The
difference in these two parts is simply that in one case the
revised CO standard is 16 g/mi, while in the other it is 11
g/mi.
1. Control Technology for Proportional Stanuaros
Including a 16 g/mi CO Requirement
As discussed in Chapters III and IX, the general control
strategy for reducing emissions from open-loop (nonfeedback)
vehicles to proportional levels is to recalibrate the fuel
metering device (e.g., carburetor) so that the excessively rich
fuel/air mixtures which normally occur at higher elevations are
corrected. This same emission control technique should be
useful in reducing emissions to the revised proportional levels
analyzed in this chapter because it corrects the same problem
of overly rich fuel/air mixtures in both cases (current versus
revised low-altitude standards). Also, in each case the
allowable high-altitude emissions from a vehicle certified at
low altitude will be limited to the same percent increase.
(This percentage is based on the ratio of the high-altitude
standard to the respective low-altitude standard.) Therefore,
the recalibration in Loth cases (i.e., leaner fuel/air
mixtures) should be somewhat similar. However, the emission
control systems for current standards are expected to be
different than for the revised standards and the effect this
may have on the effectiveness of the control technique which
has been identified needs further attention.
There are emission control devices that are currently in
use on vehicles which may not be part of the emission control
-------
XI-S
system if the standards are revised upward. These devices
include: 1) the NOx reduction catalyst on many gasoline-fueled
vehicles, and 2) the air pump on most of the gasoline-fueled
vehicles. The effects of removing these hardware items on the
effectiveness of recalibrating fuel/air ratios will be examined
separately below.
An exhaust catalyst basically functions by promoting
chemical reactions which remove a percentage of the harmful
emissions coming from the engine's combustion cylinders.
Therefore, as the amount of a particular pollutant entering the
catalyst is increased, only a portion of it will be removed,
the remainder passes through the catalyst into the atmosphere.
Under the current high-altitude standards, leaning the
fuel/air ratio to reduce HC and CO emissions will increase NOx
emissions coming from the combustion chambers. It is possible,
although unlikely with regard to proportional standards, that
this control strategy by itself could lead to excess NOx
emissions for some vehicles. The important point is that even
with the reduction catalyst the manufacturer of such a vehicle
would probably have to manipulate other engine control
parameters to reduce NOx emissions from the combustion chambers
because the reduction catalyst would remove only a portion of
the excess. These engine control parameters may include more
careful recalibration of the fuel metering to prevent excessive
leaning, recalibrating ignition timing, or recalibrating
exhaust gas recirculation (EGR) rates.
EPA assumes the degree of fuel management (i.e., leaning)
will be somewhat similar to that currently needed to comply
with proportional high-altitude regulations even if the
standards are revised upward, because both types of standards
(current versus revised) limit the increase in emissions from
low- to high-altitude by the same percentage. Although
manufacturers are not expected to use reduction catalysts to
comply with revised standards, they would be confronted with
essentially the same task as is currently the case with these
devices if NOx emissions from some vehicles were to increase
beyond acceptable levels. That is, the emissions coming out of
the engine need to be controlled since, as described above, the
existence of the reduction catalyst by itself would not
completely control the additional NOx. Therefore,
manufacturers will have to pay more careful attention to fuel
management, ignition timing, and EGR rates if inoeeo NOx
emissions increase beyond allowable levels when the fuel/air
ratio is leaned to control HC and CO. It should be emphasized,
however, that under the revised proportional standards (whether
they include an 11 g/mi or 16 g/mi CO standards), EPA expects
that few if any vehicles will have excessive NOx emissions as a
result of leaning the fuel/air ratio to achieve the requisite
HC and CO control.
-------
XI-10
Now that the effects of removing the reduction catalyst
have been discussed, the effects of removing the air pump and
replacing it with a pulse air injection system will be
assessed. The air pump adds fresh air (oxygen) to the engine's
exhaust which promotes the removal (oxidation) of HC and CO.
Under the current proportional standards the air pump, which is
mechanically driven by a belt from the engine, may be part of
the high-altitude modification on some vehicles by changing the
pump's drive pulley so that it rotates faster and delivers more
oxygen to the exhaust stream. since an air pump delivers a
particular volume of air at any given speed, this technique may
be used to recover some of the lost effectiveness of the
injection system when it is used at higher elevations where the
air is less dense and, hence, there is less oxygen per volume
of air. The effectiveness of recalibrating the air pump,
however, is limited. Care must be taken not to increase the
injection rate of fresh air to such a degree that the exhaust
gas temperature is reduced too much. This would reduce the
chemical reacton rate and could lead to greater HC and CO
emissions. Therefore, modifying the air pump would only
provide a portion of the total emission reduction which may be
required; indeed, this portion may be small because of the
limited effectiveness of increasing the pump's delivery rate.
Manufacturers must rely on the other control options which have
already been identified (e.g., fuel management and spark
timing) to ensure compliance with the current proportional
standards.
The pulse air injection system (PAIR) that is expected to
replace the currently used air pump system on some vehicles if
the standards are revised also reduces HC and CO emissions by
adding fresh air to the exhaust. The PAIR system is driven by
the pressure pulses in the exhaust stream which are caused by
the hot gases existing the combustion chambers. This system
does not possess the wide range of air delivery rates that
characterize the air pump and its effectiveness can be limited
for this reason. Like an air pump, PAIR delivery rates may be
somewhat increased as part of a high-altitude modification when
the pulse air system is not originally functioning at peak
performance. If such an increase is not sufficient, the
situation is analogous to the air pump, and manufacturers will
have to pay more careful attention to optimizing air/fuel
ratios, timing, etc. Overall, EPA expects that few, if any,
vehicles equipped with PAIR will have trouble meeting
proportional standards of 0.57 g/mi HC and 16 g/mi CO.
In summary, the recalibration techniques that are used to
comply with the current proportional standards (i.e., primarily
fuel management) also should be useful to achieve revised
proportional standards since in both cases the increase in
emissions from low to high altitude is limited by the same
percentage. The change in emission control technology which is
-------
XI-11
expected to accompany revised low-altitude standards should not
significantly affect these control techniques or the requisite
hardware.
2. Control Technology for Proportional Standards
Including an 11 g/mi CO Requirement
In the preceding section, EPA concluded that the requisite
control technology for complying with proportional standards
under the revised base scenario should be essentially the same
as projected for compliance with the current proportional
standards which were previously analyzed in Chapters III and
IX. This conclusion is primarily based on the similarity in
emission control which will be~ required by the two sets of
standards. This section investigates why essentially the same
control hardware should allow compliance with either an 11 g/mi
CO standard or a 16 g/mi CO standard at high altitude.
The assumed technical similarity between an 11 g/mi CO and
a 16 g/mi CO standard is based primarily on EPA's recent
experience with 1S82 model year LDVs that have received a CO
waiver from the current 3.4 g/mi CO low-altitude standard up to
7.0 g/mi. Waivers are allowed by the Clean Air Act in cases
where manufacturers can demonstrate that meeting a 3.4 g/mi CO
standard at low altitude would be technically infeasible
considering the cost of control. Vehicles which are granted a
low-altitude waiver automatically qualify for a less stringent
11 g/mi CO standard at high-altitude instead of the normal
proportional 7.8 g/mi CO standard.[1]
Experience with these waivered vehicles is of value in
estimating the difficulty of meeting high-altitude standards
since the waivered low-altitude vehicles are currently required
to meet the same CO standard as is assumed to be in effect at
low-altitude in this analysis of revised standards for 1984 and
later vehicles (i.e., 7 g/mi). Also, the current high-altitude
standard for waivered vehicles is equivalent to the revised CO
standard at high altitude, which is assumed in this chapter
(i.e., 11 g/mi). During the waiver process, manufacturers must
submit detailed information to EPA on the cost and technical
feasibility of complying with the statutory low-altitude
requirements in order to justify a less stringent standard.
EPA's Motor Vehicle Emissions Laboratory has not received any
comments from manufacturers of waivered vehicles stating that
compliance with the 11 g/mi CO standard at high altitude would
be technically difficult or that unique emission control
hardware would be required beyond that needed to comply with
the current proportional CO standard of 7.8 g/mi for unwaivered
vehicles.
• Of course this is not conclusive evidence that all 1984
and later vehicles will comply,, with an 11 g/mi CO standard
using essentially the same control hardware as would be needed
-------
XI-12
for compliance with a 16 g/mi CO standard. For example, it is
possible that some vehicles equipped with nonfeedback
(open-loop) emission controls systems using pulse air injection
systems (PAIR) may have more difficulty meeting an 11 g/mi CO
standard than a 16 g/mi CO standard because the effectiveness
of the PAIR system may be degraded by the lower air density at
higher elevations, as previously discussed. However, this
problem is only speculative at this time. On the other hand,
it is possible that any vehicle having greater difficulty in
meeting the more stringent 11 g/mi CO standard may be eligible
for exemption from the high-altitude certification requirements.
The above discussion illustrates that uncertainities
regarding the requisite control hardware remain to be
resolved. Based on the best available evidence, however, it is
EPA's judgment that tor the vast majority of 1984 and later
high-altitude vehicles, meeting an 11 g/mi CO standard will not
be significantly more difficult than complying with the current
proportional standard for 1982 and 1983 vehicles and, hence,
essentially the same control hardware should be required.
The emission control hardware for nonfeedback (open-loop)
LDGVs complying with fixed-point proportional standards (the
base scenario) was previously estimated in Chapter III. Since
LDGVs complying with the revised proportional standards (the
revised base scenario) are assumed to utilize the same control
hardware, the estimates in Table III-4 are applicable to this
analysis also. The information in that table, reproduced here
in Table XI-2, shows that nonfeedback LDGVs are expected to
require recalibrating and the addition of one fixed-step
aneroid to the engine's fuel metering system. An aneroid is a
simple mechanical device which senses changes in atmospheric
pressure and properly adjusts the fuel/air ratio entering the
engine's combustion chambers. This device is used on a
high-altitude vehicle to prevent excessively lean fuel/air
ratios which may cause engine damage it these vehicles are
driven at lower elevations. (Chapter III contains a detailed
discussion of this natural phenomenon.) No change to the
evaporative emission system is required to comply with the
proportional standard.
The requisite control technology for LDDVs was previously
estimated in Chapter VIII for the base scenario. Compliance
with proportional standards are expected to require limiting
the maximum fuel rate and possibly a change in fuel injection
timing (Table XI-2). Both of these involve physical
adjustments to existing hardware.
C. Revised Scenario 2 Control Hardware
The procedure used to assess emission control requirements
for the revised statutory standards in this section is similar
to that already presented for the revised proportional
standards.
-------
XI-13
Table XI-2
High-Altitude Emission
Control Hardware for the Revised Control Scenarios
Revised
Scenario Vehicle Control Hardware Description Fleet Affected
Base LDGV 1 aneroid 100%
LDDV Limit maximum fuel rate 100%
Possible fuel injection timing 100%
change
Scenario 2 LDGV 3 aneroids 100%
AIR system replacing PAIR system 40%
Larger evaporative cannister 100%
LDDV Limit maximum fuel rate 100%
Fuel injection timing change 100%
-------
XI-14
Like the proportional standards, complying with the
revised statutory standards at low and high altitude is
expected to require essentially the same control hardware as
estimated in previous chapters for meeting the current
statutory standards because of their similarity. In Chapter
III, EPA estimated that nonfeedback LDGVs are expected to
utilize additional recalibrations and two aneroids in addition
to the one aneroid that will be required in the base scenario
for a total of three (Table XI-2). The additional aneroids
will control engine parameters such as the power enrichment
circuit of the carburetor, EGR rate, or air injection rate. As
alluded to earlier, aneroids ensure proper engine operation if
a high-altitude vehicle is driven at lower elevations. As
before, the relevant question is: Will these control techniques
retain their effectiveness even though the basic emission
control systems that are currently in use will change if the
standards are revised upward?
Removal of the reduction catalyst should not significantly
change the emission control techniques that were identified
above. Even with a reduction catalyst, any modifications made
for high-altitude NOx control will focus on the engine since
changes to the catalyst system (e.g., increased noble metal
loading) are uneconomical. These engine-related modifications
(e.g., EGR and timing) will attempt to maintain high-altitude
engine-out NOx emissions at low-altitude levels/, since any
increase in engine-out levels would result in some increase out
of the catalyst. This normally would be unacceptable if
increased NOx levels from the catalyst jeopardized complying
with the emission standard at high altitude. However, the
presence of the reduction catalyst could give some added
flexibility since a moderate increase in engine-out NOx levels
should be reduced to a much smaller increase by the catalyst.
