EPA-450/3-84-012a
Evaluation of Air Pollution
Regulatory Strategies for
Gasoline Marketing Industry
OFFICE OF AIR QUALITY PLANNING AND STANDARDS
AND
OFFICE OF MOBILE SOURCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Washington, DC 20460
July 1984
-------
This report has been reviewed by the Office of Air Quality Planning and Standards and the Office of Mobile
Sources, EPA, and approved for publication. Mention of trade names or commercial products is not in-
tended to constitute endorsement or recommendation for use. Copies of this report are available through
the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle Park,
N.C. 27711, or from the National Technical Information Services, 5285 Port Royal Road Sprinafield
Virginia 22161.
-------
TABLE OF CONTENTS
Title
1.0 EXECUTIVE SUMMARY .
1.1 Operations, Emissions and Control Technology. . .
1.1.1 Bulk Terminals
1.1.2 Bulk Plants. . . .
1.1.3 Tank Trucks. . .
1.1.4 Service Stations .......
1.2 Analyses of Regulatory Strategies
1.2.1 Regulatory Strategies, Model Plants, and
Projections. ........
1.2.2 Air Pollution Emissions, Health-Risk, and
Control Cost Analyses
1.3 Results of Regulatory Strategy Analyses
i . •
1.3.1 Nonattainment Area Strategy Results . . .
1.3.2 Nationwide Strategy Results
1.3.3 Cost Per Incidence Reduction .......
2.0 GENERAL DESCRIPTION AND PROFILE. .............
2.1 General Industry Description
2.2 Gasoline Marketing Operations and Their Emissions
2.2.1 Gasoline Composition
2.2.2 Bulk Terminals
2.2.3 Storage Tanks at Terminals .
2.2.4 Bulk Plants. ...............
2.2.5 Tank Trucks .: .
2.2.6 Service Stations .
2.3 Baseline Emissions. .
2.4 References
3.0 CONTROL TECHNOLOGY
3.1 Introduction
3.2 Control Technology for Bulk Gasoline Terminals. .
3.3 Control Technology for Storage Tanks. . . .... .
3.3.1 Fixed-Roof Tanks
3.3.2 Internal Floating-Roof Tanks
3.3.3 External Floating-Roof Tanks
Page
1-1
1-2
1-2
1-4
1-5
1-5
1-7
1-7
1-12
1-17
1-20
1-22
1-29
2-1
2-1
2-3
2-3
2-8
2-9
2-13
2-13
2-15
2-16
2-20
3-1
3-1
3-1
3-4
3-4
3-5
3-5
m
-------
TABLE OF CONTENTS (Continued)
Title
3.4 Control Technology for Bulk Gasoline Plants . .
3.4.1 Submerged Fill
3.4.2 Vapor Balance System
3.4.3 Efficiency of Control Technologies . . .
3.5 Control Technology for Tank Trucks
3.5.1 Description of Control Technologies. . .
3.5.2 Effectiveness of Technologies
3.6 Control Technology for Transfers into Service
Station Underground Storage Tanks (Stage I) . .
3.6.1 Description of Technology
3.6.2 Effectiveness of Technology
3.7 Vehicle Refueling
3.7.1 Stage II Vapor Control Systems
3.7.2 Onboard Vapor Control Systems
3.7.3 Effectiveness of Technologies
3.7.4 In-Use Effectiveness of Control
Technologies
3.8 References
4.0 MODEL PLANTS AND REGULATORY STRATEGIES
4.1 Model Plants
4.1.1 Bulk Terminal Model Plants
4.1.2 Storage Tank Model Plant
4.1.3 Bulk Plant Model Plants
4.1.4 For-Hire Tank Truck Population ,
4.1.5 Service Station Model Plants ,
4.2 Gasoline, Facility, and Vehicle Projections . . ,
4.2.1 Gasoline Consumption .... ,
4.2.2 Gasoline Marketing Facilities ,
4.2.3 Light Duty Vehicles and Light Duty Trucks,
4.3 Regulatory Strategies
Page
3-6
•3-6
3-8
3-8
3-10
3-10
3-11
3-11
3-11
3-13
3-15
3-15
3-19
3-22
3-23
3-27
4-1
4-1
4-1
4-3
4-3
4-4
4-7
4-13
4-13
4-17
4-20
4-23
-------
TABLE OF CONTENTS (Continued)
Title
4.4 Source Category Control Options
4.4.1 Bulk Terminals
4.4.2 Bulk Plants . . .
4.4.3 Service Stations . . . .
4.5 References
5.0 ENVIRONMENTAL AND ENERGY IMPACTS . . .
5.1 Air Pollution Emission Impacts
5.1.1 Phase-In-Schedules for Control Options
5.1.2 Discounting of Emission Reductions . .
5.1.3 Emission Reduction Methodology . . . .
5.2 Other Environmental Impacts
5.2.1 Water Pollution Impacts. .......
5.2.2 Solid Waste Impacts. .
5.2.3 Other Environmental Impacts
5.3 Energy Impacts. .........
5.4 References
6.0 EXPOSURE/HEALTH-RISK ANALYSIS
6.1 Unit Risk Factors
6.1.1 Credibility of Risk Estimates
6.2 Exposure and Risk Methodology and Assumptions
6.2.1 General Assumptions
6.2.2 Incidence Analysis
6.2.3 Lifetime Risk Analysis
6.3 Presentation of Risk Estimates for Regulatory
Strategy . . . .
6.4 References
Page
4-29
4-29
4-32
4-33
4-41
5-1
5-1
5-1
5-6
5-7
5-19
5-19
5-24
5-24
5-24
5-28
6-1
6-1
6-3
6-4
6-5
6-7
6-18
6-27
6-40
-------
TABLE OF CONTENTS (Continued)
Title
7.0 COST IMPACTS
7.1 Introduction
7.2 Individual Facility Costs ...
7.2.1 Bulk Terminals . . . . .
7.2.2 Storage Tanks
7.2.3 Bulk Plants
7.2.4 For-Hire Tank Trucks
7.2.5 Service Stations
7.3 Nationwide Costs of Control Options
7.4 Nationwide Costs of Regulatory Strategies
7.5 Cost Per Incidence Reduction. .
7.6 References
8.0 ECONOMIC IMPACT OF THE REGULATORY STRATEGIES .
8.1 Scope and Method
8.2 Summary of Cost Comparisons •
8.2.1 Total Cost by Regulatory Strategy
8.2.2 Total Cost by Sector
8.2.3 Sources of Variation in Cost
8.2.4 Unit Cost and Quantity Impacts
8.3 Distributive Impacts «
8.3.1 Petroleum Refineries
8.3.2 Bulk Terminals and Bulk Plants
8.3.3 Service Stations
8.4 References
9.0 ENFORCEMENT STRATEGIES AND COST CONSIDERATIONS ..'...
9.1 Enforcement Strategies
9.1.1 Stage II Programs
9.1.2 Onboard Control Programs
9.2 Resources Required to Pursue Enforcement Strategies
at Various Levels of Effort
9.2.1 Installation Monitoring Resources
Page
7-1
7-1
7-2
7-2
7-8
7-9
7-15
7-20
7-34
7-37
7-44
7-50
8-1
8-1
8-3
8-3
8-5
8-10
8-18
8-26
8-29
8-31
8-34
8-37
9-1
9-1
9-1
9-5
9-5
9-5
vi
-------
TABLE OF CONTENTS (Concluded)
Tttl e
9.2.2 In-Use Inspections Resources . . .
9.2.3 Test Observation Resources
9.2.4 Legal-Clerical Resources . . .
9.2.5 Onboard Control Inspection Resources . . . .
9.3 Enforcement Costs
9.4 Enforcement Cost-Effectiveness Analysis
9.5 References
APPENDIX A HISTORY/BACKGROUND. . .
APPENDIX B BASELINE EMISSIONS ANALYSIS
APPENDIX C, ASSESSMENT OF ONBOARD CONTROLS, ...
APPENDIX D DETERMINATION OF IN-USE EMISSION REDUCTION
BENEFITS OF STAGE II PROGRAMS
APPENDIX E CUMULATIVE VALUES OF EMISSION REDUCTIONS
(1986-2020)
APPENDIX F EXPOSURE AND HEALTH-RISK ANALYSIS ........
APPENDIX G CUMULATIVE VALUES OF CAPITAL AND ANNUALIZED COSTS
(1986-2020)
Page
9-5
9-15
9-15
9-17
9-17
9-17
9-28
A-l
B-l
C-l
D-l
E-l
F-l
G-l
vn
-------
LIST OF TABLES
Title
1-1 Gasoline Marketing Regulatory Strategies ........... 1-8
1-2 Major Analytical Considerations ............... 1-H
1-3 Unit Risk Factor Summary ................... i"14
l-4a Estimated Risks from Gasoline Marketing Source Categories
(Using Plausible Upper Limit Unit Risk Factor for
Gasoline Vapors) ....................... 1-18
l-4b Estimated Risks from Gasoline Marketing Source Categories
(Using Maximum Likelihood Estimate Unit Risk Factor for
Gasoline Vapors) ....................... 1-19
1-5 Vehicle Refueling Controls in Nonattainment Areas ...... 1-21
1-6 Control of Benzene from Gasoline Marketing Sources ...... 1-23
1-7 Summary of Theoretical Impacts for Selected Regulatory
Strategies ..... ............... !-25
1-8 Impacts Based on "In-Use" Effectiveness for Stage II and
Onboard Controls (1986-2020) ................. 1-26
1-9 Economic Impact of Regulatory Strategies Based on Theoretical
Efficiencies ......................... i'30
1-10 Economic Considerations ..... .... .......... 1-31
1-11 Estimated Regulatory Costs and VOC Benefits ......... 1-33
1-12 Benzene Regulatory Costs and Incidence Reduced ........ 1-34
1-13 Benzene Regulatory Costs Per Cancer Incidence Avoided
(Assuming VOC Benefits) ................... !-35
1-14 Benzene and Gasoline Vapors Costs Per Cancer Incidence
Avoided (Using Rat Data Unit Risk Factor for Gas Vapors). . . 1-36
2-1 Uncontrolled Emissions from Gasoline Tank Truck
Loading Operations at a Typical Bulk Gasoline Terminal. . . . 2-10
2-2 Emissions from Gasoline Storage Tanks Located at a
Typical Terminal ....................... 2~12
2-3 Uncontrolled Emissions from a Typical Bulk Plant ....... 2-14
2-4 Uncontrolled Emissions from a Typical Service Station .... 2-17
vm
-------
LIST OF TABLES (Continued)
Title
Page
2-5 Summary of Baseline Emissions for Gasoline Marketing
Facilities for Base Year 1982 2-19
3-1 In-Use Efficiencies of Stage II and Onboard Technologies. . . 3-25
4-1 Bulk Gasoline Terminal Model Plant Parameters 4-2
4-2 Bulk Plant Model Plant Parameters 4-5
4-3 Method of Calculating the Number of Uncontrolled
Terminal-Owned Trucks . . . *. . - 4-6
4-4 Estimates of 1982 Service Station Population . . 4-9
4-5 Estimated 1982 Service Station Size Distribution. ...... 4-11
4-6 Alternate Gasoline Consumption Projections
(billion liters [gallons]) . . 4-is
4-7 Estimated Number of Facilities in the Base Year 1982. . . ". . 4-19
4-8 Onboard Consumption Projections 4-21
4-9 Gasoline Marketing Regulatory Strategies Standard Numbers and
Titles. 4-24
4-10 Composition of Regulatory Strategies by Source Category . . . 4-25
4-11 Gasoline Marketing Facility Model Plants . . . . . 4-30
4-12 Number of Facilities Affected by Gasoline Marketing Control
Options 4-31
4-13 Ozone Nonattainment Areas (NA) Assumed Affected by a CTG for
Gasoline Marketing Evaluation ... ...... 4-35
4-14 Relative Size of Nonattainment Areas Affected by Additional
Stage II Controls 4-37
4-15 Nonattainment Areas Committed to or Scheduling Stage II
Vehicle Refueling Controls (NA) ... 4-38
5-1 Effective Dates and Phase-In Schedules for Gasoline
Marketing Regulatory Strategies 5-5
5-2 Summary of Control Requirements and Affected Emission Factors
for Gasoline Marketing Control Options 5-9
5-3 Gasoline Vapor Emission Reductions in 1982 Associated with
Control Options . 5-11
-------
LIST OF TABLES (Continued)
Title
5-4 Summary of Projected Total Onboard Emission Impacts in
Each Year of the Study (1988-2020) .............. 5-13
5-5 Projected Total Gasoline Consumption Changes
from the Base Year ...................... 5-15
5-6 Projected Leaded Gasoline Consumption ............ 5-16
5-7 Nationwide Emission Reductions from Gasoline Marketing
Control Options ....................... 5-18
5-8 Emission Reductions from Stage II Options When Combined with
Onboard ........................... 5-20
5-9 Nationwide Emission Reductions from Gasoline' Marketing
Regulatory Strategies (1986-2020) .............. 5-21
5-10 Emission Reduction Comparison Between Stage II and Onboard
Considering In-Use Efficiencies . . ............. 5-23
5-11 Gasoline Recovery Ratios ...... ............. 5-25
5-12 Energy Savings Associated with Gasoline Marketing Regulatory
Strategies ............ • ............. 5-27
6-1 Unit Risk Factor Summary ................... 6-2
6-2 Emission Sources Considered in Risk Analysis ......... 6-6
6-3 Bulk Terminal and Bulk Plant Annual Incidence Analysis. ... 6-9
6-4 Service Station Annual Incidence Analysis .......... 6-12
6-5 Self-Service Incidence Analysis ............... 6-15
6-6 Bulk Terminal and Bulk Plant Lifetime Risk Analysis ..... 6-20
6-7 Service Station Lifetime Risk Analysis ............ 6-25
6-8 Self-Service Lifetime Risk Analysis ..... : ....... 6-28
6-9 Estimated Cumulative Incidence from Benzene and Gasoline
Vapors (Plausible Upper Limit Unit Risk Factor) from Gasoline
Marketing Source Categories under Each Regulatory Strategy. . 6-29
6-10 Estimated Cumulative Incidence from Benzene and Gasoline Vapors
(Maximum Likelihood Unit Risk Factor) from Gasoline Marketing
Source Categories under Each Regulatory Strategy ....... 6-30
6-11 Estimated Cumulative Incidence from EDB and EDC from Gasoline
Marketing Source Categories under Each Regulatory Strategy. . 6-31
-------
LIST OF TABLES (Continued)
Title
Page
6-12 Estimated Cumulative Incidence from Benzene and Gasoline
Vapors from Vehicle Operations under Selected Regulatory
Strategies 6-32
6-13 Estimated Lifetime Risk from Benzene and Gasoline Vapors
(Using Plausible Upper Limit Unit Risk Factor) from
Gasoline Marketing Source Categories under Each Regulatory
Strategy 5.34
6-14 Estimated Lifetime Risk from Benzene and Gasoline Vapors
(Using Maximum Likelihood Estimate Unit Risk Factor) from
Gasoline Marketing Source Categories under Each Regulatory
Strategy 6-35
6-15 Estimated Lifetime Risk from EDB and EDC from Gasoline
Marketing Source Categories under Each Regulatory Strategy. . 6-36
6-16 Estimated In-Use Cumulative Incidence from Benzene and Gasoline
Vapors (Plausible Upper Limit Unit Risk Factor) from Gasoline
Marketing Source Categories under Selected Regulatory Strategies
and Enforcement Levels . 6-37
6-17 Estimated In-Use Cumulative Incidence from Benzene and Gasoline
Vapors (Maximum Likelihood Estimate Unit Risk Factor) from
Gasoline Marketing Source Categories under Selected Regulatory
Strategies and Enforcement Levels ...'... 6-38
7-1 Bulk Terminal Bottom Load Control Costs (Thousands of
4th Quarter 1982 Dollars) . 7-5
7-2 Bulk Terminal Top Load Control Costs (Thousands of
4th Quarter 1982 Dollars) 7-4
Footnotes for Tables 7-1 and 7-2 7-6
7-3 Average Bulk Terminal Control Costs
(Thousands of 4th Quarter 1982 Dollars) 7-7
7-4 Bulk Terminal Average Weighted Costs (Thousands of 4th Quarter
1982 Dollars) 7-3
7-5 Cost of Installing a Bolted Internal Floating Roof on an
Existing Fixed-Roof Tank (4th Quarter 1982 Dollars) ... . . . 7-10
-------
LIST OF TABLES (Continued)
Title Page
7-6 Estimated Control Costs for Bulk Plants (No Exemptions)
(4th Quarter 1982 Dollars) 7-12
7-7 Estimated Control Costs for Bulk Plants (Exempt < 4,000 gal/day)
(4th Quarter 1982 Dollars) 7-14
7-8 Cost for the For-Hire Tank Trucks at Terminals
(4th Quarter 1982 Dollars) 7-17
7-9 Cost for the For-Hire Tank Trucks at Bulk Plants
(4th Quarter 1982 Dollars) 7-19
7-10 Service Station Stage I Capital and Net Annualized
Cost Estimates (4th Quarter 1982 Dollars) 7-21
7-11 Stage I Control Costs (Millions of 4th Quarter
1982 Dollars) ". . 7-22
7-12 Average Stage II Costs per System (4th Quarter
1982 Dollars) 7-23
7-13 Stage II Recovery Credit Calculations 7-25
7-14 Weighted Average Stage II Costs (4th Quarter
1982 Dollars) 7-26
7-15 California Air Resources Board (ARB) Stage II
Control Costs 7-28
7-16 Service Station Stage II Weighted Average Costs
(4th Quarter 1982 Dollars) 7-29
7-17 Onboard Vapor Control Hardware Costs
(1983 Dollars) 7-32
7-18 Cost Comparison between Stage II and Onboard Controls
Considering In-Use Efficiencies (4th Quarter 1982 Dollars)
(1986-2020) 7-35
7-19 Number of Facilities Requiring Controls for Gasoline
Marketing Options .... 7-36
7-20 Nationwide Costs of Gasoline Marketing Control Options. . . . 7-38
7-21 Cost Effectiveness of Gasoline Marketing Control Options. . . 7-39
7-22 Costs for Stage II Options when Combined with Onboard .... 7-41
7-23 Nationwide Costs of Gasoline Marketing Regulatory Strategies. 7-42
xn
-------
LIST OF TABLES (Continued)
Ti tl e
Page
7-24 Cost Effectiveness of Regulatory Strategies for
Gasoline Marketing. . . 7-43
7-25 Estimated Regulatory Costs and VOC Benefits 7-45
7-26 Benzene Regulatory Costs and Incidence Reduced 7-46
7-27 Benzene Regulatory Costs Per Cancer Incidence Avoided
(Assuming VOC Benefits) 7-47
7-28 Benzene and Gasoline Vapor Costs Per Cancer Incidence Avoided
(Using Rat Data Unit Risk Factor for Gas Vapors) 7-48
8-1 1986 NPV of the Costs of the Regulatory Strategies
(109 1982 dollars). . 8-4
8-2 Estimated Reduction in Benzene Content of Gasoline Resulting
From Regulatory Strategy XIV . ...".. 8-6
8-3 1986 NPV of the Costs of Benzene Reduction (109 1982 Dollars). 8-6
8-4 1986 NPV of Stage I Control and Enforcement Costs for Bulk
Terminals, Bulk Plants, and For-Hire Trucks
(106 1982 dollars) . 8-7
8-5 NPV of Stage I and II Control and Enforcement Costs for Service
Stations (10^ 1982 dollars) 8-9
8-6 1986 NPV of the Total Costs of the Regulatory Strategies Under
Alternative Assumptions (lO9 1982 dollars in 1986) 8-11
8-7 Average Unit Cost and Quantity Effects for Nationwide
Regulatory Strategies under Base Case Assumptions. ...... 8-19
8-8 Average Unit Cost Increases for Gasoline under Alternative
Cost Assumptions (1982 /liter) 8-20
8-9 Reductions in Gasoline Consumption Attributed to Average Unit
Cost Increases under Alternative Cost Assumptions
(106 liters/year) 8-23
8-10 Gasoline Quantity Impacts: Average National Reductions in
Consumption under Constant Gasoline Consumption With Various
Elasticity Assumptions (10^ liters/year) 8-24
xm
-------
LIST OF TABLES (Concluded)
Title Page
8-11 Reductions in Vehicle Consumption Attributable to Unit Cost
Increases under Alternative Cost Assumptions 8-25
8-12 Reductions in Vehicle Consumption Attributable to a Unit Cost
Increase of $13/Vehicle Tank With Various Price Elasticities . 8-25
8-13 Difference in Average Unit Cost for Small and Large Facilities
under Declining Gasoline Consumption 8-28
8-14 Increased Cost to Petroleum Refineries Due to Benzene Reduction
in Gasoline for a 10,000-Barrel/Stream-Day Refinery vs. the
U.S. Average 8-30
8-15 Increased Unit Cost to Petroleum Refineries Due to Benzene
Reduction of Gasoline by PADD District ( ?/liter, t/galIon
in parentheses) 8-30
8-16 1986 NPV of Control Cost per Station by Model Plant
(1982 Dollars) 8-35
9-1 Inspection and Re-Inspection Time Assumed for
Enforcement Cost Analysis 9-7
9-2 Probability That Individual Control Unit will have at Least
One Defect, as a Function of Enforcement Effort 9-13
9-3 Percentages of Facilities (Steady-State Average) Which
Would be in Violation as a Function of Frequency of
In-Use Inspections 9-14
9-4 Cumulative Enforcement Costs of Control Options (1986-2020). . 9-19
9-5 Cumulative Enforcement Costs of Regulatory Strategies
(1986-2020) 9-22
9-6 Theoretical and In-use Effectiveness Comparison among Stage II
Control Options 9-27
xiv
-------
LIST OF FIGURES
Title
Page
1-1 Gasoline Marketing in the U.S. (1982 Baseline
VOC Emissions) 1_3
1-2 Effect of Onboard and Stage II Controls on Benzene Incidence
(Based on Theoretical Efficiencies) 1-28
2-1 Gasoline Distribution in the U.S 2-2
2-2 Emission Equations 2-5
2-3 Mass Emission Rates ^Saturation 2-6
2-4 Mass Emission Ratio of BZ, EDC, & EDB to VOC in Vapor. ... 2-7
3-1 Example of Gasoline Loading at Bulk Terminals 3-2
3-2 Gasoline Tank Truck Loading Methods 3-7
3-3 Vapor Balance Systems at Bulk Gasoline Plants 3-9
3-4 Tank Truck Vapor Collection Equipment for
Bottom Loading Operations . 3-12
3-5 Vapor Balance System at a Service Station 3-14
3-6 Stage II Vapor Recovery Balance System 3-16
3-7 Stage II Vapor Recovery Vacuum Assist System 3-18
3-8 Stage II Vapor Recovery Hybrid System 3-20
3-9 Onboard Controls for Vehicle Refueling Emissions 3-21
4-1 Estimated Flow of Gasoline Through the U.S. Gasoline
Distribution System 4_14
4-1 Footnotes for Figure 4-1 4_15
4-2 Gasoline Consumption Projections 4-16
4-3 Total Gasoline Consumption vs. Gasoline Consumed by Onboard
Controlled Vehicles (1988-2020) 4-22
5-1 Linear Phase-In-Schedules 5-3
6-1 Map of Bulk Terminal Complex 6-21
6-2 Map of Bulk Plant Complex 6-23
6-3 Map of Service Station Complex . 6-26
9-1 Probability of an Individual Defective Unit 9-11
xv
-------
-------
1.0 EXECUTIVE SUMMARY
The purpose of the study was to evaluate the air pollution
regulatory strategies available to reduce emissions from gasoline
marketing operations of benzene (Bz), ethylene dibromide (EDB), ethylene
dichloride (EDO, and gasoline vapors (GV). Gasoline vapors or
volatile organic compound (VOC) emissions contribute to ambient ozone
concentrations and, thus, in some areas contribute to a failure to
attain the national ambient air quality standard for ozone. Benzene is a
known carcinogen, which has been listed as a hazardous air pollutant
under Section 112 of the Clean Air Act and. is present in varying amounts
in gasoTine. In addition, EDB, EDC and gasoline vapors each have been
shown to cause cancers in laboratory animals. EDB and EDC are-generally
added to leaded gasoline, but are not present in unleaded gasoline.
The following segments of the gasoline marketing industry were considered:
bulk terminals (including storage tanks and tank trucks), bulk plants
(including storage tanks and tank trucks) and service stations (both
inloading of underground storage tanks and refueling of vehicles). The
regulatory strategies examined controls on all segments of the industry,
both with and without selected size cutoffs for small facilities, as
well as controls onboard the vehicle to reduce refueling emissions.
As noted, there are still areas of the country which have"not yet
attained the national ambient air quality standard (NAAQS) for ozone.
The Clean Air Act requires that all areas achieve the NAAQS by
December 31, 1987. Some States, as part of their State implementation
plans to meet the statutory requirement, are considering control of
gasoline marketing sources, especially the refueling of motor vehicles.
Thus, an analysis of gasoline marketing regulatory strategies must
address the need to attain the ozone NAAQS in selected areas. However,
the emissions from gasoline marketing sources may induce public health
risks which require control on a national basis. The analysis evaluated
regulatory strategies which address both the more limited nonattainment
issue in part of the country and the broader question of the need for a
national control program to limit potential hazardous exposure.
1-1
-------
1.1 OPERATIONS, EMISSIONS AND CONTROL TECHNOLOGY
This section briefly outlines the operations and emissions of each
source category and major associated type of facility in the gasoline
marketing industry, as well as the commonly used control techniques.
The segments of the gasoline marketing industry analyzed in this study
include all elements and facilities that move gasoline starting from
the bulk terminal to its end consumption. Gasoline produced by refiner-
ies is distributed by a complex system comprised of wholesale and
retail outlets. Figure 1-1 depicts the main elements in the marketing
network. The flow of gasoline through the marketing system is shown
from the refinery, through bulk terminals, and sometimes bulk plants,
to retail service stations or commercial or rural dispensing facilities,
primarily via pipeline and tank truck. The wholesale operations storing
and transporting gasoline including delivery and storage in a service
station underground tank are commonly called Stage I operations-.
Retail-level vehicle refueling operations are commonly termed Stage II.
The baseline nationwide VOC emission estimates are also given for
the various source categories on Figure 1-1. VOC emission factors for
the individual point source operations at each source category were
estimated. Emissions at baseline were calculated based on the gasoline
throughput and current regulations for each source category in each
county in the nation. Emission estimates for the other pollutants (Bz,
EDB, EDO were calculated using a ratio of the vapor pressures and
thus, vapor emission rates.
1.1.1 Bulk Terminals
Bulk gasoline terminals serve as the major distribution point for
the gasoline produced at refineries. Gasoline is most commonly delivered
to terminal storage tanks by pipeline with no emissions. Gasoline is
stored in large aboveground tanks and later pumped through metered
loading areas, called loading racks, and into delivery tank trucks,
which service various wholesale and retail accounts in the marketing
network.
Most tanks in gasoline service at terminals have an external or,
less commonly, an internal floating roof to prevent the loss of product
through evaporation and working losses. Floating roofs rise and fall
with the liquid level preventing formation of a large vapor space
1-2
-------
Imported
Gasoline
Barge
Pipeline
Tanker
'200,000 MG/YR
222,000 !%/YR
Servlca
Station
407,000 MG/YR
Commercial,
Rural
Consumer
Imcortad
or
Domestic
Crude
Wholesale
Distribution
Level
208,000 MS/YR
J ~ Storage
= iransaort
Figure 1-1. Gasoline Marketing in the U.S.
(1982 Baseline VQC Emissions)
1-3
-------
and resulting emissions. Fixed-roof tanks, which are still used for
gasoline in some areas, use pressure-vacuum (P-V) vents to control the
smaller breathing losses and may use processing equipment to control
the much greater working (filling and emptying) losses. Breathing
losses result from volume variation due to daily changes in temperature
and barometric pressure. Emptying losses occur when air drawn into the
tank during liquid removal saturates with hydrocarbon vapor and expands
beyond the vapor space of a fixed-roof tank. Filling losses occur when
the vapors in the fixed-roof tank are displaced by the incoming liquid
and forced to the atmosphere. The largest potential source of losses
from external floating-roof tanks is an improper fit between the seal
and the tank shell. Withdrawal loss from exposed-wet tank walls is
another source of emissions from floating-roof tanks.
Emissions from the tank truck loading operations at terminals occur
when the product being loaded displaces the vapors in the delivery
truck tank and forces the vapors to the atmosphere. In order to con-
trol these loading emissions, the displaced vapors can be ducted to a
vapor processor such as a carbon adsorber, thermal oxidizer, or refrig-
erated condenser for recovery or destruction. The quantity of emissions
generated during loading a tank truck are dependent on the type of
loading. Splash loading from the top of the truck creates considerable
turbulence during loading and can create a vapor mist resulting in
higher emissions. Top submerged loading, which uses an extended fill
pipe, or bottom loading admit gasoline below the liquid level in the
tank and can be used to reduce turbulence and emissions (about a 60
percent reduction). The recently promulgated bulk terminal new source
performance standards (NSPS), as well as a large number of State regula-
tions, currently require the use of vapor processors and submerged
loading at bulk terminals. Most State regulations limit truck loading
emissions to 80 mg/liter transferred (equivalent to about 90 percent
reduction) and require the tank truck to be vapor tight. The NSPS is
more stringent and requires a lower emission limit of 35 mg/liter.
1.1.2 Bulk Plants
Bulk gasoline plants are secondary distribution facilities that
typically receive gasoline from bulk terminals via truck transports,
1-4
-------
store it in aboveground storage tanks, and subsequently dispense it
via smaller account trucks to local firms, businesses and service
stations. As discussed in the previous section, vapors can escape from
fixed-roof storage tanks at bulk plants due to breathing losses
even when there is no transfer activity. The majority of bulk plants
already use top or bottom submerged loading, largely in response to
State regulations. Vapor balancing is required by many State regulations,
primarily for incoming loads, but also for outgoing loads in some
instances. Vapor balancing enables the vapors from the tank being
filled to be transferred via piping to the tank being emptied. Thus, the
vapors are not forced to the atmosphere, and as a result, working losses
are greatly reduced (by 90 percent or greater).
1.1.3 Tank Trucks
Gasoline tank trucks are normally divided into compartments with a
hatchway at the top of each compartment. Loading can be accomplished by
top splash or submerged fill through the hatch, or by bottom filling.
The majority of trucks have dual capability. Either top or bottom
loading can be adapted for vapor collection. However, the trend is
toward bottom loading because of State vapor recovery regulations and
operating and safety advantages. The vapor collection equipment is
basically composed of vapor domes enclosing each top hatch along with
various connectors and pipes (some removable) that enable the vapors
from the tank being filled to be transferred to the tank being emptied
of liquid. Tank trucks with vapor collection equipment can become a
separate source of emissions when leakage occurs (estimated to average
about 30 percent of potentially captured emissions). Many States
require gasoline tank trucks equipped for vapor collection to pass an
annual test of tank vapor tightness and pressure limits for the tanks
and vapor collection equipment (reducing average leakage to about
10 percent).
1.1.4 Service Stations
Gasoline handling operations, emissions, and controls at service
stations are basically divided into two steps: the filling (or inloading)
of the underground storage tank, commonly called Stage I, and vehicle
refueling, commonly called Stage II. The filling of underground tanks
1-5
-------
at service stations ends the wholesale gasoline marketing chain. The
automobile refueling operation at service stations is the part of the
marketing chain that interacts directly with the general public.
Emissions from underground tank filling operations at service
stations can be reduced significantly (by about 95 percent) by the use
of a vapor balance system. Instead of being vented to the atmosphere,
the vapors are transferred into the tank truck unloading at the service
station and, ultimately, to the terminal vapor processor for recovery
or destruction. Such controls have been incorporated into many State
regulations.
Vehicle refueling emissions are another major source of emissions,
attributable to spillage and to vapor displaced from the automobile
tank by dispensed gasoline. The two basic vehicle refueling regulatory
strategies are: (1) control systems on service station equipment
(termed Stage II controls), and (2) control systems on vehicles and
trucks (termed onboard controls). Stage II controls consist of either
vapor balance systems or assisted systems. Assisted systems use a
variety of means to generate a more favorable (negative or zero) pressure
differential at the nozzle-vehicle interface so that a tight seal is
not necessary between the vehicle and the nozzle "boot" (a flexible
covering over the nozzle which captures the vapor for return to the
underground tank via a vapor hose). Stage II controls are currently
being used in 26 counties in California and the District of Columbia and
are being considered for other ozone nonattainment areas. Onboard vapor
controls consist of a fillpipe seal and a carbon canister that adsorbs
the vapors displaced from the vehicle fuel tank by the incoming gasoline.
The onboard system has undergone only limited testing to date. It is
unclear what design problems could be encountered if onboard were
required for the entire vehicle fleet; however, the technology is an
extension of a system already installed on light-duty cars and trucks.
Since 1971, new cars have been equipped with similar carbon canister
systems for collecting evaporative emissions (breathing losses caused
by temperature changes in the vehicle tank and carburetor).
Both Stage II and onboard controls can be highly effective (as
high as 95 and 98 percent, respectively). However, these high theoretical
1-6
-------
efficiencies are likely to be somewhat reduced in-use (to as low as
56 percent for Stage II programs under a minimal enforcement scenario
considering a 20 percent rate of noncompliance, and to about 92 percent
for onboard controls with the expected level of tampering).
It should be noted that vehicle refueling controls not only reduce
ambient concentrations of VOC and hazardous emissions dispersed from
the service station, but also reduce the much higher exposures to
hazardous pollutants during self-service refueling (as discussed later).
In addition, it has been found that the present canisters for controlling
evaporative emissions on many models of new vehicles are undersized.
The expansion of the onboard system to control refueling emissions
could achieve additional evaporative emission reductions roughly equal
to the emission reductions achieved through refueling control. The
estimate of excess evaporative emissions is preliminary because EPA
testing is not yet complete.
1.2 ANALYSES OF REGULATORY STRATEGIES
The regulatory strategies selected for this evaluation were assessed
with regard to their air pollution emissions, health-risk and cost
impacts. Impacts analyses were conducted using a model plant approach
for most industry segments and source categories, along with certain
key assumptions. Economic impacts were also assessed, as were the
effects of various enforcement levels on in-use effectiveness of the
vehicle refueling control systems. The following sections summarize
the regulatory strategies and analytical methods and assumptions used.
1.2.1 Regulatory Strategies, Model Plants, and Projections
A total of 14 industry-wide regulatory strategies were selected for
evaluation. These strategies, which are presented in
Table 1-1, are composed of a mixture of control options for the individual
source categories. For the strategies calling for additional controls
assessment was made of the relative emissions, risks, costs, and cost
effectiveness of:
(1) nationwide control of Stage I sources,
1-7
-------
TABLE 1-1. GASOLINE MARKETING REGULATORY STRATEGIES3
No Additional Controls (Baseline)
Stage II - Selected Nonattainment Areas (NA*)b
Stage II - All Nonattainment Areas (NA)C
Stage I - Nationwide
Stage II - Nationwide
Stage I and Stage II - Nationwide
Onboard - Nationwide
Stage II - Selected Nonattainment Areas & Onboard - Nationwide
Stage II - All Nonattainment Areas & Onboard - Nationwide
Stage I & Onboard - Nationwide
Stage II - All Nonattainment Areas and Stage I & Onboard - Nationwide
Stage II & Onboard - Nationwide
Stage I & Stage II & Onboard - Nationwide
Benzene Reduction in Gasoline^
Facility Size Cutoffs:
Stage I:
(1) bulk plants with throughputs <4000 gal/d from balance controls on
outgoing loads; and
(2) service stations with throughputs <10,000 gal/mon.
Stage II:
(1) all service stations with throughputs <10,000 gal/mon; and
(2) all independent service stations with throughputs <50,000 gal/mon,
Ozone nonattainment areas needing vehicle refueling controls to help meet
their ozone attainment goals by 1987.
Areas predicted by State or EPA to be nonattainment for ozone in 1982.
d
Benzene reduction:
A. removal of 94.5 percent of Bz from reformate fraction for total
reduction of 62.4 percent;
B. removal of 94.5 percent of Bz from reformate and fluid catalytic
cracker (FCC) fractions for total reduction of 81.3 percent.
1-8
-------
(2) nationwide control of vehicle refueling emissions
(Stage II controls, onboard controls, or both),
(3) ozone nonattainment area control of vehicle refueling
emissions (selected or.all ozone nonattainment areas), and
(4) combinations of the above.
The regulatory strategies consider these approaches either singly or in
combination, both for controlling all facilities and for including size
cutoffs for some facilities. The facility size cutoffs were assumed
based on the relatively higher costs of control for small facili-
ties, existing size cutoffs under State and local regulations, and
statutory requirements for small and medium throughput independent
service stations under Section 325 of the Clean Air Act (section titled
"Vapor Recovery for Small Business Marketers of Petroleum Products").
If a Section 112 standard is pursued requiring Stage I or Stage II
controls, the actual size cutoffs could vary from these assumptions
based upon a more thorough economic analysis for small businesses and
an assessment of whether Section 325 applies to Section 112 standards.
For the purposes of the analysis, initial installation of Stage I and
II control equipment was assumed to occur in 1986 for the nonattainment
area strategies, in 1987 for nationwide strategies, and on new vehicles
beginning with the 1988 model year for onboard controls. All of the
strategies were compared with a baseline reflecting 1982 Federal, State,
and local regulations. The base year of 1982 was selected because this
represented the final implementation year for many State regulations
affecting gasoline marketing sources, and because at the beginning of
the analysis the most recent complete data reflected 1982 totals.
A number of model plants were developed to represent the entire
spectrum of facilities in the analyses. Four model plants were used
for both bulk terminals and bulk plants while five model plants were
used for service stations. The model plants for each source category
were differentiated on the basis of size, in terms of gasoline throughput.
Estimates of typical costs, emissions, and resultant health-risks could
then be generated for each model plant and, thus, for the entire population
spectrum of each facility type.
1-9
-------
Because onboard controls would be installed only on new vehicles
(this analysis did not consider a retrofit option), the onboard regulatory
strategies take a number of years after initial implementation to
control the entire vehicle fleet. Therefore, in order to evaluate the
comparison of onboard with other controls during both phase-in and full
Implementation, the analyses examined the time period from 1986 (when
the first controls would begin to be implemented) through 2020. Thus,
the projection of certain basic parameters was necessitated.
Total and leaded gasoline consumption were extrapolated to the
year 2000, based on the projections by EPA through 1990 for the phasedown
of lead in gasoline (47 FR 49329). Due to a lack of confidence in
extrapolating beyond 15 years, gasoline consumption was assumed to
remain constant from the year 2000 to the year 2020. The number and
fuel consumption of onboard controlled vehicles in a given year, were
estimated based on projections of new vehicles, retirement rates, fuel
economy, and mileage accrual rates through the year 2000. These parameters
were also assumed constant from 2000 through 2020. Although the number
of bulk plants and service stations have been decreasing, no quantitative
data was readily available with which to project facility populations.