In cases where there is a significant margin available at
low altitude for increased NOx emissions, the small increase
seen after the catalyst may be acceptable where the larger
increase occurring in front of the catalyst would not be. In
these cases, the removal of the catalyst could require the
addition of more EGR under revised low-altitude standards to
meet the high-altitude standards than would be needed for the
same high-altitude control under the current low-altitude
standards. However, these cases should arise ver^ infrequently
and the absolute increase in EGR and any resulting fuel economy
effect should be quite small, if measurable at all. Thus,
removal of the reduction catalyst should not have any
significant negative impact on the fleet's fuel economy average.
To comply with revised statutory standards, manufacturers
are expected to replace currently used air pump injection
systems with PAIR systems on many gasoline-fueled vehicles. As
previously stated, the effectiveness of air infection systems
at higher elevations may be degraded because of the accompaning
-------
XI-15
reduced air densities. While air pump systems have the ability
to regain their effectiveness by increasing the amount of
oxygen available to the catalyst, PAIR systems are not as
flexible and may not have this capability.
Under revised proportional standards this should present
no significant difficulty to manufacturers because modifying
the air pump at high altitude would only provide a small
emissions benefit which could be offset by more carefully
recalibrating other engine control parameters when PAIR is used
as a replacement. However, complying with statutory standards
for some vehicles may require that all engine parameters be
completely optimized for low emissions. If this is the case,
even the small loss in effectiveness caused by the use of PAIR
instead of AIR may not be recoverable since further
optimization of other parameters may be impossible. Therefore,
the lost effectiveness of the PAIR may only be recoverable by
replacing such systems with mechanically-driven air pumps as
part of the high-altitude modification. Although the extent of
this possible replacement is unknown, to be conservative EPA
will assume that all vehicles equipped with PAIR in the revised
base scenario 2 will have the systems replaced with AIR to meet
the more stringent requirements in the revised scenario 2.
This affects 40 percent of the LDGVs (Table XI-2. Complying
with the statutory evaporative HC standard at high altitude is
expected to require a larger evaporative emission storage
capacity. One way to accomplish this is to increase the
charcoal loading of the existing storage cannister (Table
XI-2).
There is very little that can be done to diesel engines to
reduce high-altitude emissions other than carefully controlling
the fuel injection. Of course, more exotic forms of control
are available such as turbocharging, but that has been ruleo
out in this analysis because of its excessive cost. In essence
then, the statutory levels of this scenario will have to be
achieved by further limiting the maximum fuel rate for low
emissions beyond that alreaay required to meet the proportional
levels in the revised base scenario, and also by recalibrating
the fuel injection timing. This should be sufficient to bring
most LDDVs into compliance. The reason for this optimism is
that there should be little or no EGR use at the 1.5-2.0 NOx
control level which could otherwise exacerbate HC and CO
emissions at higher elevations by reducing the oxygen which is
available to support combustion in the engine's cylinders, as
may be the case under the current NOx standard of 1.0 g/mi.
(Chapter IX contains a more detailed explanation of this
phenomenon.) Of course, some LDDVs may be difficult to control
to statutory levels but this problem should be mitigated by
providing more exemptions than in the revised base scenario.
The control technology for each revised scenario is
summarized in Table XI-2.
-------
XI-16
IV. ECONOMIC IMPACT
This section analyzes the economic impact of implementing
the revised scenario 2 (i.e., fixed-point statutory standards
with revised levels), rather than continuing with a form of the
base scenario (i.e., fixed-point proportional standards with
revised levels). The incremental cost analysis is composed of
four distinct parts: 1) manufacturing costs, 2) consumer user
costs, 3) total national or aggregate costs, and 4)
socioeconomic impact.
The first three parts deal directly with the incremental
costs that are associated with modifying LDVs to achieve the
revised statutory standards at high altitude. These
incremental costs are defined as the differences between the
costs associated with modifying vehicles to comply with the
proportional standards in the revised base scenario and those
associated with modifying vehicles to meet the statutory
stanoards in the revised scenario 2. The fourth part deals
with the effect that higher purchase prices may have on
high-altituce vehicle sales.
This analysis uses the same basic methodology as
previously documented in Chapter V. To avoid needless
repetition, the detailed explanation in that chapter will not
be repeated here. Instead, this section will include only
general descriptions of the methodology as needed. For a more
complete discussion of the analytical techniques, Chapter V
should be consulted.
A. Costs to Manufacturers
Manufacturing costs can be separated into two categories:
variable and fixed. Each of these cost categories are
discussed below.
1. Variable Costs
This category includes all of the cost components which
contribute to the incremental purchase price of a high-altitude
vehicle except for the cost of development and certification
which is regarded as a fixed cost. Variable costs are
presented in terms of the retail price equivalent (RPE) and
presume that the emission control hardware is added to the
vehicle on the assembly line. These costs are estimated
separately for LDGVs and LDDVs.
The incremental emission control technology which will be
needed by LDGVs to meet the statutory levels of the revised
scenario 2 was previously estimated to include the addition
of: 1) two fixed-step aneroids on 1GC percent of the vehicles,
2) the replacement of the normally used pulse air injection
system (PAIR) with an air pump injection system (AIR) on 40
-------
XI-17
percent of the vehicles, and 3) the use of a larger evaporative
emission storage cannister on 100 percent of the vehicles.
The RPE for two aneroids was estimated in Chapter V to be
approximately $14 to $18 dollars. The unit cost for replacing
the PAIR system with the AIR system as original equipment can
be estimated by following the methodology contained in Chapter
V and using the cost information contained in an EPA report by
Lindgren.[6] The cost of the AIR system is computed to be
about $40 per unit.
No adjustment is necessary to reflect the small production
volume of these units for high-altitude sales, as was done for
aneroids, since the economy of scale has already been achieved
by using the same equipment on about 60 percent of the
low-altitude sales volume. The PAIR system which would
normally cost about $6 will not be used, so this expense is
credited toward the cost of the high-altitude modification.
The RPE of adding an air pump to a vehicle as original
equipment is, therefore, the cost of the AIR system ($40) minus
the cost of the PAIR system ($6) or about $34 in 1981 dollars.
The RPE for increasing the vehicle's evaporative emission
storage capacity was estimated in Chapter V to be about $2 to
$3 and would be the same here.
The RPE for the average high-altitude LDGV is sum of the
sales-weighted costs for the various hardware items. Aneroids
are expected to be installed in 100 percent of the
high-altitude vehicles so its sales-weighted cost is $14 to $18
per LDGV (1.00 x $14 to $18). AIR systems will very likely be
required on 40 percent of the high-altitude vehicles so its
sales-weighted cost is $14 per vehicle (0.40 x $34). A larger
evaporative storage capacity is estimated to be required by 100
percent of the high-altitude vehicles so its sales-weighted
cost is $2 to $3 per vehicle (1.00 x $2 to $3). By adding
these sales-weighted costs, the incremental RPE for the average
high-altitude LDGV is $30-$35.
The emission control technology that will be needed by
LDDVs to comply with the requirements of the revised scenario 2
was previously estimated to include adjusting both the maximum
fuel limiter and the fuel injection timing. Thus, no added
haroware should be required. These adjustments are easily
performed on the assembly line and so there should be little or
no measurable increase in the RPE of a high-altitude LDDV.
2. Fixed Costs
Based on the analysis in Chapter V, two categories of
fixed costs may be affected by complying with statutory versus
proportional standards at high altitude: development and
certification. That chapter concluded that increases would
occur in both cost categories. However, this would not be true
-------
XI-18
when both the low-altitude and high-altitude standards are
changing in both the statutory ana proportional cases which
would be the situation here.
This analysis assumes that the revised base scenario could
be implemented in 1984. Since this scenario has different
standards (0.57 g/mi HC, 11 g/mi CO, and 1.5-2.0 g/mi NOx) than
the present proportional standards (0.57 g/mi HC, 7.8 g/mi CO,
and 1.0 g/mi NOx), the automotive fleet would have to be
redeveloped and then recertified if the revised base scenario
were adopted. Furthermore, there should be essentially no
difference in the cost of development and certification for
either proportional or statutory high-altitude standards, as
discussed further in Appendix III. Because of this, the fixed
cost increment of implementing the revised scenario 2 in 1984
is zero since the expenditures for development and
certification would already have to be maoe.
B. Cost to Users
The incremental cost to users of high-altitude vehicles is
composed of the initial increase in purchase price and the
change in fuel economy and maintenance expenditures which
accrue during the life of the vehicle. Complying with the
revised scenario 2 will increase the purchase price of an
average LDGV by $30 to $35. For an LDDV, there should be no
increase in purchase price as a result of complying with the
revised scenario 2. Stated as an average for all LDVs, the
sticker price increase will be about $25 to $30 based on the
sales data previously aiscussed in Chapter IV (represented in
Table XI-3).
A fuel economy benefit is expected from the additional
recalibration allowed by the use of two additional aneroids at
high altitude to comply with statutory standards. In Chapter V
this benefit was calculated to be $80 if discounted over the
life of the vehicle (present value in 1981 dollars). It is
assumed that this fuel economy improvement also applies in this
chapter because the same emission control strategy (e.g.,
leaning the fuel/air ratio) will be used to comply with both
the current statutory and revised statutory standards at high
altitude. The fuel savings should occur for every
high-altitude LDGV built in compliance with the revised
scenario 2, since each vehicle is assumed to be equipped with
the two additional aneroids.
The change in fuel economy that may accompany the use of
an AIR system on some high-altitude vehicles that otherwise
would not require it is less clear. The air pump is
mechanically driven by the vehicle's engine and, hence, absorbs
some power that is then not available to power the vehicle.
Adding an air pump can, therefore, result in a fuel economy
loss. Often, however, when an AIR system is used at
-------
XI-19
Table XII-3
Hi^h-Altitude Sales Preset ions
of Light-Duty Vehicles
(thousands of vehicles)
Year Gasoline-Fueled Diesel-Powered
1984 293 30
1985 287 36
1986 276 45
1987 265 53
1988 262 56
5-Year Total 1,383 220
-------
Xl-20
low-altitude, the settings of certain engine parameters which
may have been compromised for emission control purposes can be
optimized for better fuel economy because of the added
flexibility provided by more air being available to the
catalyst. If an AIR system is used in this manner, no fuel
economy change may be observed.
The question then is: Will manufacturers recalibrate
high-altitude vehicles requiring the addition of an air pump
for optimum fuel economy? The answer is unknown at this time.
It is possible that manufacturers will not take the opportunity
to optimise these sytems because of the small high-altitude
market (only 3 to 4 percent of the total). Because of this
possibility, a worst case will be assumed in this analysis
where the benefit of using aneroids is offset by the potential
penalty of using AIR systems as part of a high-altitude
modification. This fuel economy offset affects 40 percent of
the high-altitude vehicles. Therefore, the fuel economy
benefit of $80 for the revised scenario 2 will only apply to 60
percent of the high-altitude LDGVs. Thus, the fuel savings for
the average LDGV is $46 (0.60 x $80). No fuel economy
increment was found in Chapter X for LDDVs complying with the
statutory standards and there is no reason to expect this to
change with a revision of the low-altitude standards. Thus,
none is included in this chapter.
No increment in maintenance costs are expected to occur as
a result of implementing the revised scenario 2. Aneroids
should not require maintenance during the life of the vehicle
(Chapter VI). The possibility that AIR systems may require
maintenance was explored in a recent rulemaking action and the
conclusion was that any increment would be insignificant.[7]
Therefore, the maintenance costs of the revised statutory
standards should be 2,ero.
The net cost to users, then, is the purchase price
increase, less any fuel economy benefit. For LDGVs, the fuel
economy Lenefit ($46) overwhelms the initial purchase price
increase ($30 to $35) and results in a net savings to the
consumer of $13 to $18. For LDDVs, the net cost is zero since
there are no changes in purchase price, fuel economy or
maintenance. Stated as an average for all LDVs, the overall
net savings is $11 to $15.
It is apparent that the fuel economy increment for LDGVs
dominates the results of the cost analysis. Therefore, the
predicted fuel economy gain deserves to be discussed in greater
detail before proceeding. In Chapter VII, this topic was
explored and the conclusions from that chapter should apply
here as well. The fuel economy increment associated with the
use of two aneroids is based on a very limited amount of test
data. Because of this, the estimated effects on fuel economy
-------
XI-21
should be used with some caution and must be carefully
reevaluated as additional information becomes available.
The analysis of statutory high-altitude LDV standards in
this chapter is even more sensitive to the estimated fuel
savings than in previous chapters. This is accounted for by
the change in emission control technology that is expected to
be used to comply with either the current statutory standards
(scenario 2 in the previous chapters) or the revised statutory
standards (the revised scenario 2 in this chapter). More than
twice as many vehicles are assumed to experience a fuel economy
benefit in this chapter than in previous chapters. By virtue
of this great sensitivity, the fuel benefit must be used with
discretion in this analysis.