Therefore, the numbers and size distributions of facilities were assumed
to remain constant at the values estimated for the base year of 1982.
However, the throughputs per facility and corresponding recovery credits
were decreased in proportion to the projected decline in nationwide
gasoline consumption. The economic effects of alternative assumptions
(e.g., constant gasoline consumption, declining marketing facilities,
etc.) are examined in Chapter 8.
A summary of major analytical considerations that should be noted
in assessing the results is given in Table 1-2. Both costs and emissions
were summed to a cumulative value, and also were discounted (at 10 percent)
to a net present value in 1986 (by summing the equivalent worth in 1986
of each annual amount). Discounting was necessary because impacts are
not uniform over the time period analyzed due to: 1) the -slower phase-in
period of onboard controls compared to Stage I and II equipment; 2) the
respective useful lives of service station and onboard control equipment;
and 3) the declining gasoline consumption which directly influences the
1-10
-------
TABLE 1-2
MAJOR ANALYTICAL CONSIDERATIONS
A standard exposure lifetime of 70 years was used for exposure to ambient
concentrations (i.e., those away from the immediate vehicle fueling area)
A period of 50 years was used for self-service fueling exposure because
one would not be expected to operate a vehicle for a complete lifetime.
Because of the different phase-in times for Stage II and onboard and to
analyze the impact of the strategies after onboard was fully implemented
the analysis covered a period from 1986 to 2020.
In this analysis, it was assumed that a national Stage II program would be
in place in 1989, and that 1988 model year vehicles would be the first to
incorporate onboard controls. It is expected that it will take about 20
years to convert the entire vehicle fleet to onboard control.
Both costs and emissions capture over the 35-year analysis period were
discounted at 10 percent to calculate cost effectiveness. This was done
to address the difference in phasing-in Stage II and onboard controls.
— Model Plants
-- Typical Locations of Plants
— Number of Facilities Within Source Categories
-- Distribution of all Sources
o Sources were distributed by best estimate considering size and population
densities.
o Risk is assumed to be linearly related to dose (concentration x duration
of exposure) and combinations of concentration and duration yieldinq the
same dose are assumed to be equivalent for risk estimation purposes.
o The impacts of exposure to other substances beyond benzene, EDC, EDB and
gasoline vapor are not addressed. Health Effects other than cancer are
not explicitly addressed.
o Total gasoline and leaded gasoline consumption is based on EPA's lead
phase-down projection extrapolated to the year 2000. The consumption
estimate for the year 2000 is assumed for all years from 2001 to 2020.
o Fleet average cost estimate of onboard systems used in this analysis was
approximately $15 per vehicle. Recent studies by API and Ford Motor
Company estimate average onboard costs of $13 and $53 per vehicle
respectively. '
o The average capital costs (equipment and installation costs) per station
for Stage II control systems used in this analysis are $5,700 $6 100
$6,600 $9,800 and $14,800 for 5, 20, 35, 65, and 185 thousand gallon
per month throughput stations, respectively.
1-11
-------
recovery cost credits. Cost effectiveness was calculated using the
discounted costs and emissions values.
1.2.2 Air Pollution Emissions, Health-Risk, and Control Cost Analyses
Several underlying methods and assumptions were made in all of the
emissions, health-risk, and control cost analyses, in addition to the
projections noted in the previous section. Generally, emissions and
health-risk impacts of the various regulatory strategies (including
baseline, which reflects current controls) were estimated for a base
year (1982) and then extrapolated to the years 1986 through 2020 in
proportion to the total (for benzene and gasoline vapors) or leaded
(for EDB and EDO gasoline throughput for each source category. In
addition, the phasing-in of control installations with time were
considered in accordance with statutory requirements. All affected
facilities were assumed to install controls (linearly with time) within
one year for a CT6 and within 2 years for a NESHAP except for independent
service stations, which may be allowed up to 3 years in accordance with
Section 325 of the Clean Air Act. Capital costs were attributed to the
year of installation.
Estimates of annualized costs, emissions, and subsequent health-risks
during phase-in periods were based on the number of facilities and
corresponding gasoline throughput controlled for an entire year. The
impacts of vehicles with onboard controls in each year were calculated
based on the vehicle fleet projections noted previously. After nearly
the entire vehicle fleet was projected to be equipped with onboard
controls (in about 2002-2003), Stage II controls were not replaced, but
instead gradually phased out after the completion of useful equipment
lives for those strategies combining Stage II and onboard.
The health-risk analysis estimated both annual cancer incidences
nationwide and lifetime risk from high exposure, assuming a linear
dose response relationship with no threshold. The term "lifetime risk
from high exposure" is conceptually similar to the term "maximum lifetime
risk" which has been presented in other EPA documents, including those
on benzene sources regulated or considered for regulation under Section
112 of the Clean Air Act. The term "lifetime risk from high exposure"
rather than "maximum lifetime risk" is used in presenting the risk
1-12
-------
calculations for the gasoline marketing study because EPA is less
certain in this case that the assumptions used result in the maximum
exposure to any single person or group. For example, high exposure
from self-service was assumed to occur to a person pumping 40 gallons
per week. To the extent that some people may pump more gas than that,
risks may be underestimated.
The estimates of risk, in terms of individual lifetime risk from
high exposure and aggregate incidence, are applicable to the public in
the vicinity of gasoline marketing sources and those persons who refuel
their vehicles at self-service pumps. This analysis did not examine
the risk to workers from occupational exposure (e.g., terminal operators
and service station attendants). The lifetime risk from high exposure
for these workers is probably substantially higher than for the general
public. In addition, the estimates of aggregate incidence would be
higher if such worker populations were included in the analysis. Of
course, any controls to reduce gasoline marketing emissions would reduce
exposure for workers as well as for the general public.
The unit risk factors used in the analysis for the four pollutants,
presented in Table 1-3, were developed by the EPA's Carcinogen Assessment
Group based on available health studies. Two values of unit risk are
shown for gas vapors - a "maximum likelihood estimate" and a "plausible
upper limit." Both values are used in the analysis to provide a broader
base for evaluating the impacts of exposure to gasoline vapors. For a
detailed description of the derivation of the gasoline vapors unit risk
numbers, see the EPA staff paper "Estimation of the Public Health Risk
from Exposure to Gasoline Vapor Via the Gasoline Marketing System",
June 1984. The risk factor for benzene is based on studies of humans
occupationally exposed to benzene. The risk factors for gasoline
vapors, EDC, and EDB are based on animal studies only. Because of the
significance of the gasoline vapor animal studies, conducted for the
American Petroleum Institute, they are examined in detail in the EPA
staff paper cited above. The staff paper was submitted on June .22,
1984, to the EPA Science Advisory Board for review.
There can be substantial uncertainty in unit risk factors. Reasons
for this uncertainty include extrapolations which must be made from
1-13
-------
TABLE 1-3. UNIT RISK FACTOR SUMMARY
Pollutant
Unit R1ska
(probability of cancer
given lifetime exposure
to 1 ppm)
Health Effects
Summary
Comments
Gasoline Vapor
Plausible Upper Limit:b
Rat Studies
Mice Studies
3.S x ID-3
2.1 x 10-3
Maximum Likelihood Estimates:
Rat Studies
Mice Studies
Benzene
Ethylene
Oibrcmide
Ethylene
Bichloride
2.0 x 10-3
1.4 x 10-3
2.2 x 10-2
4.2 x 10-1
2.3 x 10-2
Kidney tumors in
rats, liver tumors
in mica.
Human evidence of
leukernogenicity.
Zymbal gland
tumors in rats,
lymphoid and other
cancers in mice.
Evidence of carci-
nogenicity in
animals by. inhalation
and gavage. Rats:
nasal tumors; Mice:
liver tumors.
Evidence of carci-
nogenicity in
animals. Circulatory
system, forestomach,
and glands; Mice:
Liver, lung, glands,
and uterus.
Gasoline test samples in
the animal studies were
completely volatilized,
therefore may not be
completely representative
of ambient gasoline vapor
exposures.
EPA: listed as a hazardous
air pollutant, emission
standards proposed.
IARCC: sufficient evidence
to support a causal associ-
ation between exposure and
cancer.
EPA: suspect human carci-
nogen; recent restrictions
on pesticidal uses.
EPA: Suspect human
card nogen. Draft
health assessment
document released
for review March 1984.
-i *? I" te"?n °f ttf Probability of a cancer incidence (occurrence) in
individual for a 70-year lifetime of exposure to l ppm of pollutant.
x 1 1™1 * is Ca1cu1ated as «» 95 percent upper-confidence limit of the
« f to exposure to 1 ppm of gasoline vapor, using the multistage dose-
response model for low-dose extrapolation.
IARC: International Agency for Research on Cancer.
1-14
-------
workers or animals to the general population and from the higher concen-
trations found in studies to the lower concentrations found in the
ambient air.
Estimates of risk due to exposures from bulk terminals, bulk
plants, service stations, and self-service vehicle refueling were
generated for each of the four pollutants. In order to calculate
community exposure to emissions (and the- resultant risk) from bulk
terminals and plants, assumptions were made concerning their geographical
distribution. The fundamental assumption was that facilities were
located in proportion to the gasoline throughput for an area—for
example, the largest model plants would be located in large urban areas
where throughput (and population density) were highest. Further, each
model plant type in each source category (bulk terminals and bulk
plants) was distributed over a range of ten urban area sizes. The
largest terminals, for instance, were assumed to be located in cities
ranging in size from New York City to Des Moines, Iowa; the smallest
terminals were assumed to be located in cities ranging in size from
Spokane to Effingham, Illinois. Estimates were also made of the extent
of existing control at these terminals. Most of those in the large
cities (likely to be ozone nonattainment areas) were considered controlled,
based upon existing regulations, with proportionately fewer facilities
controlled in the smaller areas.
Thus, for both terminals and bulk plants, there were 40 model plant
locations (four model plant sizes each distributed to 10 representative
areas) for which estimates of ambient concentrations, population exposure
and incidence were made. Total national incidence was calculated by
multiplying the model plant incidence by the number of facilities
represented by each model plant. In somewhat similar fashion, model
service stations were allocated to 35 localities (multi-county metropol-
itan areas or single counties) and grouped by seven population size
ranges. The model plants were selected to be representative of total
national service station distribution. The localities and seven popula-
tion size ranges were selected to be representative of the total national
population distribution.
1-15
-------
Ambient concentrations, exposure, and incidence for bulk terminals,
bulk plants and service stations were calculated using the SHEAR version
of the EPA Human Exposure Model (HEM). The HEM is a model capable of
estimating ambient concentrations and population exposure due to
emissions from sources located at any specific point in the contiguous
United States.
Annual incidence due to self-service vehicle refueling was
estimated based on benzene and VOC concentrations in the region of
the face of a person filling the tank, as measured in a study for the
American Petroleum Institute (API).* API selected thirteen gas stations
in 6 cities in which samples were collected to characterize typical
exposures to total hydrocarbons, benzene and eight other compounds.
Samples were collected using MSA-type battery operated pumps operated at
a one liter per minute flow rate and analyzed using a gas chromatograph.
Results were expressed as mg/nP air and ppm (vol.).
The lifetime risk analysis was designed to estimate high exposures
of the four pollutants. The Industrial Source Complex (ISC) dispersion
model was used to calculate annual concentrations in selected years at
a number of receptors in the vicinity of a bulk terminal complex, a
bulk plant complex, and a service station complex. Meteorological data
for several cities expected to produce high concentrations were used.
The highest concentration at any receptor under a given regulatory
strategy was used to estimate the risk over a 70-year lifetime. The
lifetime risk due to self-service exposure was estimated based on the
API measurements and an assumed lifetime exposure pattern of an individual
using a relatively high amount of gasoline (i.e. traveling salesman):
the risk was based upon an assumed exposure to four 10-gallon self-service
refuel ings per week for a working lifetime (estimated as 50 years).
The control cost estimates were developed using a method similar to
that for emissions. Capital and operating cost data were obtained
(largely from previous EPA studies) and developed on a per facility
*C1 ayton E'nvironmental Consultants, Inc. Gasoline Exposure Study for
the American Petroleum Institute. Job No. 18629-15. Southfield,
Michigan. August 1983.
1-16
-------
basis for each model plant size of each source category. These per
facility costs were then combined with data on the number of facilities
requiring controls within each source category. Capital costs over the
35 years of the analysis were incorporated during the initial phase-in
years and then repeated in the years in which the economic life of the
equipment ended if replacement equipment was required. Annualized
costs reflected the capital costs and also were adjusted each year, as
appropriate, to reflect reduced recovery credit due to the assumed
decreases in gasoline throughput.
1.3 RESULTS OF REGULATORY STRATEGY ANALYSES .
Although only results of strategies involving size cutoffs are
given in this summary, all strategies are discussed in detail in Chapters 2
through 9. The estimated risks from the various source categories are
given in Table 1-4 for baseline (no additional controls) and when
controlled. Table l-4a contains estimated risks using the plausible
upper limit unit risk factor for gasoline vapors and Table l-4b contains
the estimates using the maximum likelihood estimate unit risk factor
for gasoline vapors.
The lifetime risk from high exposure estimates the probability
that exposure by an individual to a relatively high ambient concentration
throughout his lifetime would result in a cancer incidence. The lifetime
risk from high exposure to bulk terminal emissions is higher than the
lifetime risk from high exposure to uncontrolled emissions from any of
the other source categories.
The average annual incidence for each regulatory strategy is the sum
of the estimated average annual incidences from each industry segment
expected to result from exposures during a given year. (The estimated
annual incidences decrease during the study period in proportion to a
projected decrease in gasoline consumption.) The average annual inci-
dences given are estimates of cancer incidence due to benzene and
gasoline vapors; for the latter, results are shown based on both mice
and rat health studies. Subsequent incidence numbers in this Executive
Summary will be given in terms of rat data only (for both plausible .
upper limit and maximum likelihood estimates) as the rat numbers are
higher and the tables can be simplified. The incidences due to EDB and
1-17
-------
TABLE l-4a. ESTIMATED RISKS FROM GASOLINE MARKETING
SOURCE CATEGORIES
(USING PLAUSIBLE UPPER LIMIT-UNIT RISK FACTOR FOR GASOLINE VAPORS)
A. BASELINE
Source Category
Bulk Terminals
Bulk Plants
Service Stations
Self-service
Source Category
Bulk Terminals
Bulk Plants
Service Stations0
Stage I controls
only
Stage II controls
only
Onboard controls
only
Lifetime Risk from
High Exposure
(probability of effect)
Bz(GVa)
1.2 x 10-4(2.4 or 3.9
6.4 x 10-6(1.2 or 2.0
2.4 x 10-6(4.4 or 7.2
1.1 x 10-5(5. 5 or 9.0
B. CONTROLLED WITH SIZE
x 10-3)
x 10-4)
x io-5)
x 10-5)
CUTOFFSb
Lifetime Risk from
High Exposure
(probability of effect)
Bz(GVa)
2.0 x 10-5(3.9 or 6.4
1.7 x 10-6(3.2 or 5.3
1.6 x 10'6(2.9 or 4.7
1.3 x 10-6(2.5 or 4.2
1.6 x 10-6(2.9 or 4.8
x 10-4)
x 10-5)
x 10-5)
x 10-5)
x 10-5)
Average Annual Incidence
Over 35 years (1986-2020)
Bz(GVa)
0.07(1.3 or 2.2)
0.04(0.68 or 1.1)
0.19(3.3 or 5.5)
3.2(19 or 31)
-
Average Annual Incidence
Over 35 years (1986-2020)
Bz(GVa)
0.05(0.92 or 1.5)
0.02(0.32 or 0.53)
0.17(3.0 or 4.9)
0.11(1.9 or 3.2)
0.10(1.8 or 3'.0)
Self-service0
Stage II controls 5.1 x 10~7(2.S or 4.1 x 10~6) 1.2(6.8 or 12)
only
Onboard controls 7.6 x 10~3(4.4 or 7.2 x 10-?) 1.1(5.9 or 9.8)
only
a
8z * benzene, GV = gasoline vapors (mice or rats studies). Incidences and
lifetime risk due to gasoline vapors are presented to reflect the two unit
risk factors (for liver cancer in mice or for kidney cancer in rats).
b
Based on theoretical control efficiencies.
Indicates reduction of lifetime risk from high exposure for controlled sources.
1-18
-------
TABLE l-4b. ESTIMATED RISKS FROM GASOLINE MARKETING
,llt.T SOURCE CATEGORIES
(USING MAXIMUM LIKELIHOOD ESTIMATE UNIT RISK FACTOR FOR GASOLINE VAPORS)
A. BASELINE
Source Category
Bulk Terminals
Bulk Plants
Service Stations
Self-service
Lifetime Risk from Average Annual Incidence
(proSaSlitroTeVfect) . ^^ ** ^ (1986-2020>
Bz(Gva) BzfGvai
1.2 x 10-4(1.6 or 2.2 x 10-3)
6.4 x lO-^S.Z x ID'5 or 1.1 x 10"4)
2.4 x 10-6(2.9 or 4.1 x 1Q-5)
1.1 x 10-5(3.7 or 5.1 x 10-5)
0.07(0.90 or 1.3)
0.04(0.46 or 0.64)
0.19(2.2 or 3.1)
3.2(13 or 18)
B. CONTROLLED WITH SIZE CUTOFFSb
Source Category
Bulk Terminals0
Bulk Plants
Lifetime Risk from
High Exposure
( probability of effect)
Bz(GVa)
2.0 x 10-5(2.6 or 3.7 x 10~4)
1.7 x 10-6(2.2 or 3.0 x 10'5)
Average Annual Incidence
Over 35 years (1986-2020)
8z(GVa)
0.05(0.62 or 0.86)
0.02(0.22 or 0.30)
Service Stations0
Stage I controls
only
Stage II controls
only
Onboard controls
only
1.6 x 10-6(1.9 or 2.7 x 1Q-5)
1.3 x 0-6(1.7 or 2.4 x 10-5)
1.6 x 10-6(2.0 or 2.8 x lO'5)
0.17(2.0 or 2.3)
0.11(1.3 or. 1.8)
0.10(1.2 or 1.7)
Self-service0
Stage II controls
only
Onboard controls
only
5.1 x 10-7(1.7 or 2.3 x 10"6)
7.6 x 10-3(3.0 or 4.1 x 10~7)
1.2(4.8 or 6.6)
1.1(4.0 or 5.6)
Bz - benzene, GV = gasoline vapors. Incidences and lifetime risk due to
gasoline vapors are presented to reflect the two unit risk factors (for
liver cancer in mice or kidney cancer in rats. racwrs i ror
b
Based on theoretical control efficiencies.
Indicates reduction of lifetime risk from high exposure for controlled sources.
1-19
-------
EDC only increase the incidences due to benzene by less than 3 percent
in most cases and by 5 percent or less in all cases. Because the
estimated incidences due to EDB and EDC are relatively small, they were
omitted from the summary tables. The average annual incidence from
self-service refuel ings at service stations contributes about 80
percent of the total incidence from all source categories. The annual
incidences due to service stations without any additional controls are
approximately 3 for benzene and from about 15 to 36 for gasoline vapors,
considering both community exposures to ambient concentrations and
individual exposures to self-service refueling concentrations.