Therefore, the uncertainty in the fuel economy benefit for
vehicles that use two additional aneroids and do not incur the
possible fuel economy offset from adding an air pump as part of
their high-altitude modification (60 percent of the LDGV fleet)
will be accounted for in the remainder of the analysis by using
a fuel savings range of $0 to $80. The cost to users of
high-altitude vehicles is restated as follows using this range
as a basis for the calculation. For the average LDGV the
overall net cost is -$18 to $35. The net cost for the average
LDDV remains unchanged at $0. The average net cost for all
LDVs is -$15 to $30.
C. Aggregate Cost to the Nation
The aggregate cost to the nation is defined as the present
value of the incremental costs of implementing the revised
scenario 2 for the first five years (i.e., 1S84 through 1988).
For this time period, the LDV production volumes were
previously discussed in Chapter IV, and are presented in Table
XI-3. Using these sales projections and the costs as discussed
above, the 5-year aggregate cost is $32 to $40 million if no
fuel economy benefit is assumed and -$15 to -$23 million if a
fuel economy savings is assumed (Table XI-4).
D. Socioeconomic Impact
The incremental purchase price for the average
high-altitude LDV was found to be about $30. As part of the
1982 and 1983 interim high-altitude rulemaking, the impact of a
$42 price increase was reviewed and it was concluded that such
an increase should not affect a dealer's sales or a consumer's
ability to purchase a vehicle.[2] Since the price increment
that was estimated in this chapter ($30) is less than the
increment analyzed in the interim rulemaking ($42), this
conclusion also should be true for the revised scenario 2.
Thus, the sticker price increase for vehicles built on the
assembly line should not unduly affect high-altitude sales.
-------
Table XI-4
Summary of Costs for a Combined Fleet
of LDGV and LDDV Under the Revised Scenarios
Case
With Fuel Economy Bene-
fit
Without Fuel Economy
Benefit
Average
Hardware
Per
Vehicle
$25-30
$25-30
Development
Per Vehicle
$0
$0
Certification
Per Vehicle
$0
$0
Average
Fuel
Economy
Increment
Per
Vehicle
$0
-$41
Average
Net
Total
Per
Vehicle
$25 to 30
-$15 to -11
Capital
Costs to
Manufacturers
$0
$0
5-Year
Aggregate
Costs[a]
(millions)
$32 to 40
-$23 to -15
[a] Present value in 1984.
IX
M
I
hJ
hJ
-------
XI-23
There is one aspect of the revised scenario 2 that may
have a significant negative impact on the ability of low- and
high-altitude dealers to trade new vehicles among themselves.
As discussed in the technology section of this chapter, EPA has
made a worst case assumption that 40 percent of the
high-altitude fleet may require the addition of an AIR system
to meet the statutory standards in the revised scenario 2. If
this is indeed true, ana there is no data at this time to prove
or disprove the assumption, these particular vehicles may be
prohibitively expensive to modify from a low-altitude
configuration to a high-altitude configuration after
production. Replacing the PAIR system with an AIR system on a
vehicle after it is originally built could cost over $150.
When this cost is added to the others (e.g., changing the
carburetor), modifications may be quite expensive for certain
vehicles. However, more data is needed before a final judgment
or. this issue can be made.
V. ENVIRONMENTAL IMPACT
This section discusses two basic measures of environmental
impact. The first measure simply involves estimating the
emission reductions resulting from implementing the revised
scenario 2 which should accrue beyond those expected by
continuing proportional standards under the revised base
scenario. These overall reductions in LC, CO, and NOx
emissions are used primarily to determine the incremental cost
effectiveness of the revised scenario 2. The second measure of
environmental impact is an air quality modeling analysis which
projects the effect of the revised scenarios on high-altitude
air quality. These models are valuable aid evaluating the
ability of high-altitude Air Quality Control Regions (AQCRs) to
achieve and maintain compliance with the National Ambient Air
Quality Standards (NAAQS).
To avoid needless repetition of detail in this chapter,
the methodology for determining both the total emissions and
the air quality consequences will not be specifically
presented. Instead only a general Description will be
provided. More specific information is available in Chapter IV
which contains the environmental impact analysis of the
high-altitude control scenarios that correspond to the current
low-altitude standards.
A. Total Emissions
The incremental emissions benefit of implementing the
revised scenario 2 over continuing the proportional standards
of the revised base scenario is found by determining the total
emissions for each scenario and then taking their difference.
The gaseous and evaporative emission factors which are used to
calculate the total high-altitude emissions are shown in Table
XI-5.
-------
Table XI-5
Emission Rates for 1904-00 Light-Duty
Gasoline-Fueled and Diesel-Fueled Vehicles at 5,300 Feet l a J
Revised Base Scenario Revised Scenario 2
Pollutant LDGV IjDDV LDGV LDDV
HC Zero-Mile Emission
Level (g/iai) 0.36 0.54 0.26 0.39
Deterioration Rate
(9/ iai/10,000 miles) U.23 0 .03 0.23 0.03
50, 000-Mile Emission
Level (g/mi) 1.51 0.69 1.4i 0.134
CO Zero-Mile Emission
Level (g/mi) b.63 2.22 3.50 2.22
Deterioration Rate
(^/mi/10, 000 miles) 1.92 0 .05 i.92 0.u5
50 ,000-Miie Emission
Level (g/mi) 15.2 j 2.47 13.10 2.47
NOx Zero-Mile Emission
Level (g/mi) 0.95-1.27 1 .11-1.49 0 .95-1.27 1.11-1.49
Deterioration Rate
(y/wi/10,000 0.09-0.08 0.06 0.09-0.00 0.00
miles)
50,000-Mile Emission
Level (g/mi)[L>] 1.4-1 .67 1.41-1.79 jl.4-1.67 1.4j.-i,79
-------
Table XI-5 (cont'd)
Emission Rates tor I9y4-dtf Liyht-Duty
Gasol ine-Fueled and Diesel-Fueled Vehicles at 5,3UU Feet l a j
Revised Liase bceuuriO Revised Sce.idLiu 2
Pol lutant LDdV LDDV LDGV i-.UDV
Eva^j. Zero-Mile Emission
HC Level (y/mi) J . i3 — J. ill
Deterioration Rate
(^/rai/10, UUl) miles) U — I)
b0/U00-Miie Emission
Level (y/mi) U . 13 — U.lu
[a]Emission Rate = ZHL + (U) (DR)
Where: ZHL = Zero-Hile Level
M = Cummulative Mileaye/lU,UOU
DR = Deterioration Rate
[b] The 1.4 and 1.41 y/mi levels reflect a 1.5 y/mi WOx
standard, while the 1.67 and 1.79 y/mi levels reflect a
2.0 y/mi NOx standard.
-------
XI-26
The total emissions and incremental benefit of the two
scenarios are shown in Table XI-6 for LDVs produced during the
first five years of the standards (i.e.f 1984 through 1988).
Implementing the revised scenario 2 should result in an
incremental benefit of about 20,000 metric tons of HC, and
283,000 metric tons of CO. Of these totals, LDGVs account for
about 85 percent of the HC reductions and 100 percent of the CO
reductions.
B. Air Quality
This section reviews the air quality effects of
high-altitude automotive emission control strategies that are
based on revised low-altitude standards. This review will be
conducted in three parts. First, the incremental air quality
benefits of implementing the revised statutory standards
instead of the revised proportional standards will be
addressed. Second, the need or justification for the more
stringent revised statutory standards will be discussed.
Third, the impact of revising the low-altitude standards on
high-altitude air quality will be reviewed. After these parts
are presented, they will be summarized and conclusions
regarding the need for more stringent automotive standards will
be made.
The air quality projections in this section are based on
the use of computer models that were easily applicable to all
high-altitude areas. The reader is cautioned that such
computer studies utilize a variety of assumptions in an attempt
to forecast the events which will affect air pollution levels
in the future. As with any projections of this type, some of
the assumptions may prove to be invalid as better information
becomes available. Also, due to time and resource constraints,
many input data were national averages and not site-specific
values. Therefore, the modeling results are most useful for
determining the differences between control scenarios, while
the absolute values associated with any single strategy are
subject to greater error.
1. Incremental Effects
The air quality effects on each pollutant are discussed
separately based on Tables XI-7 through XI-10. Nitrogen oxides
(NOx) are not addressed here because there is no difference in
the emission rates associated with the two revised control
scenarios (Table XI-6). Hence, there would be no difference in
the modeling results.
For ozone, Table XI-7 shows that under high-growth cases
with inspection/maintenance (I/M) and both low- and high-growth
cases without I/M, implementing the revised scenario 2 is
projected to provide only a one percent improvement in the
ambient concentration for a few of the years. The results are
-------
XI-27
Table XI-6
Lifetime Emissions from 1984-88 LDVs
Sold Above 4,000 Feet
(lp3 metric tons)
Reductions
Revised Revised (Base Scenario
Base Scenario Scenario 2 minus Scenario 2)
Pollutant
LDGV
LDDV
LDGV
LDDV
LDGV
LDDV
Total
HC
208
15
195
12
13
3
16
CO
2,102
54
1,819
54
283
0
283
NOx
193-230
31-39
193-230
31-39
0
0
0
Lvap. HC
17.9
0
13.8
0
4.1
0
4.1
Total HC
226
15
209
12
17
3
20
-------
Table XI-7
Average Percent Reduction in Expected
Maximum 1-Hour Ozone Concentrations from 1979 Base
Year in Denver and Salt Lake City (low and high growth)[a]
With Inspection/Maintenance
Revised Base
Revised §2
Revised Base
Revised 92
1986
Low High
24-32 22-28
24-32 22-28
1986
Low High
20-27 18-23
21-27 18-24
1987
Low High
26-34 23-30
26-34 23-30
1987
Low High
22-29 19-25
22-29 19-25
1988
Low High
27-35 24-31
27-35 24-31
Low High
23-30 20-26
23-30 20-26
1989
Low High
28-36 24-32
28-36 24-32
Low High
24-31 20-27
24-32 21-27
1990
Low High
28-37 24-32
28-37 24-33
1990
Low High
24-32 21-28
25-33 21-28
1993
Low High
28-36 24-31
28-36 24-31
1993
Low High
25-33 21-27
25-34 21-28
1995
Low High
27-35 22-29
27-36 23-30
1995
Low High
25-32 20-26
25-33 20-26
Without Inspection/Maintenance
1988 1989
[a] Note that a range of values is reported here to reflect two different ratios of HC/NOx ambient concentrations,
as discussed in Chapter IV. Results from the higher ratio are reported first.
-------
Table XII-8
Number of Violations of Ozone NAAUS in
Denver and Salt Lake City (low and high growth)jaj
With Inspection/Maintenance
1986
1987
1988
1989
1990
199J
Low
High
LOW
H iqh
Low
High
Low
High
Low
High
Low
High
LOW
Hign
Revised
Base
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
U —1
U-l
0-2
Revised
#2
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
U-l
U-l
Without
Inspect
ion/Maintenance
1986
1987
1988
1989
1990
1993
199b
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised
Base
1-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
1-2
Revised
§2
0-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
1-2
0—J.
1-2
[a] Note that a range- of values is reported here to reflect two different ratios of HC/NOx ambient concentrations,
as discussed in Chapter IV. Results from the lower ratio are reported first.
X
M
I
to
vc
-------
Table XI-9
Average Percent Reduction in Expected
Second Highest 8-Hour CO Concentrations from 1979
Base Year in Six High-Altitude Cities (low and high growth)la]
With Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low High
Revised
Base
55
48
60
54
65
58
68
61
71
65
76
69
cc
Revised
#2
55
49
61
55
6b
58
69
6 J
72
65
77
7U
7a 71
Without
Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
LOW
High
Low
High
LOW
High
Low Hign
Revised
Base
46
38
51
43
56
48
60
52
64
56
71
62
73 64
Revised
f 2
46
38
52
44
57
48
61
53
65
57
11
63
74 65
TaT The cites investigated are Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque, and salt Lake City.
-------
Table XI-10
Number of Violations of CO NAAQS in
Six High-Altitude Cities (low and high growth)[a]
With Inspection/Maintenance
1986 1987 1988 1989 1990 1993 1995
Low High Low High Low High Low High Low High Low High Low High
Revised Base 7 18 28 03 01 00 00 00
Revised |2 7 16 27 03 01 00 00 00
Without Inspection/Maintenance
1986 1987 1988 1989 1990 1993 1995
Low High Low High Low High Low High Low High Low High Low High
Revised Base 20 43 10 26 3 16 1 8 0 3 0 0 0 0
Revised #2 20 43 9 24 3 14 0 5 0 2 0 0 0 0
[a] The cites investigated are Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque, and Salt Lake City. m
u>
-------
XI-.32
somewhat similar for the projected number of National Ambient
Air Quality Standard (NAAQS) violations tor ozone. Table xi-8
shows that the revised scenario could produce one less
violation in 1995 at high-growth rates with I/M. Without I/M
there is a potential for one less violation in 1986 under
low-growth and in 1990 under high growth.
Table XI-9 shows that the revised scenario 2 could reduce
the ambient concentration of CO by 1 to 2 percent in most years
under both high- and low-growth rates, with or without I/M.
Some reductions in the number of CO NAAQS violations are
associated with these reduced emission levels as shown in Table
XI-10. Without I/M/ two fewer violations are projected in 1987
and 1988/ three less in 1989, and one less in 1990 under
high-growth conditions. Under low-growth conditions, one less
violation is observed in 1987 and 1989. With I/M, two fewer
violations are projected in 1986 and one less in 1987 under
high-growth conditions. For the low-growth case no differences
are observed.
2 . Air Quality Needs
The justification for greater emission reductions at high
altitude can be evaluated by reviewing the projected dates of
compliance with the NAAQS. This will show whether
high-altitude areas should be in compliance by the 1987
statutory deadline and whether the implementation of the
revised scenario 2 could help attain the NAAQS sooner. As
before, the two pollutants of interest, HC and CO, will be
reviewed separately.
From Table XI-8, it is clear that although the number of
ozone NAAQS violations is small, there is a possibility that
the standard could never be attained by all high-altitude areas
under any of the cases studied. Therefore, although the HC
emission reductions associated with implementing revised
statutory standards will not by themselves guarantee attainment
of the ozone NAAQS, they may be needed along with reductions
from other sources to assure compliance with the ambient
standard in all high-altitude areas. Certainly, the additional
control of some HC sources in high-altitude areas will be
necessary for the ozone NAAQS to be attained if these
projections are accurate.
Attainment dates for the CO NAAQS are more variable (Table
XI-10). Without I/M, attainment by all high-altitude areas is
project to occur two years beyond the statutory deadline in
1989 with low-growth rates under the revised scenario 2.
Without it, compliance could be delayed until 1990. When
high-growth rates are assumed, attainment could occur sometime
in the early 1990's under both revised scenarios. With I/M,
the effects are similar to the without I/M case except that all
of the attainment dates are earlier. Attainment is predicted
-------
XI-33
by 1988 with low-growth rates and by 1990 with high-growth
rates, under both revised scenarios. Thus, implementing more
stringent standards appears to make no difference. Therefore,
statutory standards may only be effective in helping to attain
the CO NAAQS sooner in the absence of i/h. Implementing these
automotive standards alone, however, may not be adequate to
achieve the 1987 statutory deadline in any of the cases
analyzed.
As mentioned throughout this report, any conclusions based
on the absolute number of projected NAAfcS violations must be
conditional due to the potential errors involved. Only further
study of those areas just in or out of compliance with the
NAAC-S will yield firm conclusions in this area.
3. Comparison of Air Quality Under "Revised" Standards
and "Current" standards
The air quality effects of adopting high-altitude
standards that are based on revised low-altitude standards
("revised" scenarios) versus high-altitude standards that are
based on the current low-altitude standards ("previous"
scenarios) can be reviewed by comparing the information
presented in this chapter to that previously presented in
Chapter IV (Tables XI-11 through 14). Such a comparison shows
that there is no significant difference for ozone between the
two general types of standards (revised versus current). This
is to be expected since the emission standards for HC are the
same for the respective statutory (0.41 g/mi) and base (0.57
g/mi) control scenarios.
The same conclusion can be reached for CO with I/h, that
is, there is no difference between the number of violations
under the two types of standards (revised base versus current
base and revised scenario 2 versus current scenario 2).
Without I/h, however, the number of CO NAAQS violations uncier
the scenarios based on revised standards is generally less than
that under the previous scenarios based on current standards.
Looking at the NAA^S attainment dates rather than the number of
NAA^S violations, no difference between the two types of
standards is evident regardless of whether or not I/M is
implemented.
These results for CO appear surprising at first glance.
It seems unreasonable for the generally less stringent
standards of the revised scenarios to provide greater air
quality benefits than are associated with the scenarios that
are based upon current standards. However, this difference can
be explained by recalling the discussion in Chapter IV which
pointed out that many manufacturers are using computerized
feedback emission control systems to comply with the current
statutory standards. These systems currently exhibit a
significant rate of catastrophic failure. The CO emissions
from vehicles with such failed systems are very high which in
-------
Table XI-11
Average Percent Reduction in Expected
Maximum 1-Hour Ozone Concentrations from 1979
Base Year in Denver and Salt Lake City (low and high growth)[a]
1986
1987
With Inspection/Maintenance
1988 1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised Base
24-32
22-28
26-34
23-30
27-35
24-31
28-36
24-32
28-37
24-32
28-36
24-31
27-35
22-29
Revised f2
24-32
22-28
26-34
23-30
27-35
24-31
28-36
24-32
28-37
24-33
28-36
24-31
27-36
23-30
Previous Base
24-32
22-28
26-34
23-30
27-35
24-32
28-36
25-32
29-37
25-33
29-38
25-32
28-36
24-31
Previous §2
24-32
22-29
26-34
23-30
27-35
24-32
28-36
25-32
29-37
25-33
29-38
25-33
28-37
24-31
Without Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised Base
20-27
18-23
22-29
19-25
23-30
20-26
24-31
20-27
24-32
21-28
25-33
21-27
25-32
20-26
Revised #2
21-27
18-24
22-29
19-25
23-30
20-26
24-32
21-27
25-33
21-28
25-34
21-28
25-33
20-26
Previous Base
21-27
18-24
22-29
19-25
24-31
20-27
24-32
22-28
25-33
22-28
26-34
22-29
26-34
21-28
Previous |2
21-27
18-24
22-29
19-25
24-31
20-27
25-32
22-29
25-34
22-29
26-35
22-29
26-35
21-28
CO
[a] Note that a range of values is reported here to reflect two different ratios of IIC/NOx ambient concentrations,
as discussed in Chapter IV. Results from the higher ratio are reported first.
-------
Table XI-12
Number of Violations of Ozone NAAQS in
Denver and Salt Lake City (low and high growth)[a]
With Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
LOW
High
Low
High
Low
High
Revised Base
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-2
Revised #2
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
Previous Base
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-0
0-1
0-0
0-1
0-1
0-1
Previous §2
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-1
0-0
0-1
0-0
0-1
0-1
0-1
Without
Inspect!
on/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised Base
1-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
1-2
Revised #2
0-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
1-2
0-1
1-2
Previous Base
0-2
1-2
0-1
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
0-2
0-1
1-2
Previous i2
0-2
1-2
0-1
1-2
0-1
1-2
0-1
0-2
0-1
0-1
0-1
0-2
0-1
1-2
[a] Note that a range of values is reported here to reflect two different ratios of HC/NOx ambient concentrations,
as discussed in Chapter IV. Results from the lower ratio are reported first.
-------
Table XI-13
Average Percent Reduction in Expected
Second Highest 8-Hour CO Concentrations from 1979
Base Year in Six High-Altitude Cities (low and high growth)[a]
With Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised Base
55
48
60
54
65
58
68
61
71
65
76
68
78
71
Revised #2
55
49
61
55
65
58
69
63
72
65
77
70
79
72
Previous Base
55
48
60
54
64
58
68
61
71
65
76
69
78
70
Previous #2
55
49
61
55
66
59
69
63
72'
65
77
71
80
73
1986
1987
Without Inspection/Maintenance
1988 1989
19 90
1993
199 5
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised Base
46
38
51
43
56
48
60
52
64
56
71
62
73
64
Revised #2
46
38
52
44
57
48
61
53
65
57
72
63
74
65
Previous Base
45
37
51
42
55
47
60
51
63
54
70
61
72
62
Previous §2
46
38
51
43
57
48
61
53
64
56
71
62
74
65
X
n
I
u>
a\
[a] The cites investigated are Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque, and Salt Lake City.
-------
Table XI-14
Number of Violations of CO NAAQS in
Six High-Altitude Cities (low and high growth)[a]
With Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
Hiqh
LOW
High
Low
High
Revised Base
7
18
2
8
0
3
0
1
0
0
0
0
0
0
Revised §2
7
16
2
7
0
3
0
1
0
0
0
0
0
0
Previous Base
7
18
2
8
0
3
0
1
0
0
0
0
0
0
Previous #2
7
16
2
7
0
3
0
1
0
0
0
0
0
0
Without Inspection/Maintenance
1986
1987
1988
1989
1990
1993
1995
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Revised Base
20
43
10
26
3
16
1
8
0
3
0
0
0
0
Revised #2
20
43
9
24
3
14
0
5
0
2
0
0
0
0
Previous Base
21
43
10
28
3
17
1
9
0
4
0
0
0
0
Previous §2
20
43
10
26
3
14
0
8
0
3
0
0
0
0
fa]
The cites investigated are Denver, Colorado Springs, Ft. Collins, Greeley, Albuquerque, and Salt Lake City.
-------
XI-38
turn causes the average in-use emissions from this type of
vehicle to be much higher than would be expected based on the
standard it is certified to meet.
Under the revised standards, which are being analyzed in
this chapter, manufacturers are expected to comply with the
less stringent emission control requirements by using
nonfeedback systems. These systems do not exhibit the
catastrophic failures associated with current feedback
systems. Therefore, nonfeedback systems have lower average
in-use emissions despite the equivalent or higher emission
standards to which they are certified. This explains why the
modeling results generally show better air quality results
under the scenarios with "revised" standards when the benefits
of i/M are excluded. of course, it was also pointed out in
Chapter IV that the in-use catastrophic failure rates that are
now exhibited by feedback systems may be significantly reduced
in the future as more experience with these new systems is
gained. If this reduction takes place, the trend now shown in
the analysis may be reversed so that the high-altitude
scenarios based on current low-altitude standards show better
air quality results than the control scenarios based on revised
low-altitude standards.
Although the impact of the high-altitude NOx standards was
not relevant to the preceding air quality discussions, it is
important here, since under the current standards the allowable
level is 1.0 g/mi and under the revised standards the level may
be 1.5 to 2.0 g/mi. Tables XI-15 ana 16 contain both the
results for the "previous" scenarios (based on current
standards) and those for the "revised" scenarios (based on
revised standards) . Table XI-15 shows that the ambient NOx
concentrations are projected to be greater under the revised
scenarios. Table XI-16 shows, however, that when high-growth
rates are assumed, the NOx NAAQS may never be attained in
Denver beyond 1987 by either set of scenarios. For low-growth
rates, violations could begin in the early 1990's under the
revised scenarios and in the iuid-1990's if the previous
scenarios are implemented. Regardless of the growth rates, the
trend is toward an increase in violations near the end of this
century, although the actual number of violations is small.
4. Conclusions of the Air Quality Analysis
There appears to be a small incremental benefit in ozone
and CO air quality associated with controlling automotive
emissions beyond the levels prescribed by the revised base
scenario under some of the cases analyzed. In other cases,
there appears to be no significant difference between the
revised base scenario and the revised scenario 2. The current
air quality modeling studies indicate that additional emission
control may be needed beyond the levels of the revised base
scenario since all high-altitude areas may not attain the NAAQS
by the 1987 statutory deadline. However, further study would
-------
Table XI-15
Average Percent Increase in Expected NOx
Concentrations in Denver (low and high growth)[a]
1986 1987 1988 1989 1990 1993 1995
Low High Low High Low High Low High Low High Low High Low High
Revised
Scenarios 2 10 4 14 6 16 8 20 10 24 16 35 20 43
Previous
Scenarios 0 8 2 10 2 12 4 16 4 18 8 27 12 35
t a ] Note that this table reflects increases in pollutant concentrations rather than reductions as was the convention
for earlier tables on CO and ozone'.
-------
Table XI-16
Number of Violations of NOx NAAQS in Denver (low and high growth)
1986 1987 1988 1989 1990 1993 1995
Low High Low High Low High Low High Low High Low High Low High
Revised
Scenarios 00 01 01 01 01 11 11
Previous
Scenarios 00 00 01 01 01 01 11
X
M
£*
O
-------
XI-41
be needed to firmly make this conclusion. Finally,
implementing high-altitude standards that are based on revised
low-altitude standards rather than retaining the current
standards may have a small negative impact on NOx NAAQS
violations near the end of this century.