1.3.1 Nonattainment Area Strategy Results
The effects of vehicle refueling controls in nonattainment areas
are shown in Table 1-5. The primary focus of the nonattainment area
strategies is to reduce VOC emissions in order to attain the national
ambient air quality standard (NAAQS) for ozone; reduction of risk due
to hazardous pollutants is an added benefit. If onboard controls
nationwide under Section 202(a)(6) of the Clean Air Act are used to
replace or supplement nonattainment area regulatory strategies, in
addition to the primary aim of reducing VOC emissions in some or all
nonattainment areas, VOC and hazardous emissions are also reduced
nationwide. It should be noted that the strategy of Stage II controls
in all nonattainment areas also includes the costs, emissions, and
risk reductions for Stage I controls at service stations in two non-
attainment areas (Atlanta and Phoenix) where they currently are not
installed.
The average annual baseline level of VOC emissions (and zero addi-
tional control cost) from service stations of 91,300 Mg/yr is given in
parentheses in Table 1-5. The average annual VOC emission reduction
from the baseline, net present value of control costs, and discounted
cost effectiveness due to Stage II refueling controls were estimated to
be 20,600 Mg/yr, $210 million, and $940/Mg VOC, respectively, for the
"selected nonattainment areas" (NA*) strategy and 60,900 Mg/yr, $570
million, and $870/Mg VOC, respectively, for the "all nonattainment
areas" (NA) strategy. The costs and cost effectiveness values for
strategies involving onboard controls are much higher (costs about $2
1-20
-------
oo
^^
LU
Q£
^f
"^
Ly
S
• i
*?
^^
"^
LU
l
•!•
«? 1
|_
1
1
C r-
o 10 .
-o li =Jaj ^
C CJ CJ O
o a> cn-o i _.
4J -^ S. "tj vo N
"3 > 3) C CO 03
SE ^ ^ i"^ en
a> < — i
oo •«-*
tn
tn
t3 a>
^^ CJ C3
C 4J > >
3 tn oo
z 4-> en
o ^™*-
§ =
=" 5« -~
>— CO
ro c - cj encM
3 O<0 = E O
S-r- S S C\J
c tn O •*- .
a> "s o -u S to
C»LU 3 ,
3
IO
1 i
»
> en co r- rt 5
^f ^P CO
O O O O o
C f-l !•« O O O
i CM LO en — i 5
-H CM CM
•^
§O O C3 O a
O C3 O O o
co vo en (*) en ^^
ca* c^ i«* r^ *
*.
^ =2
•*-* WJ «^» *W 4^ «
5 S til jS 2 afi^
*^- SO i> 3 S (/^ '
1 S C +4 Z h4 44 4-> !
00 Z 44 (U •*- 44 3 !
•-•a; c—o»<0 Z t? U ^ 2 Z N S
C < r— ^ -— ^—44 "S P— JO ^-)o
O C£ CJ = rs o*j^- o»i- O crjojc: a> &. c '
t/» ^ UJ C5 "3(Oi/J (*3tO .2 ^SO fO(OO J
CD OO Q£ V> (/} O <-O OO
i
i
I
i
,1
1
in
•2
u
c
CJ
u
'£
s
4J
0
en
c
1O
>>
o
en
1
Ik.
jj
U
u-
c
n reductio
0
(/}
t/»
ff
LU
/«
CJ
u
2
a>
tn
"3
en
c
*^
3
"°
tn
"ie
3
^
>
•*••
C
u_
o
xposure
CJ
§
^
c
0*
CJ
)
s s
*fc- o
t/1 O)
•i *^
^ Hz
s- o
c
*•»
.££
42
£
c
3
a)
re
^
4J
a>
•o
o
£
P~
0)
^£
r~*
maximum
en
^
tn
3
tn
a>
IO
!
c •
CJ >,
tn I—
a>
s. u
O)
c o.
> O)
tn s,"
CJ 4-1
.a u
3 <2
0 "«
4J S.
— * •*—
^ 3
C
0 4->
10 "=
4-» •*—
4-> CJ
10 a.
i- a.
~ 3
in c;
S- i—
o ja
CL-^
10 tn
> 3
IO
a> •—
c a.
i— -a
o c
ie
u; s-
a
;» 10
s
«
c
Ol
IO
4->
«
C
•*-
c
1
1
^a)
y
>^
.c
CJ
L.
o
>
•—
; emission
c
*i
fO
_m
I^J
CO
en
_>>
o
en
4->
c
i
J:
10
'4->
4->
rO
j:
4J
er
S
CJ
en
4J
•5
c
S
«
t
^
^
Z
>
(O
OJ
44
§
i Nonattai
OJ
o
CD
'S
OO
w
0)
'2
?
"§
u
»
•
S
en
c
a>
c
o
N
o
ti_
c
CJ
1
IO
4J
4->
C
J
48
cu
.£? *
o "a.
4J =
IO
4J
S 4J
is
0.1
w >s
S ,3
4ta C
•—
IQ IO
«s
•a c
'wli
*io i— i
4J
. 4->
• OO
"c o*
in °.
a> o
i
1 1
5"^
44
« -*•
O *" *
c *
cS
'"* *^^
o "^
o *w
a>
O C3
in r^
tn en
•a -a
10 10
c: "5
0 0
o*
i —
T.
-1
s-
IO
o
c
°
-(•
^
^
-------
billion and cost effectivenesses about $5,000/Mg VOC) when considering
only the emission reductions in the affected nonattainment areas and
nationwide costs for onboard controls. The annual average incidences
presented are the cumulative nationwide cancer incidences expected to
result from service station community and self-service exposures during
the 35-year period under the given regulatory strategy averaged over the
35 years. As can be seen from Table 1-5, the incidence reduction due to
Stage II controls in nonattainment areas are relatively small (less than
one incidence per year due to benzene) compared with that associated with
onboard controls nationwide (more than two incidences per year due to ben-
zene reduced). The results shown in Table 1-5 are based on theoretical
control efficiencies and do not account for reduced efficiency in-use.
1.3.2 Nationwide Strategy Results
The issues involved in gasoline marketing operations do not" relate
to ozone attainment only -- there is concern for hazardous exposure also.
This section presents summaries of the results of analyses of alternative
nationwide regulatory strategies and combined nationwide/nonattainment
strategies. Table 1-6 shows the estimated baseline annual average
incidence for benzene and gasoline vapors, and the residual incidence
and cost for each of the three primary nationwide strategies: Stage I,
Stage II and onboard. The baseline benzene incidence attributable to
vehicle operations emissions (tailpipe and evaporative) is also shown;
this incidence is more than twice as large as that from gasoline marketing
operations. In addition, the baseline incidence from possible additional
evaporative emissions is shown separate from vehicle operations. This
value was separated because of its uncertainty, being based on preliminary
test results. Controls on gasoline marketing sources will not reduce
incidences associated with vehicle operations.
For gasoline marketing sources, the average annual reduction in
benzene incidence is estimated to be 0.1 with Stage I controls, 2.1
with Stage II controls and 2.2 with onboard. For gasoline vapors, the
comparable numbers are 1.0 or 1.8, 12.3 or 21.6, and 13.5 or 23.7.
Only limited benzene incidence reduction is achieved by removing benzene
from gasoline (2.4-3.2 per year). This reduction primarily results
from reduced exposure from gasoline marketing sources. Benzene tailpipe
1-22
-------
TABLE 1-6. CONTROL OF BENZENE AND GASOLINE VAPORS
FROM GASOLINE MARKETING SOURCES
Regulatory
Strategy
Average Annual
Incidence
Of Cancers Expected
From Exposures
Over 35 years
(1986-2020)
8z (GV«)
Cost Impacts'
of Strategies
(SBillion)
BASELINE
Gasoline Marketing
Vehicle Operations0
Evap. Emissions Mot Captured6
Total
IMPACTS AFTER SELECTEDf
NATIONWIDE STRATEGIES
Stage I - Nationwide
(with size cutoffs)
Stage II - Nationwide
(with size cutoffs)
Onboard Controls
Nationwide
- w/o Evap
- w/ Evap
Benzene Reduction
in Gasoline
Gasoline Marketing
Vehicle Operations0
Evap. Emissions Not Captured
3.5 (23 or 40)
9.7 (NAd)
0.2 (4.0 or 7.0)
13 (NA)
AVERAGE ANNUAL
INCIDENCE REDUCTION
0.1 (1.0 or 1.8)
2.1 (12 or 22)
2.3 (14 or 24)
2.5 (18 or 31)
2.2-2.9 (0 or 0)
0.18-0.23 (NA)
0.09-0.12 (1.5 or 3.3)
1986 NPVb Total 1982
of Costs Dollars Spent
(1986 - 2020) (1986 - 2020)
0.9
1.6
1.9
1.9
7.4-22
3.2.
6.3
9.7
9.7
30-90
GV = gasoline vapors (rat data only), two numbers given represent estimates using
maximum likelihood estimate unit risk factor and plausible upper limit unit risk
factor, respectively.
b
MPV = Net Present Value.
Incidences due to exhaust and evaporative benzene emissions during vehicle operations
£I!,?!*1T!! "Sin9 !" area.source approach similar to that used for service stations.
Jnanticipated evaporative emissions were not considered here (see Footnote e).
d
Not applicable.
Based on preliminary estimate of possible emissions not captured by the existina
Impacts based on theoretical control efficiencies.
1-23
-------
emissions are not affected substantially by benzene removal, since
benzene is formed in the combustion process.*
Costs of additional controls beyond baseline are presented both as
the net present value of cost (discounted at 10 percent to 1986} and the
cumulative value of the estimated expenditures from 1986 through 2020
(all in 1982 dollars). The costs of all available nationwide strategies
are greater than $800 million net present value of costs or $3 billion
cumulative costs. The cost of benzene reduction in gasoline is a
factor of 2 to 10 greater than for the next most costly strategy.
Table 1-7 summarizes the estimates of average annual incidence,
emission reductions, cost, and cost effectiveness for the nationwide
control strategies (with size cutoffs) evaluated.. These impacts also
are based on theoretical control efficiencies. This table presents
more detailed results: average annual incidence under the strategy,
cumulative VOC and benzene emission reductions, net present value of
costs, and discounted cost effectiveness for both basic regulatory
strategies and combinations of strategies. Although Stage I controls
result in large emission reductions at relatively low cost, they result
in substantially less incidence reduction than Stage II or onboard.
Strategies with either Stage II or onboard refueling controls achieve
greater incidence reductions because of their effect on self-service
emissions.
Table 1-8 presents costs, emission reductions, incidence reduction
and cost effectiveness with theoretical and in-use efficiencies for
Stage II and onboard, and gives two levels of enforcement for Stage II.
The in-use efficiency of Stage II programs is highly dependent on the
level of enforcement used, varying from 56 percent with no inspections
to 86 percent with annual inspections. Enforcement costs are not
included in the cost-effectiveness figures given (their impact is addressed
in Chapter 8). Although average annual enforcement costs for Stage II
nationwide with annual inspections 'are about $7.7 million, including
* Black, P.M., I.E. High, and J.M. Lang. Composition of Automotive
Evaporative and Tailpipe Hydrocarbon Emissions. Journal of the Air
Pollution Control Association. 30:1216-1221. November 1980.
1-24
-------
CO
LjJ
oo
a:
o
UJ
a
UJ
vo
Q±
O
4-1 0) 0
§ •££=•
o o 4-> en
O CJ O S
in oj ^
5 i*-
|jj
=
4J r— O
= 14. d ^
m O E CM
ai 15
S- CD 1
a. 3 *
r— .Q IO
•w re in co
a) >• 4>* en
Z in --
CJ
re en—
3 s'o"
C C CM
C O "O
< •— = CM
in o
01 in "^ i
cn-r- 4J
re s o ic
i- UJ 3 CO
ai -a cn
> N cu X!
< eo ai —
_
=3 0 SO
C i- CM
C in «o
•C in c CM
•— O
o-o en
5 * a-""
i
3 O
C 0) CM —
c u o re
*C C CM ^
CD en
at -a i — .
re u to M
1- C CO CO
aj *— » en
i
•
i
Regulatory
Strategies
with size cutoffs)
!
1
1
i
1
0 O
i i O ro
1* f
|
1
1
O O
0 O
i to en
1 * 0
-o
g1"1 o o
o o
"— '
r- 1
O O O
O O O
O » "
.H ro r*»
f^ to en
U^ 1—4 *— 1
O co -<
t- !. 0
O O
CM
ro o •
CM — en
uo ^- ro
m — i — ^
IV4
Basellnef No Addition;
Controls)
Stage II - Nationwide
Onboard -
Nationwide
o
CO
_4*
o
0
CM
O
o
i
o
s
CM
in
CO
CM
—i
CD
in -r-
re 3
CD C
s. o
* re
«c z
z
•o
1— 1 S-
— re
o
CD .a
en c
re o
S °
G"> — t
ro ^
— *
8 3
"*!. ^
CM
0 0
0 0
"1 ro
O o
0
O* f."
CM CM
— CO
ro
0 °
» S
— <
sn
re
CO
o
•o -c
re "3
— 0
re -z
o
en
o
co
CM
i
in
CM
0
2
s
o
1—4
co
CM
_i
Stage I and Onboard-
Nationwide
o
o
_J"
o
m
«/
o
0
^
CM
|
!?
CM
cn •
o
P^ |
in
CO |
o i
:
'
Stage I, Stage II,
Onboard - Nationwide
]
i
!
i
!
!
i
.;
^
e •»-
3 4->
S U
— a;
x a.
re in
fs qj
t
en
c *
in o
3 4-1
u
in re
01 14.
+•*
JJ 11
in
c 3
CD
in 4^
a) *v
s- s
Q.1—
a) p—
^
c oj-
CD Q.
> CL
•<- 3
CD
*• S
.a in
= 3
3 re
c: r—
O-
f-o
•*-» c
re
— t-
>. o
*= 0
o re
re
re in
•o —
gasoline vapors (rat
11 hood estimate unit r
u a>
O t—
re x
in
O
u
tfc.
O
jr
^
§
4J
C
CD
in
a>
CL
O)
(4.
O
3
in
o>
-C
in
present value of costs
counted at 10 percent)
in
OJ "d
z —
3
4>J
C
CD
CD
CL
O
C
CD
J=
4-1
-a
•c
01
•o
•5
W)
in
0
u
o
CD
3
> •
01 (J
ai CD
i- CX
CL
a) ""
C 4-1
ai
4J
in c
M— 3
O
in y
in in
a) i—
c -o
CD
> j=
— 4J
4J 0
<-> 43
£ .
K- in
CD C
« °
in in
O in
u "^
•o S
a)
4^ 14—
= O
3
o a>
U 3
a >
U 1
in
-------
CJ
CO
CT»
CO
c u-
E -r-
3 4->
S U
S CD
£
to
to
LU
SS
Ul
! "S
0) 4->
UJ
O A9r-
^ !
UJ
UU
CO
T5 8)-—
-i
! = r-
a
LU
to
(00 —
3 -a
ca
to
-< o
co
UJ
£"5
,-:
LU O>
•r- t- O 1-
18 T- JJ — +J
"
i 7- 4-> — 4J !
•—-os- i- i. — -a
.„ -
01 -a 10 iJ -a •—
— . ! Q -P
O 4
a. >
(O = —
S — .
^- — 4)
i— U CTi
1-26
-------
them would cause only a slight increase in cost effectiveness. The
in-use efficiency of onboard controls is expected to be about 92 percent
(based on current levels of tampering to use leaded gasoline, disregarding
the phase-out of leaded gasoline). The enforcement costs for onboard are
lower than Stage II (average annual cost of $0.1 million) since they
are for the incremental cost above the current certification and in-use
testing program, and the incremental costs of inspecting onboard rather
than evaporative control systems on selected vehicles at the assembly
line.
Figure 1-2 graphically depicts annual incidences due to benzene
with either Stage II or onboard regulatory strategies and with no
additional controls (baseline). Estimated incidence with Stage II is
shown both for the assumed statutory phase-in requirements and for an
alternative phase-in schedule suggested by the American Petroleum
Institute (API). The assumed statutory phase-in requirements used in
this analysis are installation within 2 years for non-independents under
a NESHAP (Section 112) and 3 years for independents assuming Section 325
of the Clean Air Act applies to CTG's and Section 112 standards. The
API phase-in schedule assumes 3 years for nonindependents and 7 years
for independents. Installation of Stage II controls was assumed to
begin in 1987 for both the statutory and API phase-in scenarios.
Onboard controls were assumed to be installed on new vehicles beginning
with the 1988 model year. Therefore, all of the strategies shown begin
at baseline levels of about 4.8 incidences expected from 1986 benzene
emissions from the entire gasoline marketing system. The baseline (no
further controls) levels of annual incidence also decrease with time in
proportion to the projected decrease in gasoline consumption. The
Stage II strategies reduce incidence more rapidly than the onboard
strategy. The numbers in parentheses on the graph indicate the differences
in cumulative incidence before or after 1995 when the onboard strategy
is projected to reduce incidence to below the level reached by the
Stage II strategies. Thus, although Stage II can achieve incidence
reduction sooner than onboard, by 2020 the cumulative incidence reduction
with onboard controls has surpassed the cumulative reduction with
Stage II controls, since the steady-state levels of annual benzene
incidence are about 1.2 for Stage II versus 0.5 for onboard.
1-27
-------
Figure 1-2. Effect of Onboard and Stage II Controls on
Benzene Incidence
(Based on Theoretical Efficiencies)
5.0
STAGE II (WITH SIZE CUT-OFFS)
STATUTORY PHASE-IN (2 YRS. NOH-INO., 3 YRS. IND.)
STAGE II (WITH SIZE CUT-OFFS)
API PHASE-IN (3 YRS. NON-INO., 7 YRS. INC.)
ONBOARD (BEGINS IN 1988)
BASELINE INCIDENCE £r
A A
A
A A
( ) DIFFERENCE IN INCIDENCE DUE TO EFFECT OF CONTROL OPTION
A
(4.9 FROM 1986 THROUGH 1995 )
(4.0 FROM 1986 THROUGH 1994 )
(14.7 FROM 1996 THROUGH 2020 )
36 38 90
16 18 20
1-28
-------
The control costs associated with each of the regulatory strategies
are assumed in this study to be passed on by producers to consumers of
gasoline and vehicles in the form of higher prices. The magnitude of
these price increases for components of nationwide regulatory strategies
with size cutoffs are presented in Table 1-9. Most show gasoline price
increases of less than half a cent per gallon of gasoline. Price
increases for benzene reduction strategies range from 1.5 to nearly
5 cents per gallon. These are average figures; in practice they would
vary both with location and over time.
Consumer resistance to these price increases can reduce the sale
of vehicles and gasoline. Estimates of the reductions in the rate of
consumption are displayed in percentage terms in.Table 1-9. The
estimated reduction in gasoline consumption ranges from 38 million
gallons a year for Stage I to 128 million gallons a year for a-combination
of Stage I, Stage II, and onboard controls, and to over 1,200 million
gallons a year for the most costly benzene reduction option. For
regulations involving onboard controls, annual LDV and LOT rate of sales
are estimated to decline by 17.7 and 5.3 thousand vehicles, respectively.
Other economic impacts of the regulatory strategies were also
examined. These included an analysis of the sensitivity of cost
calculations to underlying assumptions and consideration of distributional
impacts of the regulatory strategies by firm size. Results are summarized
in Table 1-10.
1.3.3 Cost Per Incidence Reduction
An analysis was performed to determine the residual costs expended
per cancer incidence avoided for selected nationwide and nonattainment
area regulatory strategies. The residual costs were determined by
obtaining the annualized costs of the controls associated with the
regulatory strategy and subtracting a range of assumed benefit values
per megagram of YOC emissions reduced. The assumed YOC benefits are
those in addition to cancer prevention, such as reduction in non-cancer
health effects and agricultural damage due to ozone. The residual cost
per incidence was then calculated by dividing residual costs by the
appropriate amount of cancer incidences avoided.