VI. COST EFFECTIVENESS
Cost effectiveness is one measure of the economic
efficiency of reducing air pollution. The incremental cost
effectiveness of the revised scenario 2 for HC and CO is found
by dividing half of the net cost per vehicle by the emission
reduction per vehicle for each pollutant. This computation is
performed in Table XI-17 for LDGVs, LDDVs, and a combination of
the two vehicle types. Controlling HC emissions from LDGVs
ranges up to $1,460 per metric ton. Carbon monoxide control
for these vehicles ranges up to $85 per metric ton. For LDDVs,
controlling both HC and CO costs nothing ($0) per metric ton
because no added hardware or fixed costs are involved. When
LDGVs and LDDVs are combined into a single control strategy,
controlling HC emissions ranges up to $1,250 per metric ton
while CO control ranges up to $85 per metric ton.
The wide range of incremental cost-effectiveness values
displayed in Table XI-17 for LDGVs and the combination of LDGVs
and LDDVs under the revised scenario 2 are caused by the
inclusion or exclusion of the estimated fuel economy benefit
that may accompany implementation of revised statutory
standards at high altitude. The low estimates which include
the fuel economy benefit are tenuous at this time because of
the uncertainties associated with estimated change in fuel
consumption.
In the worst case analyzed (i.e., no fuel economy
benefit), scenario 2 is nearly twice as costly per ton of HC
than the most expensive control strategy shown in Table XI-18.
For CO, it is more comparable to the other strategies. On the
other extreme (i.e., inclusion of the fuel economy benefit),
the revised scenario 2 is more cost effective than all but one
of the other control strategies. Until additional information
becomes available with which to more precisely define the cost
effectiveness of this scenario, implementing revised statutory
standards at high altitude should be considered a viable, but
unproven, alternative to the revised base scenario.
VII. SUMMARY
The analysis in this chapter supports the conclusions of
the earlier chapters. Statutory standards at high altitude
appear to provide a small but real air quality benefit in a
potentially cost-effective manner and should be seriously
considered in the overall program to reduce pollution in
high-altitude areas. Nevertheless, different conclusions are
possible under scenarios which assume other revised
low-altitude standards.
-------
XI-42
Table XI-17
Incremental Cost Effectiveness of Revised Scenarios
Vehicle
Low
LDGV
-18
LDDV
0
LDGV
and
LDDV
-15
Cost (dollars)
per vehicle)[a]
High
35
0
30
Emission
Reductions
(10~3 metric
tons per vehicle)
HC CO
12
14
12
205
0
176
Incremental
Cost Effectiveness
(dollars/metric ton)
HC
Low High
neg. 1,460
0 0
CO
Low High
neg,
85
0
neg. 1,250 neg.
85
Ta~] The low estimates for gasoline-fueled vehicles are tentative at
this time because of the uncertainties associated with the
estimated fuel economy benefit.
-------
XI-43
Table xi-18
Cost-Effectiveness Comparison With
Other Emission Control strategies
Emissions
After
Control
Cost
Baseline
Strategy Effectivenes
Control Program
Emissions [a]
Implemented
HC
CO
LDV Statutory[2]
HC
0.9
HC
0.41
734
67
Standards
CO
15
CO
3.4
LDV 1/M[3]
—
—
640
58
L'DT 1984 [4]
HC
1.7
HC
0.8
195
1'5
Standards
CO
18
CO
10
HDE 1984
HC
1.5
HC
1.3
Standards[5][b]
CO
25
CO
15.5
(gasoline)
HC
1.5
HC
1.3
305
10
CO
25
CO
15.5
(aiesel)
HC
1.5
HC
1.3
325
—
CO
25
CO
15.5
Motorcycle
HC
9
HC
8-22.5 [c]
582
Neg
Stanaards[6]
CO
34.67
CO
27.4
HDG Evap.[7][d]
hC
1.8
HC
0.17
200
Interim 1982-83
HC
1.47(cars)
HC
1.33(cars)
393
12
HA Standards [8]
4 .19 (trucks)
3 .78(trucks)
CO
16.23(cars)
CO
13.21(cars)
73 . U2(trucks)
55.65(trucks)
[a] Emission levels are in g/mi except for the hDE 1584 standaros
which are in grams per brake-horsepower-hour.
[b] The baseline ana after control strategy emission values were
based on different test procedures (see Reference 4 in
Chapter VI).
[c] Slicing scale based on engine displacement (cubic
centimeters).
[d] The evaporative standard is in terms of g/test and converted
to g/mi here to facilitate comparison.
-------
XI-44
References
1. "Control of Air Pollution from New Motor Vehicles
and New Motor Vehicles Engines; Final High-Altitude Emission
Standards for 1982 and 1983 Model Year Light-Duty Motor
Vehicles," U.S. EPA, 45 FR 66984, October 8, 1980 .
2. "Final Regulatory Analysis - Environmental and
Economic Impact Statement for the 1982 and 1983 Model Year
High-Altitude Motor Vehicle Emission Standards," U.S. EPA,
OANR, OMS, ECTD, SDSB, October 1980.
3. "Update on the Cost Effectiveness of Inspection and
Maintenance," U.S. EPA, OANR, OMS, ECTD, IMS, EPA-AA-IMS/81-9,
April 1981.
4. "Summary and Analysis of Comments on the Notice of
Proposed Rulemaking for High-Altitude Emission Standards for
1982 ana 1983 Model Year Light-Duty Motor Vehicles," U.S. EPA,
OANR, OMS, ECTD, SDSL, October 1980.
5. "Technical Feasibility of the Proposed 1982-83
High-Altitude Standards for Light-Duty Vehicles and Light-Duty
Trucks," CTAB/TA/80-3, U.S. EPA, OANR, OMS, ECTD, SDSB, August
1980.
6. "Motor Vehicle Emission Standards for Carbon
Monoxide ana Nitrogen Oxides," U.S. EPA, OANR, OMS, ECTD, SDSB,
April 1981.
7. "Cost Estimations for Emission Control Related
Components/Systems and Cost Methodology Description," Rath and
Strong, Inc., Lindgren, Leroy H., EPA-480/3-78-002, March 1978.
8. "Regulatory Analysis and Environmental Impact of
Final Emission Regulations for 1984 and Later Model Year
Light-Duty Trucks, U.S. EPA, OANR, OMS, ECTD, SDSB, May 1980.
-------
Chapter XII
Conclusions and Recommendations
This report examined various high-altitude control
scenarios, based on the current emission standards, for their
effects on LDGVs, LDGTs, and LDDs. Generally, the existing
proportional high-altitude emission standards for these vehicle
classes have proven valuable in cost effectively improving the
air quality of high-altitude areas. At a minimum, these
standards should be continued for 1984 and later model years.
I. EFFECTS OF THE ALTERNATIVE SCENARIOS BASED ON CURRENT
STANDARDS
A. LDGVS
EPA considered six alternative scenarios to continuing the
current fixed-point proportional standards for LDGVs. The
costs of these alternatives vary with their technical
requirements, which in turn are based on three basic factors:
1) the maximum elevation for which control must be
demonstrated; 2) the extent of exemptions, if any; and 3) the
level of standards. Based on these three factors, the Agency
concludes that:
1) continuing the current statutory high-altitiude
requirements, as mandated in section 206 of the Act, is
extremely costly, may significantly limit model availability at
both low and high altitudes, and is extremely cost ineffective;
2) there is no air quality justificaton tor controls
above 6,000 feet;
3) any statutory standards at high altitude can provide
a small incremental improvement in air quality;
4) exemptions, or similar waivers, can significantly
reduce compliance costs, while maintaining acceptable model
availability at higher elevation;
5) exemptions, or similar waivers, can prevent the
potential for adversely affecting model availability
throughtout the nation that may accompany implementing
statutory standards at higher elevations beginning in 1964, as
required by the Clean Air Act; and
6) fixed-point statutory standards which require
vehicles sold above 4,000 feet to comply with the standards
when tested at 5,300 feet and which provide for some exemptions
are the most cost-effective alternative beyond continuing the
current proportional requirements, of the six alternatives
-------
XII-2
analyzed. Of course, there are many other possible
alternatives, one of which may be better than any of the six
analyzed here.
B. LDGTs
While not specifically required by the Act, EPA finds that
controlling LDGTs in addition to LDVs results in a positive
impact on the ambient air quality of high-altitude areas.
Controlling light trucks to statutory standards would reduce
vehicle emissions of HC by 50 percent more and of CO by 40
percent more than if LDGVs were controlled alone. in addition,
the Agency finds that controlling truck emissions under
fixed-point statutory standards is more cost effective than the
same degree of control for LDGVs.
C. LDDS
The Agency analyzed high-altitude standards for both
gaseous emissions and particulate emissions for LDDs. For
gaseous emissions, EPA concludes that achieving fixed-point
statutory standards should be no more difficult for aiesel
engines than for gasoline engines, and that the cost over a
5-year period should be small. The Agency finds that
particulate emissions will be reduced by the same techniques
that reduce gaseous emissions, although it is too early to
determine if the statutory particulate standards can be met
with these techniques alone. Also, controlling gaseous
emissions from LDDs to statutory high-altitude standards is
expected to be more cost effective than controlling LDGVs.
D. Effects of the Scenarios Based on Revised Low-Altitude
Standards
The Agency also considered the effects of the same six
alternative scenarios under revised low-altitude standards.
The major difference is that under revised fixed-point
statutory standards of 0.41 g/mi HC, 7.0 g/mi CO, and 1.5-2.0
g/mi NOx for passenger cars, the control options might tend to
reduce model availability somewhat more than they would under
the current fixed-point statutory standards. Otherwise, EPA
concludes that:
1) the technical difficulties of compliance would
remain about the same;
2) exemptions, or similar waivers, would retain their
positive effects on overall costs and model availability; and
3) fixed-point statutory standards would likely be the
most cost-effective alternative to proportional standards.
-------
XII-3
Based on the assumption that revised LDV standards at low
altitude are as stated above (i.e., 0.41 g/mi HC, 7.0 g/mi CO,
and 1.5-2.0 g/mi NOx), these conclusions would remain valid for
both gasoline-fueled and diesel-powered cars and trucks.
Nevertheless, different conclusions are possible under
scenarios which assume other revised low-altitude standards.
E. Recommendations
EPA recommends that section 206 of the Clean Air Act be
amended to:
1) Provide the Administrator the flexibility to adopt
two-car compliance strategies, and to establish high-altitude
standards, within the range from proportional to statutory, for
any class of motor vehicles that is necessary to attain the
NAAQS for ozone and carbon monoxide after considering the
technical feasibility, impact on model availability, and
economic impact of any such requirements; and
2) Confirm the Administrator's authority to exempt
certain vehicles from the high-altitude certification
requirements or waive the high-altitude standards for certain
vehicles, and to decide on the maximum number of such
exemptions or waivers.
-------
API-1
Appendix I
Perspective on the Interim High-Altitude Standards
The 1982 and 1983 high-altitude standards were promulgated
on October 8, 1980. [1] During the rulemaking process, the
proportional gaseous emission standards contained in those
regulations were the subject of intense analysis and public
debate. EPA believes that little would be served to again
present the detailed analysis[2,3,4] which supports that final
rulemaking in this document since the value of fixed-point
proportional standards has already been demonstrated. Instead,
EPA chose to concentrate on determining if any alternative
control options could provide greater protection of the public
health and welfare, while at the same time remaining cost
effective.
However, it is important to familiarize the reader with
the details of the interim high-altitude standards since these
facts form the basis for one of EPA's recommendations in this
report which calls for the authority to continue proportional
standards at high altitude if, after further study, more
stringent controls are found to be unnecessary or not cost
effective.
Some of the topics discussed in this appendix are
discussed in various chapters of the report but will be briefly
repeated here for clarity. The information in this section was
taken from EPA's final regulatory analysis of the 1982 and 1983
high-altitude regulations.[2]
I. COSTS TO MANUFACTURERS
At the time the 1982 and 1983 high-altitude standards were
promulgated, manufacturers were expected to incur increased
costs in three main areas: development, certification, and
emission control hardware. These costs are summarized in Table
API-1. The total cost to manufacturers is $23.36 million
(undiscounteo, 1981 dollars).
II. COST TO USERS
As a result of this regulation, users of high-altitude
motor vehicles were expected to pay an average of $22 more for
light-duty vehicles (LDV) and $39 more for light-duty trucks
(LDT) in 1982 than in 1981 (1981 dollars) . Stated as a
combined average, the increase for a high-altitude motor
vehicle would be $25 (1981 dollars). Furthermore, there would
be no change in maintenance costs, but a small positive effect
on fuel economy was expected, although because of a lack of
data, no fuel economy benefit was included in the final
regulatory analysis. If such a benefit had been included, the
total cost of the regulation would be less.
-------
API-2
Vehicle
Category
LDV
Subtotal
Year
1981
1S82
1983
Table API-1
Total Cost to Manufacturers
for the 1982 and 1983 Model Years[a]
Development Cost
(million dollars)
4.78
4.78
9.56
Hardware
Cost Total
Certification (million (million
Cost dollars) dollars)
341,200
341,200
682,400
1.2
1.2
2.4
5.12
6.32
1.2
12.64
LDT
Subtotal
Total
1981
1982
1983
2.95
2.S5
5.SO
15.46
113,700
113,700
227,400
909,800
2.3
2.3
4.6
7.0
3.06
5.36
2.3
10.72
23.36
[a] Unaiscounted, 1981 dollars.
-------
API-3
III. IMPACT ON HIGH-ALTITUDE DEALER
As a result of the interim standards, EPA estimated that
50 percent of the 1,000 high-altitude dealerships would lose
about one sale during the 2-year period (1982-83) because of
higher vehicle prices. The remaining dealers would not be
adversely affected. EPA also estimated that "dealer trades"
between high- and low-altitude dealerships would not be unduly
affected because low-altitude cars must be capable of being
modified to meet the high-altitude standards at a reasonable
cost if they could not automatically do so.
IV. AGGREGATE COST TO THE NATION
The present value of the expected costs of the interim
regulations are shown in Table API-2. The aggregate cost of
$23.98 million is equivalent to a lump sum investment maae at
the beginning of 1962.
V. AIR QUALITY IMPROVEMENTS
Tables API-3 and API-4 show the change in Denver, Colorado
area emissions and the total pollution reduction at high
altitude which were expected as a result of the 1982 and 1S83
proportional standards. By 1987, when Denver must be in
compliance with the National Ambient Air Quality Standards, HC
emissions would be reduced by 1.0 percent and CO would be
reduced by 3.4 percent. The total air quality benefit was
estimated to be a reduction of 33,100 tons for HC and 1,195,000
tons for CO.
VI. COST EFFECTIVENESS
Table API-5 summarizes the total pollution reductions,
total cost, and resulting cost effectiveness of the
high-altitude proportional standards. Under these regulations,
HC and CO were expected to be. reduced in high-altitude areas at
a cost of $365 per ton and $10 per ton, respectively.
Expressed in metric tons, the cost would be $393 for HC ana $12
for CO.
-------
API-4
Table API-2
Year
Aggregate
1981 and 1S83
Cost to the Nation for
High-Altitude Standards[a]
Development
Cost
(million
dollars)
Certification
Cost
Hardware
Cost
(million
dollars)
Total
Discount
Factor
Discounted
Total
(million
dollars)
1981 7.73 457,000 8.19 1.10 9.01
1982 7.73 457,000 3.5 11.69 0.0 11.69
1983 3.5 2.5 0.91 3.19
Total 23.98
[a] Present value in 1982, 1981 dollars, 10 percent discount
rate.
-------
API-5
Table API-3
Denver Area Emissions (tons/day)
1980 1982 1984 1S87
HC CO EC CO HC CO HC CO
Without Stas 231.7 1,927 196.8 1,687 162.2 135.8 133.7 1,011
With Stds 231.7 1,927 196.3 1,670 160.8 131.2 132.4 977
Reduction 0 0 0.5 17 1.4 4.6 1.3 34
(percent) (0) (0) (0.3) (1.0) (0.9) (3.4) (1.0) (3.4)
-------
API-6
Table API-4
Total Pollution Reductions for 1582 and 1983
(thousands of tons)
HC CO
LDV 11.6 258
LDT 21.5 537
Total 33.1 1,155
-------
API-7
Table API-5
Cost Effectiveness for the High-Altitude
Control Strategies
Cost
Reductions Cost[a] Effectiveness
Pollutant (thousands of tons) (million dollars) (dollar/ton)
HC 33 12 365
CO 1,195 12 10
[a] The total cost to the nation is divided equally between
the pollutants (1981 dollars).
-------
API-8
References
1. "Control of Air Pollution from Nevv Motor Vehicles
and New Motor Vehicle Engines; Final High-Altitude Emission
Standards for 1982 and 1983 Model Year Light-Duty Motor
Vehicles," U.S. Environmental Protection Agency, 45 FR 66984,
October 8, 1980.
2. "Final Regulatory Analysis - Environmental and
Economic Impact Statement for the 1982 and 1983 Model Year
High-Altitude Motor Vehicle Emission Standards," U.S. EPA,
OANR, OMS, ECTD, SDSB, October 1980.
3. "Summary and Analysis of Comments on the Notice of
Proposed Rulemaking for High-Altitude Emission Standards for
1982 and 1983 Model Year Light-Duty Motor Vehicles," U.S. EPA,
OANR, OMS, ECTD, SDSB, October 1980.
4. "Technical Feasibility of the Proposed 1982-1983
High-Altitude Standards for Light-Duty Vehicles and Light-Duty
Trucks," CTAB/TA/80-3, U.S. EPA, ONAR, OMS, ECTD, SDSB, August
1980.
-------
APII-1
Appendix II
Supplemental Information for the Environmental Analysis
This appendix contains supplemental information to the
environmental analysis. Contained are the individual model
year emission rates for each high-altitude strategy analyzed
(Tables APII-2 through APII-4), the low-altitude emission rates
for scenarios 1/ 2 and the base scenario (Table APII-5), the
annual growth rates for stationary and off-highway sources
(Table APII-6), and the air quality analyses performed for CO
and NOx based on zero-growth rates (Tables APII-7 through
APII-14).
The reader should be aware of some differences between the
convention used to identify the various scenarios in this
appendix and that used in the report. The relationship between
these conventions is explained in Table APII-1.
-------
APII-2
Table APII-1
High-Altitude Report to Congress Control Scenarios
Appendix II
Convention
2
3
4
5
9
10
Report
Convention
1 (without truck
control)
2 (without truck
control)
3 and Base
(without truck
control)
2 (with truck
control)
3 and Base (with
truck control.)
2 (with revised
standards
3 and Base (with
revised stan-
dards
Not discussed,
included for
completeness
No standards
No standards
1984+ Standards [a]
LDV
LDT
-/-/-
-/-/-
.41/3.4/1.0
.41/3.4/1.0
.57/7.8/1.0
.41/3.4/1.0
.57/7.8/1.0
.41/7.0/2.0
.57/11.0/2.0
.57/16.0/2.0 -/-/-
Technology Mix
for LDGV [b,c]
40.8/5S.2/0
10.2/59.2/30.6
10.2/59.2/30.6
.8/10/2.3 10.2/59.2/30.6
1.0/14/2.3 10.2/59.2/30.6
-/-/-
-/-/-
-/-/- 0/0/100
-/-/- 0/0/100
0/0/100
-/-/- 10.2/59.2/30.6
-/-/- 0/0/100
[a] When no specific standards are given, the LDGT standards
at low altitude are assumed to be .8/10/2.3. The strategy
9 low-altitude standards for LDGVs are assumed to be
.41/3.4/1.0 and the standards assumed are .41/7.0/2.0 for
strategy 10.
[b] The technology mix for the LDGTs is 100 percent oxidation
catalyst (open-loop carbureted).
[c] The LDGV technology mix of x/y/z indicates that x percent
of fleet is assumed to be closed-loop carbureted, y
percent of the fleet is expected to be throttle body fuel
injected, and z percent of the fleet is assumed to be
open-loop carbureted.
-------
APII-3
Table
APII-2
Hydrocarbon Emission
. Rates at
5,300 Feet
/ehicle
Emission
Rate[a]
Type
Model Year
ZM
DR
Standard[b]
Scenar io
LDGV
1981
.59
.21
All
1982-83
.48
.21
.57
All
1984 +
.38
.18
.41
1
.34
.19
.41
2,4
.44
.19
.57
3,5
.26
.23
.41
6
.36
.23
.57
7,8
.55
.19
( .41)
9
.47
.23
( .41)
10
LDGTs
1984 +
1.13
0.14
( .8)
1-3,6-10
0.63
0.14
.8
4
0.78
0.14
1.0
5
HDGV
1979-83
6.90
0.32
(1.5)
All
1984-85
2.67
0.22
(1.3)
All
1986 +
2.34
0.22
(1.3)
All
LDDV
1984 +
0.39
0.03
.41
1/2,4,6
0.54
0.03
.57
3,5,7,8
0.90
0.03
( .41)
9,10
LDDT
1984 +
1.40
0.06
( .8)
1-3,6-10
0.61
0.06
.8
4
0.76
0.06
1.0
5
HDDV
1984-85
8.03
0.04
(1.3)
All
1986 +
6.83
0.04
(1.3)
All
[a] Note emission factors, EF, are calculated from EF = ZM +
(DR)Y, where ZM is the zero-mile emission rate, DR is the
deterioration rate per 10,000 miles and y is the number of
miles divided by 10,000.
[b] () = low-altitude standard.
-------
APII-4
Table APII-3
Carbon Monoxide Emission Rates at 5,300 Feet
/ehicle
Emission
Rate[a]
Type
Model Year
ZM
DR
Standard[b]
Scenario
LDGV
1981
10.78
3.07
(3.4/7)
All
1982
8.13
3.12
7.8/11
All
1983
7.75
3.12
7.8
All
1984+
5.22
2.76
3.4
1
3.75
2.24
3.4
2,4
6.25
2.22
7.8
3,5
3.58
1.92
7.0
6
5.63
1.92
11.0
7
8.19
1.92
16.0
8
7.44
2.21
(3.4)
9
11.39
1.92
(7.0)
10
LDGTs
1984 +
22.69
1.35
(10)
1-3,6-10
7.13
1.35
10
4
9.85
1.35
14
5
HDGV
1979-83
324.55
8.37
(2.5)
All
1984-85
104.16
5.63
(35)
All
1986 +
82.29
5.63
(35)
All
LDDV
1984 +
1.27
0.05
3.4
1,2,4
2.22
0.05
7.8
3,5
2.22
0.05
7.0
6
2.22
0.05
11
7
2.22
0 .05
16
8
2.22
0.05
(3.4)
9
2.22
0.05
(7.0
10
LDDT
1984 +
3.47
0.09
(10)
1-3,6-10
1.98
0.09
10
4
2.77
0.09
14
5
[a] Note emission factors, EF, are calculated from EF = ZM +
(DR)Y, where ZM is the zero-mile emission rate, DR is the
deterioration rate per 10,000 miles and y is the number of
miles divided by 10,000.
[b] () = low-altitude standard.
-------
APII-5
Table API1-4
Oxides of Nitrogen Emission Rates at 5/300 Feet
Vehicle Emission Rate[a]
Type
Model Year
ZM
DR
Standard[b]
Scenario
LDGV
1981
0.52
0.09
All
1982-83
0.57
0.10
1.0
All
1984 +
0.54
0.10
1.0
1
0.58
0.10
1.0
2-5
1.27
0.08
2.0
6-8
0.52
0.09
(1.0)
9
0.99
0.06
(2.0)
10
0.95
0.09
1.5
10 with
1.5 Std.
LDGTs
1984 +
0.98
0.03
(2.3)
1-3,6-10
1.26
0.04
2.3
4,5
HDGV
1984 +
7.55
0.09
(10.7)
All
LDDV
1984 +
0.75
0.05
1.0
1-5
1.49
0.06
2.0
6-8
0.75
0.05
(1.0)
9
1.49
0.06
(2.0)
10
1.11
0.06
(1.5)
10 with
1.5 std.
LDDT
1984 +
1.89
0.07
(2.3)
1-3,6-10
1.89
0.07
2.3
4-5
HDDV
1984 +
22.90
0.12
(10.7)
All
[a] Note emission factors, EF, are calculated from EE = ZM +
(DR)Y, where ZM is the zero-mile emission rate, DR is the
deterioration rate per 10,000 miles and y is the number of
miles divided by 10,000.
[b] () = low-altitude standard.
-------
APII-6
Table APII-5
Selected Emission Rates at 1,800 Feet
for 1984 and Later Model Years
Pollutant
HC
Vehicle Type
LDGV
Emission Rate[a]
ZM
0.37
0.33
PR
0.17
0.18
Scenario
1
2,3
LDDV
0.39
0.03
All
CO LDGV 4.27 2.13 1
3.21 1.88 2,3
LDDV 1.27 0.05 All
NOX LDGV 0.56 0.10 1
0.63 0.11 2,3
LDDV 0.75 0.05 All
[a] Note emission factors, EF, are calculated from EF = ZM +
(DR)Y, where ZM is the zero mile emission rate, DR is the
deterioration rate per 10,000 miles and y is the number of
miles divided by 10,000.