1-29
-------
y?
o
o
LU
)
«C
03
^O
UJ
3
UJ
•* 2£
0^
H- 0
15 tJU
"•••% 1 1_
O LU
LU
06 a*
fcS
«C 0
a, uj
o
»~j
C^l
•y
o
o
LU
CTl
»— 4
LU
i
H*
?
i
• i
t
i i
1 "i
i
' £
i
" & i
J 11
« «-i
C£ O— •
' U JJ
, r-0. C
j «•£;. CJ
; o o u
! 3 4J a.
! S S
! CJ in
j SPg
: fe°
-p ^
s ^^
1
1
1
r
i
i
i
f
t
i
! ** o
' i»
flj U
S ""*
< 0
u
1
i
i
1
n
t
t
i
*
t
j
i
i
*
§
•
i
— .
CO
O
*-4
^— """*
fl^
tn •— •
•M l_ *^ O
JS:| f
o"~"
^J
ft*
•
i
•&
^~^
^^
*n — .
g ^
S S
o u
^
fel
8^-S
•^ fMV C7
pHB "^^ «^-
d W> W
S 4J 4J
re c c
c3 a> a>
~*
S
^Ij'lj
4> »*-
«3 O
*•* 3
U) O
>» w
4^ CO
•is
a —
co en
in in
CO CO
d d
r*» r^
to 10
^4 *•*
d d
r~ m CM r~
r~I — en i~l
in » ao in
So in in
^4 ^H O
odd d
CM CM
CM CM
en
-H CM tn CM
odd "^
^^ *-• ~~ o
v in en ej
m to en <»
o o o O
*O •— .fl
*» p c
^ *; o
^ c ^
e o c
re u re
_ _ _ -a —
cj cj cj re cj
en en en o en
crt co en o vi
fn
in
co
d
^
S
to
^M
d
in
§
in
s. re
a. 1
c o
CJ u-
M CJ
1= S-
ej
CQ
.:•,: i:l:,,,::
tA
0)
E
t«-
^^
o
i.
0)
e
e
re
tA
e
o
u
£
X
e
o
ex
tA
IA
re
iJ
•a
CJ
u
>^
u
CJ
o
0
U)
re
1 m
! re
!::' "J:""
tA
e
o
4J
a
3
V
S-
i
o
0
ii
IA
e
a.
e
0)
j=
o
a)
re
, 4M
3
J3
lil
4->
4^
a
V)
O
(J
3
T3
CJ
a:
:~s • :'"
4J
e
0)
u
a
C 0)
0 >,
u — .
IA
0) i.
C CJ
"o i^
IA
reen
^ ""^
a x
3
re -i
IO
g,~
re ..
u s_
cj re
> CJ
e "IA
o e
o
•a i—
01 r—
u> re
re cn
cn
c
CJ X
CJ IO
a- ^^
-o
1-30
-------
TABLE 1-10
ECONOMIC CONSIDERATIONS
The costs of nationwide strategies without facility exemptions are 80
to 100 percent higher than for the comparable strategies with exemp-
tions. For strategies covering only nonattainment areas, costs with-
out facility exemptions are 15 to .35 percent higher than for comparable
strategies with exemptions.
Stage I and Stage II controls will cost 1/2 to 2 1/2 cents more per
gallon of throughouput at small gasoline marketing facilities than at
large ones. Facility exemptions reduce this cost differential, but
do not eliminate it.
Facility exemptions improve the competitive position of the smallest
facilities, but small facilities tend to be less efficient than large
f aci 1 i ti es.
This analysis assumes the per vehicle cost for onboard controls is.
$13 per tank. If, however, the cost really is $25, the net present
value of nationwide onboard control costs would increase by more than
50 percent, and would then exceed those of nationwide Stage I and
Stage II controls.
This analysis assumes gasoline consumption will decline in the years
ahead. If, however, consumption holds at current levels, then costs
for Stage I and Stage II controls would be less because there would be
more recovery credits.
This analysis assumes the number of gasoline marketing faci1ities re-
mains constant in the years ahead. If, however, the number declines,
then costs for Stage I and Stage II controls would be less because
there would be less control equipment needed.
1-31
-------
In Table 1-11, several of the regulatory strategies are presented
with their corresponding emission reductions. The emission reductions
are presented as the net present value of all the annual emission
reductions over the study period and as'an annualized value representing
equal emission reductions for each year of the study period. The
residual costs were determined assuming several different dollar values
for the benefit of reducing each megagram of YOC emissions. For example,
in Table 1-11 the annualized emission reduction associated with Stage I
is 0.218 million Mg. Multiplying this emission reduction by each of
the assumed YOC benefit values yields the annualized VOC benefit in
dollars.
Table 1-12 presents annualized cost (control equipment and
enforcement costs) and annualized incidence reduction due to benzene
exposure associated with several of the regulatory strategies. " The
cost per cancer incidence avoided, assuming no additional benefits, is
calculated by dividing the annualized costs by the annualized incidence
reduction. Table 1-13 takes this one step further by incorporating the
annualized VOC benefits into the analysis. The values presented represent
the residual cost, assuming varying benefits for reducing VOC emissions,
of reducing cancer incidences due to benzene exposure.
Table 1-14 contains a similar analysis to that used in Table 1-13,
except that Table 1-14 was developed using the sum of the incidences
due to benzene and gasoline vapors. It is assumed that the incidences
due to benzene exposure and the incidences due to gasoline vapor exposure
are additive since the respective exposure results in different types
of cancer incidences (leukemia in the case of benzene exposure and
liver or kidney tumors in the case of gasoline vapor exposure). Net
costs per annual incidence avoided are given using both plausible upper
limit and maximum likelihood estimate risk factors for gasoline vapors.
1-32
-------
OO
UJ
ca
Q
«C
00
OO
o
UJ
UJ
<:
UJ
CO
1 t s
i ' §
o
i CM
1 1
) i cn
c s
0 1 -^
— •• 0
f— 14- I O
f— O if}
SO) — i
4^ 3 1 Vt-
*fO 1
W 2.
+•> i cn
— +; , ^
CO *^ O
c o> i o
o> c o
CO QJ 1 .
Tl .
So '"
> 25 i
! -a *" i cn
co cn s
| M C 1 ^.
fO 3 in
3 I/> 1 4*
C in
5< •
| i cn
O
I Si
i
i
CO
S-
^3 C CO
o> c o sa.
N 0 •—
i— > U S
(a in 3
3 i- -o c
= S co o
i
w -g;
o c
O QJ i^ O
™ 3 « —
a. »— « •—
fa *p- f^
O) tu £
z —
en ">i*3)
CO (J U_
4-J C u_
n 3 JT
3 | +J
cn = —
cc "" S
1
10 co in in _ «• cn
co •— * r*. cn cn r*^ ,— i
•a- <-< CM — i CM in
I
1
1
I
!
p^. CTI vo •- * n in cn
CM co in CM <}• o co
f> CM ^H CM f*0
j
eo cn r— r~- in r-» cn
— i in n —i — < OJ '
o o o o o o o ;
i
^-t ^o *f rr cn mm;
CM O O -H O -* CM
^ tJ
co uo > to ;
^^r:r;r:-o 5 •
i. o
QJ <1> OJ <3J 4) fQ "*^» *^^
cncnCTioicno S3
-------
TABLE 1-12. BENZENE REGULATORY COSTS AND INCIDENCE REDUCED
Regulatory Strategy
(with size cutoffs)
(In-use efficiency)
Stage I
Stage II-NA (87%)
Stage II-NA (56%)
Stage II-Nation (86%)
Stage II-Nation (56%)
Onboard (92%);
w/o evaporative
w/ evaporative
Annual i zed
Costs
($ Millions)3
91
62
52
183
146
199
199
Annuali zed
Benzene
Incidence
Reduction"
0.06
0.83
0.41
1.92
1.13
1.44
1.66
Costs
($ Millions
per Benzene Cancer
Incidence Avoided)
1,564
75
126
95
-128
138
120
Includes control equipment and annual enforcement costs.
Incidence reduction after controls. Before-control annualized incidence:
Stage I - 0.18, Vehicle Refueling = 4.09.
1-34
-------
UJ
o
UJ
0
1— 4
0
I-H
o;
UJ
o
^C •*— ™*
<_5 00
1—
CU H-.
UJ U_
Q- UJ
z
OO UJ
K- 03
OO
0 0
0 0
^>
>•
o =i
h- >-i
< s:
_1 =>
=3 00
•
oo
i
i— i
UJ
_j
ca
•a:
1—
^—,
•a
CO
•o
• fH.
0
>
CO
c i-'
CU 0
•o
•i- CO
(J 3
C p™
(-4 (B
s_ >
CO -P
O •!-
C M-
Q.
cn
to c
C -1-
0 &
•i- 3
t— )
S
^^
*M •
^
+•>
to
0
o
•a
CO CO
M 0
•r- C
r— CO
3 -f
C u
C C
«t 1—4
^^
cn
+j
(^
s_
4^
OO
>^
SM.
O
4»
rO
i—
3
CD
CO
a;
cn
s:
o
0
o
CM
CD
21
O
O
tn
i— i
*«•
cn
s:
o
o
o
t— 1
cn
^F
^•^
o
o
uo
4V^
cn
2!
0
to
1 ^,J
1 ^^
C -^•^
O CO
•r— C
•P CO
3 £=
•o co
CU CO
o;— •
^~»
>>
c
CO
•r—
0
*4™
^4—
CO
CU
to
3
1
C
1—4
— •
O O O O O O O
O O O O CM O O
o ^r uo cri «* LD o
.a
^3 ^^ t-H i^* ^O CO f^O
CO CO LO CO O^ ^"
^^ ^^ ^^ ^^ r^^ ^7 co
CM lf> O f-> O CM CO
IO r- 4 ,— 1 '^4
*«o
co co ^^ CT^ *"•' ^f r*^
O O O ^H r— 1 r-l »-H
S* 3*
to vo
CO UO CO CO
&$ ^S *p™ *r-
r-» to c: c: -P -p
CO IjO ^D ^3 * * ^O ^3
— ~— -r- -i- — . i. S.
•P -*-> &S O O
^ ^ (O fQ CM CL CL
'z. -z. "zz. "z. cn >
i— i t— i i— i i— i CO CO
1— 1 1— 1 )— 4 . )— 4 1 1 "O
1- O
CO CO CO CO CU (O *^> • —
cn cn cn cn cn o 5 !5
(T5 (O rt3 (O (O <^
•P 4*^ 4~* P P C
oo oo oo -oo oo o
^^
1—4
,— H
1
r— 1
CO
"re
1 —
s_
"~^
I/)
^(J
>rw
<4-
co
CO
ca •
o
o
-a
cu
N
•r—
to
J=
"*"*
CO
-p
o
•p-
•a
(^
•t—
t/1
o
i.
CO
M
1-35
-------
03
O
f*S
O
^g*
LU
jgg* i^i
>-« o
0 Q.
•Si <
»— I *rm
^^ ^^
O± CO
LU <£
0 OS
22
00
fHft
of O
I—
//S ^tf*
C? utv
^-<
ooi
o.
S
• •>» 1 Q-
* '8 ,
<0 | in •
VI *» | UJ
CJ ill ' vt | _J
O I •
k. . E
(Q 19 _J
O > 1
.Q *J 1 Cn 1 •
^ *** E Cw
O»- 1 •». 1
0) o
•f* C 1 CO 1 •
01 o LU
jMca i - i
CO -M — 1
^ ^^ ^*
of
ra i _j
§e i .
— 1 =5
•»•• g | O •
>- 9 s i a-
I— U> t ^
•»- at o I
E< t 0
in i uj
44- | V* •
^•^ 1 _J
1 •
4** 1
O 1
O •
1 _J
1 3
i E ol
. 0 '
in i •
i cv ui
i _J
( '
i
•a aj c
2 o *^
r— -OotS — I- 4J _J
C C CJ CQ (O QB>^ •
< a. —
•a ffl c= 'o
.
*^cj4J^ ^j=:iouj
r— T3 O O •<- — S •
<0— 3 Hh X^— -^ _J
3U13IM W U *» .
c c 01 m EJKV>E
«C «J
Si s;
, •£
: 2 «
* M Si
' -3 ?
en c
1 QJ »^«
r** •*—
o o o o o o o
o o o o o o o
o o o o o o o
O O O O O 'O O
O o **"> e»j ^* ^r o
O — « Lf) CO ^JO IO O
O C*) *-O IO CO GO PO
.
o ir> ^-* eo <*) ro in
*— * f— t ^H
O ^0
1—4
CO ^f — i ^O IO O
ro
01 «-l
•M o t" CNJ f r^» **^
^* OU ^ -HCJ
^4 —I —4 CO
~« to ro en ^ o^j
9* &* 4) 4J
>
co u^ *^ ^
6-Q (V<9 (Q to
r- uo c c i- i-
co to o o o o
< «C « (0 CSJ > >
^ z z sz en o> — t »— 1 *— « O
S- 5 S
OJ OJ «3 ^
(/> t/> CO tO t/> O
1-36
«««
^4
f— t
1
«
i
H-
2
**•
(/I
4J
•^
4->
•^ w)
— o
> °
IO IT
HI
CU N
e 1^
HI IO
•O 3
•f- C
I 5
s_
D
u
IO
(J
u
CJ
VI
o
ia
o
4^
u
u_
&.
V)
s-
4J
c
3
+J
"s
r—
HI -
a.
a.
3
0)
•^
t/>
3
"S.
•a
c
ia
^
o
*J
(J
IO
JH
VI
u
4->
e
0)
4->
13
VI
at
•o
o
o
j=
1 S
IO
> S
IO f-
cn x
10
^ "™
i— CU
a. ^
j->
HI
3
+. (J
N (O
.-.-
,
VI
4J
VI
o
u
HI
J=
4-* •
J=
O>
^
VI
4->
•t—
(fa.
OJ
e
01
s
>.
J=
4J
(O
-C
4->
CD
y
•o
c
VI
2
HI
rvi
•a
-------
2.0 INDUSTRY DESCRIPTION AND PROFILE
2.1 GENERAL INDUSTRY DESCRIPTION
The gasoline marketing network consists of all the storage and
transfer facilities that move gasoline from its production to its
end consumption. The network includes tanker ships and barges, pipelines,
tank trucks and rail cars, storage tanks, and service stations. Crude
petroleum is shipped to refineries, which manufacture the wide range of
liquid petroleum products. Finished gasoline is then distributed in a
complex system comprised of wholesale and retail outlets. Figure 2-1
depicts the main elements in the marketing network.
Gasoline is delivered to bulk terminals from refineries by way of
pipeline, ship, or barge. Large transport trucks (30,000 to 38,000
liters, or 8,000 to 10,000 gallon capacity) deliver the gasoline to
service stations or to intermediate bulk storage facilities known as
bulk plants. Generally, a terminal is defined as any bulk wholesale
gasoline marketing outlet that receives product by pipeline, ship, or
barge, and delivers it in tank trucks to customers. A bulk plant
typically receives product by truck from a terminal and has a smaller
storage capacity than a terminal. In addition, daily product throughput
at a terminal is much greater, averaging about 950,000 liters
(250,000 gallons), in contrast to about 19,000 liters (5,000 gallons)
for an average size bulk plant.
Both bulk terminals and bulk plants deliver gasoline to private,
commercial, and retail accounts. Bulk plants, using 5,700 to 11,000
liter (1,500 to 2,900 gallon) capacity delivery trucks, service primarily
agricultural accounts and service stations that are either long distances
from terminals or inaccessible to the large transports. The trend in
recent years has been toward more terminal deliveries at the expense of
bulk plant deliveries. Retail and commercial level businesses include
the familiar service stations, as well as commercial accounts such as
fleet services (rental car agencies, private companies, governmental
2-1
-------
Imported
Gasoline
Ref i nery
w i
p
Barge
Pi pel-
Tanker /
A
me )
Bulk
Terminal
'Imported
or
Domestic
Crude
Wholesale
Distribution
Level
Commercial,
Rural
Consumer
Figure 2-1 . Gasoline Distribution in the U.S.
= Storage
= Transport
2-2
-------
agencies), parking garages, and buses. Another important consumer
category is about 2.7 million small farms.
2.2 GASOLINE .MARKETING OPERATIONS AND THEIR EMISSIONS
The pollutants from each of the gasoline marketing facilities are
essentially the same; however, the operations which occur at each of
the facilities differ, and the rates of emissions to the atmosphere
differ. The emissions consist of a mixture of volatile organic compound
vapors and air, and are discussed in Section 2.2.1. Because of the
complexity of the industry, Sections 2.2.2 through 2.2.6 will present
separate discussions of the operations at each industry sector and of
the associated emission rates of gasoline vapors, benzene, ethylene
dibromide (EDB), and ethylene dichloride (EDC) from typical facilities.
2.2.1 Gasoline Composition
Motor gasoline is a complex mixture of varying amounts of paraffins,
naphthenes, olefins and aromatics (such as benzene). In addition,
small amounts of additives (such as tetraethy lead, 1,2-dichloroethane
[or Ethylene dichloride (EDC)] and 1,2-dibromoethane [or ethylene dibromide
(EDB)] are used to achieve specific product qualities. This section
discusses and provides estimates of the amount of benzene, EDB and EDC
vapor in gasoline vapor.
It is well known that wide variations in benzene content exist in
gasoline. Analysis of 1977 gasoline pools produced by various refin-
eries showed that benzene content in the liquid gasoline varied from
0.15 to 4.26 volume percent with a national average of 1.3 volume
percent.1 In a review of more limited data over the period since 1977,
the benzene content variation was within the 1977 range while no
difference below or above the 1977 average was apparent.2 Only one
published source was found reporting the concentration of EDB and EDC
in liquid gasoline.3 This study estimated that the EDB and EDC
concentrations varied from 80 to 150 and 150 to 300 ppm, by volume, in
leaded gasoline, respectively.
An earlier EPA study on benzene emissions from gasoline marketing
concluded that about 0.008 grams of benzene per gram of hydrocarbon
vapor at 26.7°C (SOT) was an appropriate figure to use for 1.3 volume
percent benzene in liquid gasoline.^ The figure of 0.008 was based on
laboratory and field tests conducted at temperatures varying from
2-3
-------
25 to 31°C.5»6>7 The study also warned that any attempt to adjust the data
to other temperatures would introduce an indeterminable degree of
error. In addition, it is well understood that temperature has a major
influence on vapor-liquid equilibrium concentrations. The remainder of
this section describes the approach used to estimate the relationship
of temperature on the gasoline vapor-liquid equilibrium.
The equations shown in Figure 2-2 were used to calculate the
emission estimates for VOC, Benzene, EDC, and EDB. Equation 1 is the metric
equivalent to the AP-42 emission formula for YOC loading and was derived
from the ideal gas law. Instead of obtaining the true vapor pressure
from the ASTM distilation curve and the molecular weight from tables,
equations 2 and 3 were used to add the capability of varying temperature
and Reid vapor pressure (RVP).8»9 Equations 4, 5, and 6, which were
used for determining Benzene, EDC and EDB emissions, were developed
from the ideal gas law, assuming equal vapor to liquid volume ratios
and temperatures, an activity coefficient expression, and the Antoine
vapor pressure equation and coefficients.