[b] () = low-altitude standard.
-------
APII-7
Table APII-6
Annual Growth Rates for Stationary and Off-Highway Sources
Growth Rates 1577-95
Pollutant Source Zero Growth Low Growth High Growth
Off-Highway
N/A[a]
+ 2.5
+ 2.5
Stationary Area
N/A
+0.0
0.0
Petroleum
N/A
+ 0.8
+1.9
Storage
N/A
+ 0.8
+1.9
Industrial
Processes
N/A
+ 0.8
+ 3.1
Other Solvent
N/A
+ 0.8
+ 0.8
Industr ial
Surface
N/A
+0.8
+ 3.3
Off-Highway
+2.5
+ 2.5
+ 2.5
Stationary Point
+ 2.4
+ 2.4
+ 2.4
Combustion
+0.8
+ 0.8
+ 0.8
Stationary Area
0.0
0.0
0.0
Off-Highway
+ 2.5
+ 2.5
+ 2.5
Stationary Point
+3.5
+ 3.5
+ 3.5
Combustion
+ 0.8
+0.8
+ 0.8
Stationary Area
0.0
0.0
0.0
[a] Not Applicable. A zero-growth rate case was not analyzed
for ozone.
-------
Table APII-7
Average Percent Change in tne Ambient CO Concentration Level
froi.i the Base Year Witn Inspection/Maintenance
(aero-yrowth rate)
Scenar io
Description
1986
±987
±988
±909
^.yyu
±y 93
j.9 95
1
Continuous All Alt. Stds.
-52
-57
-62
-66
-69
-74
-7 0
2
Fixed
Pt. Ail Alt. Stus.
-52
-5 8
-G3
-t>7
-69
-75
-77
3
Fixed
Pt. Proportional Stds.
-52
-57
-61
-65
-69
-74
-76
4
Fixed
Pt. All Alt. Stds.
-53
-58
-63
-6 8
-7 U
-76
-78
5
Fixed
Pt. Proportional Stds.
-52
-58
-62
-6 6
-69
-75
-77
6
Fixed
Pt. All Alt. Stds.
-52
-58
-62
-67
-69
-74
-77
7
Fixed
Pt. Proportional Stds.
-52
-57
-62
-65
-6y
-74
-76
8
Fixed
Pt. Proportional Stds.
-51
-57
-61
-65
-67
-72
-75
-------
Table APII-8
Total Number of CO NAAyS Violations With Inspection/Maintenance
(zero-yrowth rate)
Scenar io
Descr iption
1*86
1987
1988
I98y
iyyu
iyy3
xyya
1
Continuous All Alt. Stds.
i3
6
2
u
u
u
u
2
Fixed
Pt. All Alt. Stus.
13
2
0
u
u
u
3
F ixed
Pt. Proportional Stds.
13
6
3
u
u
u
i)
4
Fixed
Pt. All Alt. Stus.
12
5
2
u
u
u
0
5
F ixed
Pt. Proportional Stds.
13
b
2
u
u
u
u
6
Fixed
Pt. All Alt. Stds.
13
5
2
u
u
u
u
7
Fixed
Pt. Proportional Stds.
13
6
2
u
u
u
u
8
Fixed
Pt. Proportional Stds.
13
6
3
i
u
u
u
-------
Table APII-9
Averaye Percent Change in the Ambient CO Concentration Level
from the Base Year Without Inspection/Maintenance
( zero-growth rate)
Scenar io
Description
1986
1987
1988
1989
199U
1993
j.995
1
Contii
nuous All Alt. Stds.
-42
-48
-53
-57
-61
-6o
-69
2
Fixed
Pt. All Alt. Stds.
-43
-48
-54
-58
-62
-6y
-72
3
Fixed
Pt. Proportional stds.
-42
-48
-53
-57
-61
-68
-7u
4
Fixed
Pt. All Alt. Stds.
-44
-4y
-55
-ou
-0 3
-7 U
-7 3
5
Fixed
Pt. Proportional stds.
-43
-49
-53
-58
-62
-69
-7 x
6
Fixed
Pt. All Alt. Stds.
-43
-49
-54
-59
-62
-72
7
Fixed
Pt. Proportional Stds.
-43
-48
-53
-58
-62
-69
-71
8
Fixed
Pt. Proportional Stds.
-42
-48
—o 3
-57
-61
-67
— 7 U
-------
Taole APII-10
Total Number of CO NAAQS Violations Without Inspection/Maintenance
(zero-yrowth rate)
Scenar io
Descr iption
1986
1987
1988
1989
1990
1993
1995
1
Continuous All Alt. Stds.
31
18
8
3
1
0
U
2
Fixed
Pt. Ail Alt. Stds.
30
16
7
3
1
U
U
3
Fixed
Pt. Proportional Stds.
31
18
8
J
i
0
U
4
Fixed
Pt. All Alt. Stds.
27
15
7
I
1
0
U
5
Fixed
Pt. Proportional Stds.
30
15
7
3
J.
0
0
6
Fixed
Pt. All Alt. Stds.
3U
15
7
3
1
0
0
7
Fixed
Pt. Proportional Stds.
30
16
8
3
jL
0
0
8
Fixed
Pt. Proportional Stds.
31
18
8
3
1
u
U
-------
Table APII-11
Average Percent Chanye in Ambient NOx Concentration Level
from the Base Year With Inspection/Maintenance
(zero-growth rate)
Scenario
Description
1986
1987
1988
1989
199U
1993
1995
1
Continuous All Alt. Stds.
-2
-2
0
U
U
4
6
2
F ixed
Pt. All Alt. Stds.
-2
-2
U
U
2
4
6
3
Fixed
Pt. Proportional Stds.
-2
-2
U
U
2
4
6
4
Fixed
Pt. All Alt. Stds.
-2
U
U
U
2
4
6
5
Fixed
Pt. Proportional Stds.
-2
U
U
U
2
4
0
6
Fixed
Pt. All Alt. stds.
0
2
2
4
6
1U
12
7
Fixed
Pt. proportional Stds.
U
2
2
4
0
1U
12
8
Fixed
Pt. Proportional Stds.
U
2
2
4
6
iU
12
-------
Table APII-12
Total Number of NOx NAAQS Violations With Inspection and Maintenance
( zero-yrowth rate)
Scenario
Descr ipt ion
1986
1987
lyaa
1989
199 U
1993
ly9b
1
Continuous All Alt. Stds.
U
U
U
U
U
U
u
2
Fixed
Pt. All Alt. Stds.
U
G
u
0
U
' U
u
3
F ixed
Pt. Proportional Stds.
0
0
u
U
U
U
0
4
F ixed
Pt. All Alt. stds.
0
0
u
0
u
0
u
5
Fixed
Pt. Proportional Stds.
0
U
u
U
u
0
u
6
Fixed
Pt. All Alt. Stds.
u
0
u
u
u
0
1
7
F ixed
Pt. Proportional Stds.
u
U
u
u
u
0
1
8
Fixed
Pt. Proportional Stds.
u
U
0
u
u
u
X
-------
Table APII-13
Average Percent Chanye in the Ambient WOx Concentration Level
from the Base Year Without Inspectio/i'laintenance
(zero-vdrowth rate)
Scenario
Description
iy86
iy«7
lybd
iyay
iyyu
iyy^
iyy5
1
Continuous All Alt. Stds.
-2
-2
0
u
u
4
b
2
Fixed
Pt. Ail Alt. Stds.
-2
-2
0
u
2
4
6
3
Fixed
Pt. Proportional Stds.
-2
-2
0
u
2
4
6
4
Fixed
Pt. All Alt. Stds.
-2
0
u
u
2
4
0
5
Fixed
Pt. Proportional Stds.
-2
U
u
u
2
4
D
6
Fixed
Pt. All Alt. Stds.
0
2
2
4
6
J.U
12
7
F ixed
Pt. Proportional stds.
0
2
2
4
6
±0
12
3
Fixed
Pt. Proportional Stds.
U
2
2
4
b
10
±2
-------
Table APII-14
Total Number of NOx NAAQS Violations Without Inspection/Maintenance
(zero-yrowth rate)
Scenar io
Description
1986
1987
1988
1989
1990
1993
1995
1
Continuous All Alt. Stds.
U
0
U
U
U
U
U
2
Fixed
Pt. All Alt. Stds.
U
0
0
U
0
U
U
3
Fixed
Pt. Proportional Stds.
0
0
U
U
u
U
U
4
Fixed
Pt. All Alt. Stds.
U
0
0
0
J
U
U
5
Fixed
Pt. Proportional Stds.
0
0
0
U
u
U
U
6
Fixed
Pt. All Alt. Stds.
0
U
U
U
0
U
1
7
Fixed
Pt. Proportional Stds.
0
0
U
U
u
U
1
8
Fixed
Pt. Proportional Stds.
0
u
U
U
u
U
1
M
i
<_n
-------
APIII-1
Appendix III
Supplemental Information for the Economic Analysis
I. DEVELOPMENT COSTS
All vehicles except those using "unmodified electronic
feedback systems" will require a unique calibration for high
altitude. Calibrations are historically developed through a
series of reiteration involving theoretical studies, carburetor
flow bench testing, and Federal Test Procedure (FTP) testing.
FTP testing is by far the most expensive portion of any
calibration effort; therefore, development costs can be
adequately characterized by conservatively estimating the
average number of FTP tests required per engine family.
Estimating the number of FTP tests required for compliance
with the standards is a problem. In reality the number is
likely to be different for each engine family because of the
variety of emission control systems and because calibrations
within an engine family will require different degrees of
development effort.
The difficulty of estimating the necessary development was
not diminished by manufacturers' comments to the 1982 and 1983
interim standards, which is a major source of data for this
study. Despite the fact that many manufacturers made repeated
claims that high-altitude testing facilities were inadequate, a
statement that should have been based on an estimate of the
requisite development testing, only one manufacturer provided
specific information. Therefore, in order to estimate the
quantity of development testing, EPA relied primarily on its
own experience with development programs at the Motor Vehicle
Emissions Laboratory and on the past experience of its
technical staff while they were employed in development areas
of the automobile industry.
Ford estimated that 52 high-altitude calibrations would be
needed and that 150 FTP tests would be required per
calibration. EPA's independent estimate is in basic agreement
with Ford. Historically, developing a low-altitude calibration
can indeed take 150 tests. However, it is unlikely that such a
great number of tests would be required to develop a suitable
high-altitude calibration. EPA reasons that calibrating
high-altitude hardware will be less difficult for several
reasons.
Typically, low-altitude calibrations are determined
simultaneously. Such a development program provides no
opportunity to learn from prior experience with similar
calibrations within the same engine family. Because special
durability and emission-data vehicles will not be required,
manufacturers will often develop high-altitude calibrations
-------
APII1-2
after the low-altitude hardware has' been determined. The
experience and information that were generated in producing
low-altitude calibrations can then be used to reduce the effort
required to develop high-altitude calibrations for the same
vehicle configurations. Also, the overall technical problem is
greatly reduced since the basic changes that must be made to
compensate low-altitude hardware for the effects of higher
altitude are generally well known.
Furthermore/ the actual number of calibrations per engine
family may be lower for high-altitude vehicles than for
low-altitude vehicles. Manufacturers may develop many more
low-altitude calibrations than are actually required because
the potential low-altitude market is so great that the
resulting small improvements in ariveability and fuel economy
(CAFE) justify the additional development costs. This amount
of optimization may not be needed or justifiable for the
smaller high-altitude market, i.e., one calibration may suffice
for several low-altitude calibrations. In this situation the
"worst case" calibration for several vehicle configurations
within an engine family will be developed first, and, if
suitable for other similar configurations, will be used unless
time, financial resources, and perceived benefit dictate
otherwise. Even though manufacturers may provide fewer
calibrations and, therefore, less optimization at high
altitudes as compared to low altitude, high-altitude consumers
will still benefit from the development work which will be
done. High-altitude vehicles should perform better and give
better fuel economy than unadjusted low-altitude vehicles
operated at high altitudes with much richer fuel-air mixtures.
Although no details were given, Ford may have based their
estimate of 52 high-altitude calibrations on the fact that less
optimization would be required for the high-altitude market
than the low-altitude market. In 1980, Ford certified 20
light-duty motor vehicle engine families. This figure and
Ford's estimate of high-altitude calibrations translates into
about 2.5 calibrations per engine family. This is in contrast
to Ford's 1S80 certification data which shows an average of
perhaps 10 calibrations per engine family. Therefore, it is
reasonable to conclude that Ford expects significantly less
optimization at high altitude than at low altitude.