Uncontrolled and controlled VOC emission factors and data are
readily available from the many sources referenced in subsequent sections
of this report. In order to estimate benzene, EDC and EDB losses,
uncontrolled emissions for the four pollutants were estimated by using
the equations presented on Figure 2-2. Figure 2-3 graphically presents
the mass emission factors, assuming a saturated vapor space. Benzene,
EDC and EDB to VOC emission factor ratios were calculated by simply
dividing each of the three pollutants emission factors by the VOC
emission factor. Figure 2-4, graphically presents these emission
ratios. Throughout the evaluation discussed in this report, these
ratios (for RVP 10) were used to estimate benzene, EDB and EDC emis-
sions in known amounts of hydrocarbons. Different ratios, based on
product temperature, were used to calculate emissions from vehicle
refueling operations than were used for all other gasoline marketing
sources. Ratios based on 70°F were used for vehicle refueling (BZ/VOC-
0.0066, EDC/VOC-0.00053, EDB/VOC-0.000052) while ratios based on 60°F
were used for all other gasoline marketing operations (BZ/VOC-0.0060,
EDC/VOC-0.00047, EDB/VOC-0.000046). These ratios were used for community
exposure from service stations, but were not used to calculate risks
2-4
-------
FIGURE 2-2 EMISSION EQUATIONS
= 1492
( P M SF \
\ T + 459.S/
(1)
(r / 413.0 \-| / 1042
= 6XpJ 0.7553 - \T + 459.6jl S0-5 log (RVP) - 1.854 - \T + 459.
U L
[/ 2416 \ -, / 8742 \ )
+ |_\T + 459.6 /- 2.013J log (RVP) - \T + 459.6 ) + 15.64>
0-5
M = A + B(RVP) + C(RVP)2 +. D(RVP)3
\ v SF
EBZ = 18.34
) 2788.51
exp U5.9008 - K-52.36
EEDC = 25
(_LL§F\ \ I 2927;17 \
.58\ K ybxp \ 16.1764 -\K - 50.22 /
\ . i
EEDB = 44.95
Where:
\ v SF
K
'.568 -
1903.6
10
(2)
(3)
(4)
(5)
(6)
A = Constant for Gasoline = 72.833
B = Constant for Gasoline = -1.3183
C = Constant for Gasoline = 0.15079
D = Constant for Gasoline = - 0.0087302
= Loading Emission Factor (mg BZ/liter loaded)
= Loading Emission Factor (mg EDB/liter loaded)
EEDC = Loading Emission Factor (mg EDC/liter loaded)
EVQC = Loading Emission Factor (mg VOC/liter loaded)
K = Stock Temperature (°K)
M = Molecular Weight of Vapors (Ib/lb-mole)
P = True Vapor Pressure (psia)
RVP = Reid Vapor Pressure (psi)
S = Slope of ASTM distillation curve at 10 percent evaporated;
for gasoline, S = 3
SF = Saturation Factor
T = Stock Temperature (°F)
V = Volume Percent of Compound in Gasoline (Assume EDB = 0.015% by Volume,
EDC = 0.03% by Volume)3
= Activity Coefficient (Bz = 1.17, EDB and EDC = 2.1)
exp x [ = ex
Reference 3.
2-5
-------
Figure 2-3. Mass Emission Rates ^Saturation
10000
1000
-o
O)
(A
c
ra
•4-s
52
E
rc>
to
•fj
(13
o:
c
o
0.01
30 40 50 60 70 80 90 100 110 120
Stock Temperature (°F)
2-6
-------
Figure 2-4. Mass Emission Ratio of BZ, EDC, & EDB to VOC in Vapor
0.01
- 1
j _
1
-=.•• l
• ^
~^= -1
^
~- .
1
r i n RVD ir^=-
to
a>
l/l
c:
QJ
-o
i.
o
o
o
CO
a
o
Q
O
O
to
1/1
in
1/1
(D
0.001
0.0001
0.00001
Benzene g- =^=--5=
30 40 50 60 70 80 90 100 110 120
Stock Temperature (°F)
2-7
-------
from benzene and-gasoline vapor exposure during self-service refueling,
which were based on measured exposure data.
2.2.2 Bulk Terminals
Bulk gasoline terminals serve as the major distribution point for
the gasoline produced at refineries. Movement of gasoline at a bulk
terminal involves loading, unloading, and transfer of the liquid from
storage tanks into tank trucks. Gasoline stored in large aboveground
tanks is pumped through metered loading areas, called loading racks,
and into delivery tank trucks, which service various wholesale and
retail accounts in the marketing network. Loading racks contain the
equipment (such as pumps, meters, piping, grounding, etc.) necessary
to fill delivery tank trucks with liquid products. Terminals generally
utilize two to four rack positions for gasoline, but there can be as
many as eight to ter\rack positions at large throughput terminals.
Each loading rack will typically have from one to four loading arms,
depending on the products available for loading at that rack position.
Each arm is dedicated to one product.
Emissions from the tank truck loading operations at terminals
occur when the product being loaded displaces the vapors in the delivery
tank and forces the vapors to the atmosphere. Loading may be performed
using either splash, top submerged, or bottom loading methods. Top
loading involves loading of products into the tank truck compartment
via the hatchway which is located on top of the truck tank. Gasoline
can be loaded directly into the compartment through a top loading fill
pipe (splash fill). Attachment of a fixed or extensible downspout to
the fill pipe provides a means of introducing the product near the
bottom of the tank (submerged fill). Top splash loading creates
considerable turbulence during loading and can create a vapor mist
resulting in higher emissions from the truck loading operation. Submerged
loading greatly reduces the turbulence, and therefore reduces the
emissions. Bottom loading refers simply to the loading of products
into the cargo tank from the bottom. This results in the same emission
reduction as associated with top submerged loading. The trend in the
industry is to build new terminals with bottom loading racks and to
convert existing terminal top loading racks to bottom loading. Some of
2-8
-------
the advantages cited for bottom loading include: (1) improved safety,
(2) faster loading, and (3) reduced labor costs. Emission factors and
emissions from loading rack operations at a typical 950,000 liters/day
(250,000 gallons/day) terminal are summarized in Table 2-1.
2.2.3 Storage Tanks at Terminals
A typical terminal has four or five aboveground storage tanks for
gasoline, each with a capacity ranging from 1,500 to 15,000 m^ (9,400
to 94,000 barrels).10 Most tanks in gasoline service have an external
floating roof to prevent the loss of product through evaporation and
working losses. Fixed-roof tanks, still used in some areas to store
gasoline, use pressure-vacuum (P-V) vents to control breathing losses
and may use vapor balancing or processing equipment to control working
losses. A typical fixed-roof tank consists of a cylindrical steel
shell /with a cone- or dome-shaped roof that is permanently affixed to
the tank shell. A breather valve (pressure-vacuum valve), which is
commonly installed on many fixed-roof tanks, allows the tank to operate
at a slight internal pressure or vacuum. Because this valve prevents
the release of vapors only during very small changes in temperature,
barometric pressure, or liquid level, the emissions from a fixed-roof
tank can be appreciable.
The major types of emissions from fixed-roof tanks are breathing
and working losses. Breathing loss is the expulsion of vapor from a
tank vapor space that has expanded or contracted because of daily
changes in temperature and barometric pressure. The emissions occur in
the absence of any liquid level change in the tank. Emptying losses
occur when the air that is drawn into the tank during liquid removal ~
saturates with hydrocarbon vapor and expands, thus exceeding the fixed
capacity of the vapor space and overflowing through the pressure vacuum
valve. Combined filling and emptying losses are called "working losses."
A typical external floating-roof tank consists of a cylindrical
steel shell equipped with a deck or roof that floats on the surface of
the stored liquid, rising and falling with the liquid level. The liquid
surface is completely covered by the floating roof except in the small
annular space between the roof and the shell. A seal attached to the
2-9
-------
Table 2-1. UNCONTROLLED EMISSIONS FROM GASOLINE TANK TRUCK
LOADING OPERATIONS AT A TYPICAL BULK GASOLINE TERMINALa
Emissions, Mg/yr
Loading Method
Submerged Loading^
Top Splash Loading
Balance Service
Gasoline
Vapor
Emission
Factor5
mg/ liter
600
1440
960
Gasoline
Vapors
200
500
300
Benzene EDBC EDCC
1 0.004 0.05
3 0.01 0.1
2 0.01 0.1
Typical terminal gasoline throughput =950,000 liters/day
(250,000 gallons/day).
3Ref. 11.
<%
'Occurs from leaded gasoline only. Assumes for the base year that 48
percent of total throughput at this typical facility is leaded gasoline
(Ref. 18).
Submerged loading could be either top submerged loading or bottom
loading.
2-10
-------
roof touches the tank wall (except for small gaps in some cases) and
covers the remaining area. The seal slides against the tank wall as,, —
the roof is raised or lowered.
An internal floating-roof tank has both a permanently affixed roof
and a roof that floats inside the tank on the liquid surface (contact
roof), or supported on pontoons several inches above the liquid surface
(noncontact roof). The internal floating roof rises and falls with the
liquid level.
Standing-storage losses, which result from causes other than a
change in the liquid level, constitute the major source of emissions
from external floating-roof tanks. The largest potential source of
these losses is an improper fit between the seal and the tank shell
(seal losses). As a result, some liquid surface is exposed to the
atmosphere. Air flowing over the tank creates pressure differentials
around the floating roof. Air flows into the annular vapor space on
the leeward side and an air-vapor mixture flows out on the windward side.
Withdrawal loss is another source of emissions from floating-roof
tanks. When liquid is withdrawn from a tank, the floating roof is
lowered, and a wet portion of the tank wall is exposed. Withdrawal
loss is the vaporization of liquid from the wet tank wall.
As ambient wind flows over the exterior of an internal floating
roof tank, air flows into the enclosed space between the fixed and
floating roofs through some of the shell vents and out of the enclosed
space through others. Any vapors that have evaporated from exposed
liquid surface and that have not been contained by the floating deck
will be swept out of the enclosed space. The withdrawal loss from an
internal floating-roof tank is similar to that discussed for external
floating roofs. The other losses, seal losses, fitting losses and deck
seam losses, occur not only during the working operations of the tank
but also during free standing periods.
Table 2-2 illustrates the magnitude of fixed roof and floating
roof emissions from a 950,000 liter/day terminal with four storage
tanks for gasoline.
2-11
-------
Table 2-2. EMISSIONS FROM GASOLINE STORAGE TANKS LOCATED AT
A TYPICAL TERMINAL3
Emissions, Mg/yr
Storage Method
Fixed Roofd
Working Losses
Breathing Losses
Floating Roof6
Working Losses
Storage Losses
Gasoline
Vapor
Emission
Factor"5 Gasoline
Mg/yr Tank Vapors
34.2 140
8.8 40
f 0.1
9.6 40
Benzene EDBC EDCC
0.8 0.003 0.03
0..2 0.001 0.008
0.001 <0.001 <0.001
0.2 0.001 0.009
Terminal with 950,000 liters/day (250,000 gallons/day) with four storage
tanks for gasoline.
See Appendix B (Section B.2.2).
•*
'Pertains to leaded gasoline only. Assumes for the base year that 48 percent
of the gasoline throughput for this typical facility would be leaded gasoline
(Ref. 18).
Typical fixed-roof tank based upon capacity of 2680 m3 (16,750 bbls.)
a
"Typical floating-roof tank based upon capacity of 5760 m3 (36,000 bbls.)
Emission Factor = (0.46 x 10~7 Q) Mg/yr, where Q is the throughput through the
tanks in barrels.
2-12
-------
•2.2.4 Bulk Plants
Bulk gasoline plants are typically secondary distribution
facilities which receive gasoline from bulk terminals by truck transports,
store it in above ground storage tanks, and subsequently dispense it
via smaller account trucks to local farms, businesses, and service
stations. A typical bulk plant has a throughput of about 19,000 liters
(5,000 gallons) of gasoline per day with storage capacity of about
189,000 liters (50,000 gallons) of gasoline.12 EPA defines the bulk
plant as having a throughput of less than 76,000 liters (20,000 gallons)
of gasoline per day averaged over the work days in one year.
As discussed in the previous section, vapors can escape from
fixed-roof storage tanks at bulk plants, even when there is no transfer
activity. Temperature induced pressure differentials can expel vapor-
laden air or induce fresh air into the tank (breathing loss).- Liquid
transfers create draining and filling losses which combined are called
"working losses".
The two basic types of gasoline loading into tank trucks at bulk
plants are the same as those used at terminals. The first is the
splash filling method, which usually results in high levels of vapor
generation and loss. The second method is submerged filling with
either a submerged fill pipe or bottom filling, which significantly
reduces liquid turbulence and vapor-liquid contact, resulting in much
lower emissions. Table 2-3 indicates the uncontrolled emissions from a
typical bulk plant. In a 1976 survey of bulk plants, 75 percent used either
top-submerged filling or bottom filling and 25 percent used top splash
filling.13 These bulk plants which use top splash filling are typically
located in States with no control regulations, or in attainment areas of
those States with regulations where no control is required.
2.2.5 Tank Trucks
Gasoline tank trucks have been demonstrated to be major sources of
vapor leakage. Some vapors may leak uncontrolled to the atmosphere
from dome cover assemblies, pressure-vacuum (P-V) vents, and vapor
collection piping and vents. Other sources of vapor leakage on tank
trucks that occur less frequently include tank shell flaws, liquid and
vapor transfer hoses, improperly installed or loosened overfill protection
2-13
-------
Table 2-3. UNCONTROLLED EMISSIONS FROM A TYPICAL BULK PLANT3
Emissions, Mg/yr
Emission Source
Storage Tanks
- Breathing Loss
- Filling Loss
- Draining Loss
Gasoline Loading
Racks
Gasoline
Vapor
Emission
Factorb
mg/1i ter
600
1150
460
Gasoline
Vapors
3
7
3
Benzene
0.02
0.04
0.02
EDBC EDCC
<0.001
<0.001
O.001
0.001
0.002
0.001
-
-
-
Splash loading
Submerged loading
Submerged loading
(Balance Service)
1440
600
960
8
3
5
0
0
0
.05
.02
.04
<0
<0
<0
.001
.001
.001
0
0
0
.002
.001
.001
a
Typical bulk plant with a gasoline throughput of 19,000 liters/day
(5,000 gallons/day).
Reference 11.
'Pertains to leaded gasoline vapors only.
percent of the throughput of this typical
Assumes for the base year that 48
facility is leaded gasoline (Ref. 18)
2-14
-------
sensors, and vapor couplers.14 This leakage has been estimated to be
as high as 100 percent of the vapors which should have been captured and
to average 30 percent.15 Since terminal controls usually coincide in
areas where trucks are required to collect vapors after delivery
of product to bulk plants or service stations (balance service), the
emission factor associated with uncontrolled truck leakage was assumed
to be 30 percent of the balance service truck loading factor (960
mg/liter x 0.30 = 288 mg/liter).
2.2.6 Service Stations
The discussion on service station operations is divided into two
areas: the filling of the underground storage tank and automobile
refueli ng. Although termi nals and bul k pi ants also have two di sti net
operations (tank filling and truck loading), the filling of the
underground tank at the service station ends the wholesale gasoline
marketing chain. The automobile refueling operations interact directly
with the public and control of these operations can be performed by
putting control equipment on either the service station or the
automobile.
Normally, gasoline is delivered to service stations in large tank
trucks from bulk terminals or smaller account trucks from bulk plants.
Emissions are generated when hydrocarbon vapors in the underground
storage tank are displaced to the atmosphere by the gasoline being
loaded into the tank. As with other loading losses, the quantity of
the service station tank loading loss depends on several variables,
including the quantity of liquid transferred, size and length of the
fill pipe, the method of filling, the tank configuration and the gasoline
temperature, vapor pressure, and composition. A second source of
emissions from service station tankage is underground tank breathing.
Breathing losses occur daily and are attributed to temperature changes,
barometric pressure changes, and gasoline evaporation.
In addition to service station tank loading losses, vehicle
refueling operations are considered to be a major source of emissions.
Vehicle refueling emissions are attributable to vapor displaced from
the automobile tank by dispensed gasoline and to spillage. The major
2-15
-------
factors effecting the quantity of emissions are gasoline temperature,
auto tank temperature, gasoline Reid vapor pressure (RVP), and dis-
pensing rates. Table 2-4 illustrates the uncontrolled emissions from a
typical gasoline service station. The emission factors presented in Table
2-4 are from EPA's AP-42 document.16 The California Resources Board (ARB)
has performed more recent testing and has estimated that refueling emissions
are about 1200 mg/liter.17 The AP-42 emission factors for vehicle refueling
have not yet been revised, and AP-42 factors were used to estimate emissions
from all other gasoline marketing sources. Therefore, as a matter of
consistency, the AP-42 factors were used to estimate vehicle refueling
losses throughout this report.
Emissions due to breathing losses from the vehicle fuel tanks are
controlled by existing carbon canister systems on the vehicle {evapora-
tive control systems). However, preliminary data from EPA's emission
factors program indicates that in-use evaporative emissions appear to
significantly exceed the evaporative HC standard, primarily because in-
use fuels typically have higher volatility and produce larger amounts
of evaporative HC (when compared to the fuels used for certification
testing) which cannot be absorbed by the current charcoal canisters.
Preliminary estimates of the level of these excess evaporative emissions
are further discussed in Section 3.7.3 (Effectiveness of Technologies).
2.3 BASELINE EMISSIONS
Baseline emissions are the emissions from gasoline marketing
sources in some selected "base" year. The purpose of establishing an
emission baseline is to be able to estimate the impacts of reducing
emissions from this baseline through the implementation of additional
control measures. The baseline emissions must take into account the
level of control already in place in the base year to get an accurate
assessment of the impacts of the control alternatives. The base year
for the gasoline marketing source category was selected as 1982. This
year was selected because this was the final implementation date for
many State regulations concerning gasoline marketing sources and
2-16
-------
Table 2-4. UNCONTROLLED EMISSIONS FROM A TYPICAL SERVICE STATION3
Emissions, Mg/yr
Gasoline
Vapor
Emission
Factors'3
Emission Source mg/liter
Gasoline
Vapors Benzene EDBC EDCC
Underground Storage
Tanks
-
Tank fil ling losses
e Submerged fill 880 -
• Splash fill 1380
- Breathing losses 120
2 0.01 <0.001 <0.001
3 0.2 <0.001 <0.001
0.3 0.002 <0.001 <0.001
Automobile Refueling
-
-
Displacement losses 1080
Spillage 84
2 0.01 <0.001 <0.001
0.2 0.001 <0.001 <0.001
a . .
Typical service station has a gasoline throughput of 190,000 liters/month
(50,000 gallons/month).
b
Reference 11.
c
Pertains to leaded gasoline only. Assumes for the base year that 48
percent of the total throughput is leaded gasoline (Ref. 18).
2-17
-------
because the latest data available on facilities and gasoline consump-
tion at the time of the development of this document were representa-
tive of 1982. Table 2-5 summarizes the baseline emissions calculated
for the gasoline marketing industry in base year 1982.
The general approach for establishing the emission baseline was
basically the same for each sector of the industry. Data was obtained
on the level of control already used by the States and anission factors
were selected to represent this level of control. Uncontrolled areas
were defined and emission factors were selected to represent the type
of loading or type of operations in those areas. Emissions were calcu-
lated by multiplying the emission factors by the.corresponding throughput
for the controlled and uncontrolled areas. Appendix B contains a
detailed discussion of the procedures used to establish the emissions
baseline.
2-18
-------
Table 2-5. SUMMARY OF BASELINE EMISSIONS
FOR GASOLINE MARKETING FACILITIES
FOR BASE YEAR 1982
Emissions, Mg/yr
Faci
Bulk
-
-
Bulk
Gasoline
lity Vapors
Terminals
Truck Loading 140,000
Storage Tanks 56,000
Plants 208,000
Benzene EDBa EDCa
840 3 30
340 1 15
1,250 5 50
Service Stations
-
-
Underground tanks 222,000
Automobile refueling 407,000
Total 1,033,000
1,330 5 50
2,690 10 100
6,450 24 245
a
Applies to leaded gasoline only. Leaded gasoline consisted of approximately
48 percent of total consumption in 1982.& "*„„<»MSijr
2-19
-------
2.4 References
1. "Cost of Benzene Reduction in Gasoline to the Petroleum Refining
Industry," A.D. Little, Inc. Report to EPA. EPA Publication No.
450/3-78-021. April 1978.
2. "Motor Gasolines," Semi-Annual Reports U.S. Department of Energy
by Bartlesville Energy Technology Center. Reports covering 1977 to
Winter 1982-83.
3.
4.
McDermott, H.J., and Killiany, S.E. "Quest for a Gasoline TLV"
American Industrial Hygiene Association Journal (AIHA). February 1978.
Standard Support Environmental Impact Statement for Control of
Benzene from Gasoline Marketing Industry. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards,
Emission Standards and Engineering Division. Research Triangle
Park, N.C. (Draft Report, never finalized). June 21, 1978.
5. Reference 3.
6. Runion, Howard E. "Benzene in Gasoline." American Industrial Hygiene
Association Journal. May 1975.
7. Scott Environmental Technology. Analysis of Vapor Samples from
Gasoline Storage Tanks, Colonial Pipeline Company, Greensboro.
florth Carolina. Report for U.S. Environmental Protection Agency.
'Office of Air Quality Planning and Standards. November 1977.
8. American Petroleum Institute. Evaporation Loss from Internal
Floating-Roof Tanks. American Petroleum Institute, Washington,
D.C. API Publication 2519, Third Edition. June 1983. p. 18.
9. Beychok, M.R. "Calculate Tank Losses Easier." Hydrocarbon Processing.
March 1983. p. 72.
10. Bulk Gasoline Terminals - Background Information for Proposed
Standards. U.S. Environmental Protection Agency. Research
Triangle Park, N.C. Publication No. EPA-450/3-80-038a.
December 1980. p 6-11.
11. Transportation and Marketing of Petroleum Liquids. In: Compilation
of Air Pollutant Emission Factors. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. July 1979.
12. Pacific Environmental Services, Inc. Study of Gasoline Vapor Emission
Controls at Small Bulk Plants. Report to U.S. Environmental Protection
Agency, Region VIII. EPA Contract No. 68-01-3156, Task No. 5.
October 1976.
13. Reference 12, p. 3-5.
2-20
-------
14. Reference 10, p. 3-15 and 3-17.
15. Reference 10, p. 3-15.
16. Reference 11.
17. Memorandum from Norton, R.L., Pacific Environmental Services, Inc.
to Shedd, S.A., U.S. Environmental Protection Agency. December 20, 1983.
Trip Report to California Air Resources Board.
18. National Petroleum News. 1983 Factbook Issue. Mid-June 1983
Volume 75, No. 7A. p. 103.
2-21
-------
-------
3.0 CONTROL TECHNOLOGY
3.1 INTRODUCTION
This chapter describes available control techniques which can be
used to reduce emissions from the gasoline marketing network. A large
portion of the gasoline marketing industry employs vapor control technology
which has been demonstrated and has been installed and operated at
facilities for many years. The control strategy for storage tanks at
terminals has been to reduce emissions (i.e., by use of submerged fill
and/or floating roofs). The control strategy for truck loading and
unloading areas at bulk terminals, bulk plants, and service stations,
has been to collect and transfer vapors back to the bulk terminal vapor
processor for treatment. Controls for storage tanks, bulk plants, bulk
terminals, and inloading at service stations is commonly referred to as
Stage I. Controlling emissions as a result of vehicle refueling at
service stations has been demonstrated in California and the District
of Columbia by using control technology installed at the service station
(known as Stage II controls). Another alternative for controlling
vehicle refueling emissions, but not practiced, is a control system on
the vehicle (onboard). Since these two vehicle refueling control
systems are a key consideration in this study, the in-use effectiveness
of these two control systems will also be reviewed.
3.2 CONTROL TECHNOLOGY FOR BULK GASOLINE TERMINALS
Emissions from bulk terminals occur during gasoline storage and
transfer operations as shown in Figure 3-1. This section will limit
the discussion to the emssions from transferring gasoline to outgoing
transport trucks. Control of emissions from gasoline storage is discussed
in Section 3.3. Vapor controls have been used at terminals for many
years and the baseline analysis (Appendix B) estimates that approximately
70 percent of the bulk terminals had some type of controls required in
the base year of 1982.
Emissions resulting from outgoing transfer operations 'at terminals
are controlled by two main elements, a vapor processing system (or
3-1
-------
5-
OJ
3
CO
a
-a
ea
o
(O
CD
4-
O
a>
"CL
fO
X
I
oo
01
\
\
\
\
\
\
\
4— i/)
O -^
O E
C£. (O
| —
CD
C O)
•r- OJ
•)-> (0
(0 i-
0 0
u_ oo
o
S-
-------
vapor processor) in conjunction with a vapor collection system. The
vapor collection system consists of all the piping and components
necessary to safely transfer the air-vapor mixture from the loading rack
and tank truck to a vapor processor. A properly designed vapor collection
system at the terminal should not result in excessive backpressure at
the tank truck during loading and should have no vapor leakage during
transfer. Check valves are also commonly installed in the vapor collection
system at the racks so that vapors displaced from loading at one rack
position will not be simply emitted at an adjacent unused rack position.
There are three major types of new generation vapor processors
commonly used at bulk terminals: carbon adsorbers, thermal oxidizers
and refrigeration systems.
The carbon adsorption vapor recovery system uses beds of activated
carbon to remove gasoline vapors from the air-vapor mixture. These
units generally consist of two vertically positioned carbon be'ds and a
carbon regeneration system. During gasoline tank truck loading activity,
one carbon bed is being used for absorption while the other bed is being
regenerated, usually by an air and vacuum purge.
Thermal oxidation units are used to control emissions from bulk
terminals without recovering any gasoline. The gasoline vapor-air
mixture generated from transfer operations at the loading rack can be
piped to either a vapor holder or directly to the oxidizer unit. The
vapor holder stores the air-vapor mixture from the loading rack so that
the system can process gasoline vapors at a relatively constant concen-
tration and flow. Once ignition has been initiated in the thermal
oxidizer, the air-vapor mixture serves as the fuel and the combustion
process continues until all of the vapors have been burned.
Refrigeration type recovery units recover gasoline vapors from the
loading operation in the form of a liquid product. In the refrigeration
system, the air-vapor mixture from the loading racks is routed to a
condensation chamber and passed over a series of cooling coils. Tempera-
tures in the condensation section can be as low as -84°C. The
gasoline vapors condense, with some water vapor in the air, and are
separated in a gasoline/water separator.
These three vapor processing techniques were evaluated recently
by EPA.132 The test data considered in evaluating the three control
3-3
-------
technologies represent terminals ranging in gasoline throughput from
190,000 liters per day (50,000 gal/day) to 5,700,000 liters per day
(1,500,000 gal/day). Sixty-one tests were evaluated, totaling over
100 days of testing. These data are considered representative of the
conditions at a wide range of terminal sizes and indicate that these
three vapor processors can operate at or below 35 mg/liter.3 An emission
factor of 35 mg/liter was used to represent the installation of new vapor
processors in subsequent emission estimation analysis.
Several other technologies exist and have been used for many years
at terminals. These include compression-refrigeration-absorption (CRA),
compression-refrigeration-condensation (CRC), and lean oil absorption (LOA)
These technologies were considered adequate technology to meet the CTG
requirements for bulk terminals and have been shown to reduce emissions
to 80 mg/liter.4 Therefore, 80 mg/liter was used to represent controls
equivalent to those recommended in the bulk terminal CTG.
3.3 CONTROL TECHNOLOGY FOR STORAGE TANKS
Storage tank emissions occur from breathing losses, and from
filling, and emptying losses (working losses). There are three major
types of storage vessels, fixed-roof tanks, internal floating-roof
tanks and external floating-roof tanks. Each tank type has its own
associated emission rate.
3.3.1 Fixed-Roof Tanks
A fixed-roof tank is the minimum acceptable equipment currently
employed for the storage of gasoline. Working losses (filling and
emptying losses) and breathing losses normally incurred from the storage
of gasoline in fixed-roof tanks can be reduced in the following ways:
(1) by the installation of an internal floating roof with rim
seals; or
(2) by the installation and use of a vapor processing system (e.g.,
carbon adsorption, incineration or refrigerated condensation)
or a vapor balance system
Fixed-roof tank emissions are most readily controlled by the
installation of internal floating roofs.
3-4
-------
3.3.2 Internal Floating-Roof Tanks
Internal floating roofs can be used directly as a control device
for existing fixed-roof tanks. An internal floating roof, regardless
of design, reduces the area of exposed liquid surface in the tank.
Reducing the area of exposed liquid surface, in turn, decreases the
evaporative losses. The largest emission reduction is achieved by the
presence of the floating-roof vapor barrier that precludes direct
contact between a large portion of the liquid surface and the atmosphere.
All internal floating roofs share this design benefit. The relative
effectiveness of one internal floating-roof design over another, is a
function of how well the floating roof can be sealed.
From an emissions standpoint, the most basic internal floating-
roof design is the noncontact, bolted, aluminum,, internal floating roof
with a single vapor-mounted wiper seal. The four types of losses
from this roof design are rim or seal losses; fitting losses;-deck
seam losses; and withdrawal losses. Rim or seal losses and fitting
losses constitute the largest percentage contribution to the total loss
from an internal floating roof tank.
Depending on the type of roof and seal system selected, installing an
internal floating roof in a fixed-roof tank will reduce the total emission
by 68.5 to 97.8 percent.5 The currently available emissions test data6
suggest that the location of the seal (i.e., vapor- or liquid-mounted)
and the presence of a secondary seal are the primary factors affecting
the effectiveness of seal systems. A liquid-mounted primary seal has a
lower emissions rate and thus a higher control efficiency than a vapor-
mounted seal. A secondary seal, be it in conjunction with a liquid- or
a vapor-mounted primary seal, provides an additional level of control.
3.3.3 External Floating-Roof Tanks
External floating-roof tanks do not experience the fitting losses
or deck seam losses that occur with most internal floating-roof tanks.
The external floating-roof tanks are constructed almost exclusively of
welded steel. This accounts for the absence of the deck seam losses.
Further, because of the roof design, few if any deck penetrations are
necessary to accommodate fittings.
Rim seal losses and withdrawal losses do occur with external
floating-roof tanks. The only difference between external floating-
3-5
-------
roof tanks and internal floating roofs is that the external floating-
roof seal, losses are believed to be dominated by wind induced mechanisms.'
Withdrawal losses in external floating-roof tanks, as with internal
floating-roof tanks, are entirely a function of the turnover rate and
inherent tank shell characteristics. No control measures have been
identified that are applicable to withdrawal losses from floating-roof
tanks.
Rim seal losses from external floating-roof tanks vary depending
on the type of seal system employed. As with internal floating-roof
rim seal systems, the location of the seal (i.e., vapor- or liquid-
mounted) is the most important factor affecting the effectiveness of
resilient seals for external floating-roof tanks. The relative effec-
tiveness of the various types of seals can be evaluated by analyzing
the seal factors (Ks factor and wind velocity as shown in emission equa-
tions in Appendix B). These seal factors were developed on the basis
of emission tests conducted on a pilot scale tank.8 From such an
analysis it is clear that liquid-mounted seals are more effective than
vapor-mounted seals at reducing rim seal losses. Metallic shoe seals,
which commonly are employed on only external floating-roof tanks, are
more effective than vapor-mounted resilient seals but less effective
than liquid-mounted resilient seals.
3.4 CONTROL TECHNOLOGY FOR BULK GASOLINE PLANTS
Control of gasoline working and breathing losses resulting from
storage and handling of gasoline at bulk plants can be accomplished
through submerged fill and a vapor balance system. The following
sections will describe these two types of control technologies.
3.4.1 Submerged Fill
One method of reducing vapors generated during the loading of
gasoline into tank trucks and storage tanks is by using submerged fill.
Submerged fill is the introduction of liquid gasoline into the tank
being filled with the transfer line outlet being below the liquid
surface. Submerged filling minimizes droplet entrainment, evaporation,
and turbulance. This is compared to splash loading where the transfer
line outlet is at the top of the tank (Figure 3-2a).
3-6
-------
Vapor Emissions
V
Vapors
Gasoline
Vaporsx
x-
X
Fill Pipe
Hatch Cover
Gasoline
Product
Tank Truck Compartment
A. Top Splash Loading Method
Vapor Emissions
v
Product
-*•
co
O
I
Fill Pipe
Hatch Cover
Tank Truck Compartment
B. Top Submerged Method
Vapor Vent to
Recovery Equipment
or to Atmosphere
Vapors
Product
C. Bottom Loading Method
Tank Truck Compartment
Gasoline
"ft"- Fill Pipe
FIGURE 3-2. Gasoline Tank Truck Loading Methods
3-7
-------
Submerged filling of tank trucks at outgoing loading racks can be
either by a submerged fill pipe or bottom loading. In the top submerged
fill pipe method, the fill pipe descends to within 15 centimeters of
the bottom of the tank truck (Figure 3-2b). In the bottom filling
method, the fixed fill pipe enters the tank truck from the bottom
(Figure 3-2c).
3.4.2 Vapor Balance System
The vapor balance system consists of a pipeline between the vapor
spaces of the truck and the storage tanks which essentially creates a
closed system allowing the vapor spaces of the storage tank and the
truck to balance with each other. Figure 3-3 shows the balance system
at a bulk plant. The net effect of the system is to transfer vapor
displaced by liquid in the storage tank into the transport truck during
transfer of gasoline into the storage tank. This prevents the compression
and expansion of vapor spaces which would otherwise occur in a filling
operation. If a system is leak tight, very little or no air is drawn
into the system, and venting, due to compression, is also substantially
reduced. Also vapor balancing of storage tanks and outgoing account
trucks reduces account truck filling losses and virtually eliminates
emptying losses from storage tanks (i.e., displaced vapors are returned
to the storage tank in this closed balance system).
3.4.3 Efficiency of Control Technologies
Submerged filling of tank trucks can reduce vapor loss by 58 percent
when compared to splash loading. This reduction was based upon the use
of an emission factor of 1440 mg/liter to represent splash loading and an
emission factor of 600 mg/liter for submerged loading.^
The balance system has proven to be effective in bulk plant
applications for both the delivery of gasoline by transport truck
to the bulk plant and for loading account trucks. Based upon EPA test
data, controls on bulk plant storage tanks can reduce filling losses
by greater than 95 percent, and draining and tank truck loading losses
by greater than 90 percent.10 An emission factor of 57.5 mg/liter was
used to represent the balance system control technology for tank filling
losses based upon 95 percent control of the uncontrolled emissions
3-8
-------
CD
CO
(O
SI
OJ
c
O
ea
CQ
O
a
03
>
CO
-------
(1150 mg/liter).11 Emission factors for storage tank draining losses
and tank truck loading losses were assumed to be 46 mg/liter and
96 mg/liter respectively, based upon 90 percent control of the
respective uncontrolled emission factors (draining losses - 460 mg/liter,
truck loading losses (balance service) - 960 mg/liter).12,13 Maintaining
the integrity of the storage tanks, tank trucks, and associated vapor
collection systems, and ensuring that proper connections are made, are
necessary for obtaining high efficiencies.
3.5 CONTROL TECHNOLOGY FOR TANK TRUCKS
Just as there are several loading methods and types of rack
equipment at terminals and bulk plants to fill tank trucks with gasoline,
there are several compatible truck loading systems. Gasoline tank
trucks are normally divided into compartments with a hatchway at the
top of each compartment. Top loading can be accomplished by opening
the hatch cover and dispensing product directly through the hatch by
splash or submerged fill. A top loading vapor system, compatible with
the hatch, permits loading through the hatch while vapors are collected.
A better vaportight seal is realized when bottom loading is used. A
1979 surveyl4 covering approximately 1,900 tank vehicles, or about 2
percent of the gasoline tank truck population, indicated that 22.8 percent
of tank trucks have only top loading, while the remaining 77.2 percent
can be either top or bottom loaded. The trend is toward more trucks
using bottom loading, due to State vapor recovery regulations and the
advantages cited earlier.
Tank trucks become a separate source of emissions when leakage
occurs from the truck-mounted vapor collection systems and truck
compartment dome covers. This leakage has been estimated to be as high
as 100 percent of the vapors which should have been captured and to
average 30 percent.15
3.5.1 Description of Control Technologies
Vapor leakage can be minimized by the requirement of all tanks to
pass an annual leak-tight test. To meet these annual requirements,
maintenance of the vapor containing equipment such as the hatch
cover seals and pressure-vacuum vents must be conducted, and repairs
3-10
-------
performed. Figure 3-4 Illustrates the tank truck vapor collection-
equipment. The control techniques guidelines (CTG) for gasoline tank
trucks recommends pressure limits for the annual test on the tanks and
their vapor collection equipment.^ These pressure limits represent
the way in which the P-V vent valves operate on tank trucks. The CTG
recommends that the tank trucks pass an annual certification test which
verifies the vapor tightness of the tank. The monitoring provisions of
the regulations recommended by the CTG permit monitoring as needed
using a portable combustible gas detector.
3.5.2 Effectiveness of Technologies
The effectiveness of vapor control systems at bulk terminals and
bulk plants is dependent upon the absence of leaks in the vapor-containing
equipment on the tank truck. In EPA-sponsored tests, the average
vapor loss due to tank truck leakage was determined to be 30 percent in
areas having no tank truck vapor tightness regulations.17 In June 1978
EPA conducted a series of vapor leak tests on 27 tank trucks that were
required to undergo an annual leak tightness test.18 Tests were
conducted on the tank trucks before any maintenance was performed to
establish the truck leakage rate since the last certification. Evalu-
ation of this data indicated that the average leak rate for those tanks
tested prior to maintenance was approximately 10 percent, meaning that,
on the average, approximately 10 percent of the air-vapor mixture
exhausted from a regulated gasoline tank truck during product loading
would leak to the atmosphere without reaching the vapor processor.19
3.6 CONTROL TECHNOLOGY FOR TRANSFERS INTO SERVICE STATION UNDERGROUND
STORAGE TANKS (STAGE I)
3.6.1. Description of Technology. Emissions from underground
tank filling operations at service stations can be reduced by the use
of a vapor balance system (Stage I control). In the service station
balance system, vapors which would normally be vented to the atmosphere
are routed back to the delivery truck, which unloads gasoline, through
a vapor collection system. The truck transfers the vapors to the
terminal or bulk plant for ultimate treatment by the vapor processors
at the terminal .
3-11
-------
OVERTURN
(VAPOR RETURN)
RAIL
RUBBER BOOT
OR
ETAL COVER
VENT
VALVE
OVERFILL SENSOR
DOME LID SEAL
BASE RING GASKET
TANK SHELL
FIGURE 3-4. Tank Truck Vapor Collection Equipment
For Bottom Loading Operations
3-12
-------
Gasoline is loaded by gravity into the underground storage tanks
via a flexible hose. Liquid gasoline displaces a nearly equal volume
of partially saturated gasoline vapors. The vapor is vented through a
pipe and flexible hose connected to a vapor collection system (i.e., a
manifolded pipe) on the transport truck. Liquid transfer creates a
slight pressure in the storage tanks and a slight vacuum in the truck
compartment. These pressure differences effectively cause the transfer
of displaced vapor to the truck. Because of a phenomenon known as
vapor growth caused by liquid temperature differences, the truck volume
cannot always accommodate all of the vapors. Any excess vapor is
released through the vapor vent line shown in Figure 3-5. This technology
has been demonstrated and installed in service stations across the
country for over 10 years. EPA has also provided design guidance for
Stage I controls as far back as 1975.20
3.6.2 Effectiveness of Technology. The effectiveness of the
Stage I vapor balance system is adversely affected by leaks. Truck
hatches should be closed and hose connections should be tight during
loading. Tests demonstrate balance systems to be greater than 95 percent
efficient for reducing underground storage tank filling losses.21
Note that breathing and emptying losses are not controlled by this
method. These two losses account for 5 percent of total station losses.