EPA estimates that, on the average, 100 FTP tests per
engine family should be sufficient to calibrate a light-duty,
gasoline-fueled vehicle. This, of course, assumes that some
calibrations will be more difficult to develop than others and
some will be less difficult. It appears that feedback systems
should generally be easier to calibrate than many non-feedback
(aneroid) systems. Additionally, some non-feedback systems are
also expected to be quite easy to calibrate. Manufacturers'
comments to the 1S82 and 1983 interim package indicated that
some vehicles could comply with the standards by manipulating
-------
APIII-3
adjustable parameters on existing low-altitude Hardware.
However, to be conservative, EPA will use 150 tests per engine
family to determine the manufacturer's development costs due to
this regulation. The additional 50 tests will allow for
expenses that are not explicitly accounted for in this
analysis. These expenses include costs for additional
engineering support at the manufacturers' heau^uarters,
building prototype hardware, and bench testing.
The number of LDV engine families to be certified for 1984
is, of course, unkown at this time. EPA has assumed that
approximately the same number of engine families wili be
certified in 1984 as was certified in 1980. In 1980 there were
109 nonCalifornia LDV gasoline engine families certified.
These include families for sale in either the 49 states,
excluding California, or the 50 states, including California.
Engine families which are certified for sale in California only
have been excluded because these proposed regulations uo not
apply to tnose vehicles. Thus, the maximum number of families
that could undergo engine development is 109 families, anu this
would occur in scenarios la and 1c. In scenarios where some
vehicles are exempted from sale at high altitude (scenarios u,
2, 3a and the base scenario) the exempt engine families will
not have to undergo development. As is explained in tiie
certification section below, it will be assumed that the
fraction of engine families which will not be certified for
high altitude (i.e., exempted) will be the same as that
fraction of sales exempted. Also, under scenarios 3a, 3b, and
the base scenario, some vehicles will not have to undergo
development because these vehicles have systems with inherent
capability to compensate for the effects of altitude. The
percentage of these vehicles which will not undergo development
is shown in Tables V-2 and V-3 of Chapter V. Again, the
percentage of engine families not having to undergo development
is assumed to be the same as the percentage of sales. Thus,
for scenarios lb, 2, 3a, 3b, and the base scenario, less than
the total amount of families appearing in any given model year
will need special high-altitude development.
There is, of course, the possibility of carryover from the
base scenario of emission-data results from 1983 to 1984,
thereby reducing the amount of development testing required in
1984. Since all vehicles sold at high altitude in 1982 or j.9ti3
by definition fall under the base scenario there will be no
development costs for this scenario in 1984. It is uniikeiy,
however, that manufacturers could apply tue development results
of the base scenario to each of the other scenarios. Tnus, to
be conservative, it is assumed that carryover of emission-data
results in the base scenario will not be used when determining
the costs for all other scenarios in cases where vehicles are
modified. For unmodified vehicles in scenarios 3a and 3b,
carryover is assumed as would occur under the base scenario.
-------
APIII-4
Now that the testing requirement has been estimated, the
price per test remains to be determined before the cost of
development can be found. Information obtained from commercial
testing facilities located in Denver, Colorado, indicate that a
manufacturer may run a development quality FTP test for about
$375. Of course, the cost for manufacturers with their own
private facilities will be less. In calculating the cost of
development, EPA will use $500 per test to provide an adequate
allowance for engineering and technical support, and prototype
vehicle shipping expenses.
Table V-8 shows the development costs for the families
with unique high-altitude calibrations. Again, the base
scenario requires no development. Development costs for
scenarios other than the "base scenario" are estimated to be a
total of $3.38-8.18 million.
II. CERTIFICATION
Certification for this high-altitude standard will begin
in 1984 for light-duty vehicles. The certification cost will
differ according to each prescribed scenario and according to
the certification procedure which is ultimately adopted for
1984 and later model years. For this analysis, EPA has assumed
that certification will be similar to that currently used in
the interim high-altitude program (i.e., actual vehicle tests
will be conducted).
Under scenarios with exemptions, exempted vehicle families
are prohibited from sale at high altitude and, therefore, will
not require to be certified to meet a high-altitude standard.
For fixed-point scenarios, manufacturers will be allowed to use
their low-altitude 4,000-mile data vehicles by modifying these
vehicles into the selected high-altitude configuration. For
the continuously proportional or statutory standards,
manfacturers must certify a 4,000-mile data vehicle with the
same configuration at low and high altitude.
In all scenarios, manufacturers will not be required to
build and accumulate mileage on special high-altitude
certification vehicles. Deterioration factors for
high-altitude vehicles will be the same as those developed with
low-altitude, 50,000-mile durability vehicles. In EPA's
emission factor program, deterioration rates of in-use vehicles
at high and low altitudes were compared. No statistically
significant difference was found between the vehicle.
Therefore, the assignment of high-altitude DFs based on
low-altitude DFs is justified by in-use experience.
Under all scenarios manufacturers will also be allowed to
select one emission-data vehicle per engine family which is
expected to have the worst emissions when tested under
high-altitude conditions. This emission-data vehicle will be
-------
APII1-5
one of the emission-data vehicles previously selected for
testing at low altitude. Thus, this regulation will not cause
the manufacturers to incur the additional cost of building a
new emission-data vehicle and of accumulating 4,000 miles on
this vehicle.
The fraction of exempted vehicles for each scenario was
derived in Chapter II of this report. It will be assumed that
for each fraction of the total vehicles that are exempted, the
same fraction of engine families is exempted from
certification. For example, in scenario lb, 25 percent of all
vehicles are estimated to be exempt, which translates into 25
percent of all engine families.
The net certification costs for each alternative scenario
should be those costs that are incremental to the certification
costs that would normally have occurred for high altitude in
1984 under the base scenario which is a continuation of the
existing interim standards. Therefore, manufacturers woula
still certify to the levels of the 1982/1983 interim standards,
thus already incurring certification costs. These should be
"credited" towards the total certification costs that would
otherwise be required for each scenario.
After calculating first for the base scenario itself, the
incremental cost will be calculated for scenario 2, which is a
fixed-point statutory strategy and is very similar to the base
scenario, which is a fixed-point proportional strategy. Then
the costs for the continuous statutory and continuous
proportional scenarios will be calculated (scenarios la, lb,
lc, 3a, and 3b), again incremental to the base scenario. The
summary of certification costs are shown in Table V-9.
A. Base Scenario
The base scenario is a fixed-point scenario, where
manufacturers can certify an emission-data vehicle at low
altitude and a modification of the vehicle at high altitude. A
high-altitude standard would not affect the certification
status of a "low-altitude" emission-data vehicle. Thus, only
the cost of a "high-altitude" emission-data vehicle is oue to
the promulgation of a high-altitude standard.
In 1984 it is estimated that if there were no carryover,
then 117 LDV engine families will be certified, the same number
as that in 1980. This breaks down to about 109 LDVG ana 8
LDDV. For the base scenario in 1984 and thereafter, it is
estimated that 10 percent of emission-data vehicles will obtain
carryover from the previous year. This figure was based on
recent certification data for emission-data vehicles. Thus, SO
percent of the total number of LDGV engine families, or about
98 families, will be certified under the base scenario in 1984,
if there were no vehicle exemptions. However, it is estimated
-------
AP111 - 6
that 5 percent of engine families will be exempted, so that 93
families will undergo certification.
The estimated cost per test is $1,900. [1] This figure
includes $1,000 for testing ana $900 for vehicle
transportation. Approximately 1.5 tests will be performed per
engine family. This estimated testing cost may be high for
manufacturers with their own high-altitude facilities. These
manufacturers have one less profit center to account for than
do manufacturers who contract for certification at commercial
facilities. To be conservative, however, this potential cost
savings will not be included in this analysis.
The certification costs for the base scenario is $265,000
for 1984 and each year after. These costs have been calculated
keeping in mind that 5 percent of engine families are
exempted. This certification cost should be subtracted from
the total certification costs of alternate scenarios.
B. Scenario 2
In this scenario, as with the base scenario, only the cost
of recalibrating a "high-altitude" emission-data vehicle is due
to the promulgation of a high-altitude standard. Thus, with an
estimated engine family exemption of 15 percent it is estimated
that 93 gasoline-fueled LDVs will have to undergo certification
in 1984 and 90 percent of this number in 1985 and thereafter.
The incremental cost to the base scenario is then $0 for 1984
and each year after. This cost is also shown in Table V-9 .
C. Continuous Statutory and Continuous Proportional Scenarios
(Scenario la, b, c and 3a, b)
Under these scenarios manufacturers must show that each
nonexemptea vehicle can meet certification requirements at both
low and high altitude. In some instances this may require the
recertification of a vehicle at low altitude due to significant
changes in the engine families that were already certified in a
previous year. In other cases, if the manufacturer can
demonstrate that new control hardware devices show no effect on
previous certification results, then the low-altitude
certification process need not be repeated.
In 1984, it is expected that all LDV vehicles will have to
undergo certification testing at high-altitude to meet a
continuous high-altitude standard with the exception of
exempted vehicles or for vehicles that would not require
modification to meet proportional as in scenario 3a. In this
particular scenario the testing requirements for unmodified
vehicles at 5,300 feet from the previous model year would be
sufficient to demonstrate compliance. Therefore, the
high-altitude certification costs for scenario 3a are similar
to the base scenario except for the absence of exemptions.
-------
AP111-7
However, not all of these vehicles will have to be recertified
at low altitude, because a high-altitude standard should not
always cause manufacturers to incorporate a major change in
engine design that would affect previous low-altitude emission
results. Many of the control hardware items previously
discussed in this chapter are more or less "add-on" devices
that do not affect engine operation (ana, hence, emission
results) at low altitude. These devices would automatically
perform their emissions compensating function as the vehicle is
driven at higher altitude. For example, a continuous aneroid
compensates for the decrease in air density encountered at high
altitudes by allowing the "bleeding" of more air. At low
altitude, the continuous aneroid would not affect the air
intake, thus not affecting engine operation and emission
results.
However, some control hardware components may indeed cause
engines to operate differently at low altitude as well as high
altitude and thus, require certification at low altitude. Such
control hardware would be the ELCS, a turbocharger, and the
change to a feedback control system. For example, a
turbocharger increases a vehicle's power for each cylinder
stroke cycle regardless of altitude. Thus, upon implementation
of the ELCS, turbocharger, or feedback control system,
low-altitude certification is required. It is assumed that for
scenarios la, lb, lc, and 3b, the percentage of vehicle
families equipped with ELCS, turbochargers, or feedback
control, is equal to the percentage of total vehicles sold with
these components. For example, 33 percent of vehicles in
scenario lc require either ELCS, turbocharging, or feedback
control, and this corresponds to 33 percent of engine
families. This relationship was based on the belief that these
engine families will represent the lower power to weight
vehicles with higher fuel economy, and that vehicles sold from
these engine families will represent an approximately equal
percentage of the total vehicle market in 1964.
The number of LDV emission-data vehicles certified after
1984, or 1985-88 in this analysis, will be 90 percent of the
number of vehicles certified in 1S84. As previously stated
this figure was based on recent certification data for
emission-data vehicles. Low-altitude certification costs after
1984 should not be attributed to this 1984 high-altitude
standard, since it would be normal practice for manufacturers
to certify at low altitude regardless of a 1984 standard.
As discussed above, it is estimated that 109
gasoline-fueled engine families will be certified in 1984,
unless families are exempted. Each durability vehicle for the
low-altitude portion of certification costs about $197,000,[2]
and each emission-data vehicle for normal low-altitude
certification costs $27,000.[2] Each engine family has one
durability vehicle and about 3 emission-data vehicles at low
-------
APIII-8
altitude. Again, the certification test cost is about $1/000
and approximately 1.5 tests are performed for each engine
family.
The incremental certification costs are shown in Table
V-S. These costs were calculated in the same manner as for
scenario 2. Thus, when the certification cost of $265,000 for
LDVs is "credited", the incremental costs are $0-$18,773,000
for the first year, and $0-14,000 each year after. As
discussed above, this large range of costs is due to the fact
that some scenarios do not require all vehicles to be certified
at low altitude, and thus these vehicles will not incur the
expense associated with durability and emission-data vehicles.
These incremental certification costs will be carried out
through the remainder of this report.
-------
APIII-9
References
1. "Final Regulatory Analysis - Environmental and
Economic Impact statement for the 1982 and 1983 Model Year
High-Altitude Motor Vehicle Emission Standards," U.S. EPA,
OANR, OMS, ECTD, SDSB, October, 1980.
2. Light-Duty Vehicle Certification Cost, EPA
Memorandum to Edmund J. Brune, from Daniel P. Hardin, Jr.,
March 13, 1975.
------- |