In order for the vapor balance system's performance to be maintained
at design efficiency levels, the following objectives must be met:
(a) Assure that the vapor return line will be connected during
tank filling,
(b) Assure that there are no significant leaks in the system or
tank truck which reduce vacuum in the truck or otherwise inhibit vapor
transfer,
(c) Assure that the vapor return line and connectors are of
sufficient size and sufficiently free of restrictions to allow transfer
of vapor to the tank truck and achieve the desired recovery, and
(d) Assure that gasoline is discharged below the gasoline surface
in the storage tanks.
3-13
-------
MANIFOLD FOR RETURNING VAPORS
VAPOR VENT LINE
TRUCKSTQRAG
COMPARTMENTS
INTERLOCKING VALVE
UNDERGROUND
STORAGE TANK
/HI l\m 11 ltt rrrrr
SUBMERGED FILL PIPE =
FIGURE 3-5. Vapor Balance System at a Service Station
3-14
-------
3.7 VEHICLE REFUELING
In addition to service station tank loading losses, vehicle
refueling operations are considered to be a major source of
emissions {see Table 2-4.) Vehicle refueling emissions are attributable
to vapor displaced from the automobile tank by dispensed gasoline and
to spillage. The quantity of displaced vapors is dependent on gasoline
temperature, auto tank size and temperature, fuel level, gasoline RVP,
and dispensing rates.
The two basic refueling vapor control alternatives are: (1) control
systems on automobiles (onboard), and (2) control systems on service
station equipment (Stage II). Onboard controls are basically limited
to two systems - the carbon canister and the collapsible fuel holder.
This study limits its analysis to the carbon canister system. The
collapsible fuel holder or bladder, which attempts to eliminate the vapor
emissions source by eliminating the vapor space in the fuel tank, is a
recently developed system and will not be considered in this document.
3.7.1 Stage 11-Vapor Control Systems
Loading losses due to refueling motor vehicles can be significantly
reduced by Stage II systems. There are currently three types of Stage II
systems in limited use in the United States: the vapor balance, the
hybrid, and the vacuum assist systems. These systems are currently
installed on many of the stations in California and the District of Columbia,
Stage II systems have been required in parts of California since the
early 1970s. The performance of each of these three types of Stage II
systems is discussed below.
3.7.1.1 The Vapor Balance System. The simplest of the three
Stage II systems is the vapor balance system. A generalized schematic
drawing of a balance Stage II system is presented in Figure 3-6. As
gasoline vapor in the automobile fuel tank is displaced by the incoming
liquid gasoline, it is prevented from escaping to the atmosphere at the
fill neck/nozzle interface by a flexible rubber "boot." This boot is fitted
over the standard nozzle and is attached to a hose similar to the liquid
hose. The hose is connected to piping which vents to the underground tank.
An exchange is made--vapor for liquid—as the liquid displaces vapor to
the underground storage tank. The underground storage tank assists
3-15
-------
LL.
UJ
a:
3
j_i
en
OJ
a
C
(U
o
a
-------
this transaction by drawing in a volume of vapor equal to the volume of
liquid removed.
The effectiveness of this system is dependent on a tight seal
between the boot and vehicle fillneck. If a tight seal is not maintained
the collection efficiency of this system is severly impaired. Since a
slight pressure is generally created at the nozzle/fillpipe interface
with balance systems, effective operation requires that a seal be made
at the interface during vehicle refuel ings to minimize vapor leakage to
the atmosphere. The balance system nozzle is equipped with a spring-
loaded bellows which must be compressed before dispensing can occur.
To assure that the bellows is sufficiently compressed, the spout has a
latch band which when properly hooked onto a fillpipe lip, causes an
interlock mechanism to deactivate to allow dispensing, and to open the
vapor passage valve.
3.7.1.2 The Vacuum Assist System. The vacuum assist system
differs from the balance system in that a "blower"—a vacuum pump—is
used to provide an extra pull at the nozzle/fill neck interface
(Figure 3-7). Assist systems can recover vapors effectively without a
tight seal at the nozzle/fillpipe interface because only a close fit is
necessary. A slight vacuum is maintained at the nozzle/fill neck interface
allowing air to be drawn into the system and not allowing the vapors to
escape. Because of this assist, the interface "boot" need not be as
tight fitting as with balance systems. Further, the vast majority of
assist nozzles do not require interlock mechanisms. Assist systems
generally have vapor passage valves located in the vapor passage somewhere
other than in the nozzles, resulting in a nozzle which is less bulky
and cumbersome than nozzles employed by vapor balance systems.
Because of the vacuum, the hydrocarbon/air mixture volume drawn
into the underground storage tank is more than the tank can accommodate.
Consequently, a vacuum assist system results in some venting of excess
vapors to the ambient air from the storage tank, requiring some form of
secondary processing such as adsorption, incineration, or condensation.
Vacuum assist systems typically in operation use incineration for
secondary processing.
3-17
-------
'.O
OJ
0)
-p
00
u
(O
O)
u
O)
Q.
03
OJ
cn
ra
en
uj
3-1-8
-------
3.7.1.3 The Hybrid System. The hybrid system borrows from the
concepts of both the balance and vacuum assist systems. It is designed
to enhance vapor recovery at the nozzle/fill neck interface by vacuum,
while keeping the vacuum low enough so that a minimum level of excess
vapor/air is returned to the underground storage tank (Figure 3-8).
With the hybrid system a small amount of the liquid gasoline (less
than 10 percent) pumped from the storage tank is routed (before metering)
to a restricting nozzle called an aspirator. As the gasoline goes
through this restricting nozzle, a small vacuum is generated. This
vacuum is used to draw vapors into the rubber boot at the interface.
Because the vacuum is so small, very little excess air, if any, is
drawn into the boot, hose and underground storage tank, and thus there
is no need for a secondary processor, such as the vacuum assist's
incinerator.
3.7.2 Onboard Vapor Control Systems
Onboard vapor control systems consist of carbon canisters installed
on the vehicle to control refueling emissions. The carbon canister
system adsorbs, on activated carbon, the vapors which are displaced
from the vehicle fuel tank by the incoming gasoline. A schematic
diagram of a carbon canister system is presented in Figure 3-9. Onboard
control systems have been installed and evaluated on test vehicles.
Appendix C contains an assessment of the carbon canister-type onboard
control system.
Such a system first adsorbs the emissions released during refueling
and subsequently purges these vapors from the carbon to the engine
carburetor when it is operating. This system is essentially an expansion
of the present evaporative emissions control system used in all new
cars to minimize breathing losses from the fuel tank and to control
carburetor evaporative emissions. However, unlike the present system,
a refueling vapor recovery system will require a tight seal at the
nozzle/fill neck interface during refueling operations to ensure vapors
flow into the carbon canister and are not lost to the atmosphere.
As shown in Figure 3-9, the vapors displaced during fuel tank
refueling first flow from the fuel tank to a vapor/liquid separator
where any entrained liquid is separated from the vapor and returned to
the fuel tank. The vapor then flows into a canister filled with
3-19
-------
CM
0)
or
O)
•o
•I—
1-
O
O
OJ
i.
O
Q.
fO
oo
CO
I
co
LU
3-20
-------
ONBOARD VAPOR CONTROL SYSTEM
CARBURETOR
t
CARBON
CANISTER
PURGE VAPOR/LIQUID
CONTROL SEPARATOR
FILL PIPE MODIFICATIONS
TRAP DOOR
SEAL
GUIDE
LEAD NOZZLE RESTRICTOR .
SPOUT
FIGURE 3-9 . Onboard Controls for Vehicle Refueling Emissions,
3-21
-------
activated carbon. The canister size required is estimated to be about
two or three times the size of the present canisters used for evaporative
emissions. The vapor line from the fuel tank to the canister would
need to be larger than that presently used in order to accommodate the
higher vapor flow rate during refueling.22'23
The carbon is regenerated during the driving cycle with air drawn
through the canister to desorb the gasoline vapors. The air/vapor
mixture is sent to the carburetor and burned in the engine. Onboard
control systems require purge control systems which will purge
hydrocarbons without producing significant increases in exhaust or
evaporative emissions.
One issue concerning onboard control systems is the best technique
to assure a tight seal at the nozzle-fill pipe interface. Three concepts
were investigated, a fillpipe modification, a combination nozzle/fill-
pipe modification, and a nozzle modification. Any nozzle modification
would be simpler in design than the nozzle used for service station
control systems (Stage II) because there is no need for a double hose or
vapor passage in the nozzle. In order to simplify emissions and cost
analyses performed in this document, the fillpipe modification approach
was employed as all components of the onboard control system would be
located on the vehicles.
3.7.3 Effectiveness of Technologies
The California Air Resources Board (ARB) performs certification
testing on all Stage II vapor recovery systems installed in the state.
Results of this testing indicate that all of the Stage II vapor recovery
systems discussed are capable of achieving an emission reduction of
95 percent.24'25
In addition to controlling automobile refueling losses, a well
maintained-Stage II system eliminates underground storage tank emptying
losses since gasoline vapors, not fresh air, are drawn into the tank
when liquid gasoline is pumped into the motor vehicle. Because a
secondary processor is employed, the vacuum assist system is also
effective in reducing breathing losses and Stage I inefficiencies.
Onboard vapor control systems are effective at reducing refueling
vapor emissions, but unlike Stage II systems, they provide no control
of underground storage tank breathing, emptying or Stage I inefficiency
3-22
-------
losses. An efficiency of 98 percent has been reported for control
of automobile refueling losses using onboard control systems (see
Appendix C, Section II).
The discussion on onboard controls to this point has only
considered the emission reductions derived from eliminating refueling
losses. However, preliminary data from EPA's emission factors program
indicates that in-use fuels typically have higher volatility than the
fuel specified for certification testing and, therefore, produce
larger amounts of evaporative HC which cannot be adsorbed by the
current charcoal canisters (see Appendix C, Section VI). Preliminary
estimates of the level of these excess evaporative, emissions can be
made using the data currently available from EPA's evaporative emission
• • • . .- .-,-•,•. £.* rr ...
factors testing program which is now in progress. This program
involves evaporative emission testing using Indplene {certification)
and commercial fuel in carbureted and fuel-injected vehicles. - Based
on preliminary data from'this program, it is estimated that light-duty
vehicles (LDVs) have evaporative emissions in the range of 0.23 to
0.44 g/mi using commercial fuel and 0.16 to 0.24 g/mi using certifica-
tion fuel, yielding an excess in the range of 0.07 to 0.20 g/mi. A
best estimate at this time based on this preliminary data is evaporative
emissions of 0.33 g/mi using commercial fuel and 0.20 g/mi using
certification fuel, for an excess of 0.13 g/mi. Although data is not
available for light-duty trucks (LDTs) , one would expect results in
the same ranges since LDV and LOT evaporative control systems are
very similar. Since refueling only occasionally coincides with the
occurrence of evaporative emissions, the larger charcoal canister
associated with an intergrated onboard/evaporative emission control
system could also control these excess evaporative emissions at
little or no extra cost.
3.7.4 In-Use Effectiveness of Control Technologies
It would be unrealistic to assume that Stage II and onboard control
systems would achieve, in normal use, the level of recovery efficiencies
achieved during certification tests, which are performed under idealized
conditions. The reliability of components, routine maintenance, and
public acceptance of the systems will all affect "downtime." An assessment
of actual (i.e., in-the-field) effectiveness is based upon engineering
3-23
-------
considerations respecting the technologies involved, in-field experience
to date with Stage II systems (California and the District of Columbia),
and upon EPA's Mobile Source Enforcement Division's past experience
with the enforcement of mobile source and mobile-source-related programs
(e.g., Fuels, Stage I Vapor Recovery).
Potential failure modes were identified for each type of vapor
recovery system (balance, hybrid, vacuum assist and onboard cases).
These failure modes were grouped under the three general headings of
misinstallation, improper maintenance and tampering (see Appendix D for
explanation of what system defects are covered by each mode).
For Stage II and onboard technologies, steady-state in-use
efficiencies were determijied at enforcement frequencies of quarterly,
annual, bi-annual and minimal enforcement under both State and federal
enforcement scenarios (Table 3-1). As expected, the greater the
enforcement frequency or effort, the closer 'the i'n-use efficiency
approaches the theoretical efficiency. "Minimal enforcement" or
"Voluntary compliance" means, in the case of Stage II programs, the
situation resulting if no resources, State or federal, were allocated
to program enforcement. In the case of onboard programs, the enforcement
effort would consist solely of incorporation of the onboard vapor recovery
function into the ongoing certification and in-use testing programs.
Appendix D discusses the analysis for estimating in-use effectiveness
in more detail.
EPA experience with contractor personnel involved in pollution
control program enforcement at the Federal level indicates that an
inspector, on average, spends about 75 percent of a man-year in the
field doing inspection work. Enforcement officials in California's
South Coast and San Diego areas indicate that their inspectors also
spend about 75 percent of their time in the field. The slightly higher
in-use efficiency observed for Stage II systems under the State enforcement
scenario (shown in Table 3-1) is a result of the assumptions made
regarding the method in which violations are handled. Direct enforcement
is considered more effective in bringing violators into compliance than
the notice of violation (N.O.V.) type enforcement. The analysis
3-24
-------
Table 3-1. IN-USE EFFICIENCIES OF STAGE II
AND ONBOARD TECHNOLOGIES
Program
Option
Theoretical
Efficiency
Frequency of
Enforcement
Steady-State
Average Efficiency
of Installed Systems
State
Federal
ONBOARD
Modified
Fill pipe
99%
92%
STAGE II
Balance
Systems
Hybrid
Systems
Vacuum
Assist
Systems
Weighted3
Average
of All
Systems
95% Quarterly
Annual
Bi-Annual
Minimal
95% Quarterly
Annual
Bi-Annual
Minimal
95% Quarterly
Annual
Bi-Annual
Minimal
95% Quarterly
Annual
Bi-Annual
Minimal
91
88
81
93
90
85
93
87
81
92
88
82
89
86
77
54%
93
88
82
62%
92
84
76
55%
90
86
78
56%b
aDetermined by using an average weighted according to the population of
Stage II systems currently installed (i.e., 80 percent balance, 15 percent
hybrid, 5 percent vacuum assist).
bProgram efficiency reduces to 56 percent when the rate of noncompliance
percent) is considered. Actual average control efficiency of installed
systems is estimated at 70 percent.
(20
recovery
3-25
-------
assumed all N.O.V. type enforcement on the Federal level and half N.O.V
and half direct enforcement on the State level (see Appendix D). The
efficiency of onboard systems is estimated to decrease from 98 percent
to 92 percent with the expected level of tampering (Appendix C, Section
III).
3-26
-------
3.8 REFERENCES
1. Bulk Gasoline Terminals - Background Information for Proposed
Standards. U.S. Environmental Protection Agency. Office of Air
Quality Planning and Standards. Research Triangle Park, N.C.
Publication Number EPA-450/3-80-038a. December 1980.
2. Bulk Gasoline Terminals - Background Information for Promulgated
Standards. U.S. Environmental Protection Agency. Office of
Air Quality Planning and Standards. Research Triangle Park, N.C.
Publication Number EPA-450/3-8Q-038b. August 1983.
3. U.S. Environmental Protection Agency. Federal Register, Vol 45,
Number 244, December 17, 1980. p. 83126-83153.
4. Polglase, W., W. Kelly, and J. Pratapas. Control of Hydrocarbons
from Tank Truck Loading Terminals. U.S. Environmental Protection Agency,
Research Triangle Park, N.C. Publication Number EPA-450/2-77-026.
October 1977.
5. Control of Volatile Organic Compound Emissions from Volatile Organic
Liquid Storage in Floating and Fixed Roof Tanks - Guideline Series.
U.S. Environmental Protection Agency. Office of Air Quality Planning
and Standards. Research Triangle Park, N.C. Draft. August 1983,
pg. 3-4.
6. VOC Emissions from Volatile Organic Liquid Storage Tanks-Background
Information for Proposed Standards. U.S. Environmental Protection
Agency, Research Triangle Park, N.C. Publication No. EPA/450/3-81-003a
June 1983. Appendix C.
7. The American Petroleum Institute (API) Draft Document, Evaporation
Loss from Internal Floating Roof Tanks, API Publication 2519.
Third Edition. 1982.
8. Reference 4, p. 4-18 and Appendix C.
9. Transportation and Marketing of Petroleum Liquids in Compilation of
Air Pollutant Emission Factors. Third Edition. U.S. Environmental
Protection Agency, Research Triangle Park, N.C. April 1977. p. 4.4-7.
10. Pacific Environmental Services, Inc. Compliance Analysis of Small
Bulk Plants. Report to U.S. Environmental Protection Agency,
Region VIII. Denver, Colorado. Contract 68-01-3156, Task 17.
December 1976.
11. Control of Volatile Organic Emissions from Bulk Gasoline Plants.
U.S. Environmental Protection Agency. Research Triangle Park, N.C.
Publication Number EPA-450/2-77-035. December 1977.
12. Reference 11.
3-27
-------
13. Reference 9.
14. Hang, J.C. and R.R. Sakaida. Survey of Gasoline Tank Trucks
and Rail Cars. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Publication Number EPA-450/3-79-004.
March 1979. p. 3-15.
15. Reference 1, p. 3-15.
16. Shedd, S.A. and N.D. Mclaughlin. Control of Volatile Organic
Compound Leaks from Gasoline Tank Trucks and Vapor Collection
Systems. U.S. Environmental Protection Agency. Research Triangle
Park, N.C. Publication Number EPA-450/2-78-051. December 1978.
17. Reference 1, p. 4-2.
18. Scott Environmental Technology. Leak Testing of Gasoline Tank
Trucks. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Contract No. 68-02-2813.
August 1978 (Draft).
19. Norton, R.L. Evaluation of Vapor Leaks and Development of Monitoring
Procedures for Gasoline Tank Trucks and Vapor Piping. U.S.
Environmental Protection Agency. Research Triangle Park, N.C.
Publication Number EPA-450/3-79-018. April 1979. 94 p.
20. Design Criteria for Stage I Vapor Control Systems, Gasoline Service
Stations. U.S. Environmental Protection Agency. Office of Air
Quality Standards and Planning. November 1975.
21. Report to the Legislature on Gasoline Vapor Recovery Systems for
Vehicle Fueling for Service Stations (Sacramento, California:
California Air Resources Board, March 1983). p. 9.
22. Luken, Ralph A. Cost and Cost-Effective Study of Onboard and
Stage II Vapor Recovery Systems. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. August 1978. 39 p.
23. Reineman, Martin E. Recommendation on Feasibility for Onboard
Refueling Loss Control. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. December 1978. 29 p.
24. Reference 14, p. 8.
25. Memorandum from Norton, R.L., Pacific Environmental Services, Inc.
to Shedd, S.A., Environmental Protection Agency. December 20, 1983.
Trip Report to California Air Resources Board.
26. Report of the Stage II Gasoline Vapor Recovery Study Group on the Impact
of the Implementation D.C. Law 3-127, Section 2:2:707(d): Phase I.
Bureau of Air and Water Quality, Department of Environmental Services,
District of Columbia Government. October 20, 1982.
3-28
-------
4.0 MODEL PLANTS AND REGULATORY STRATEGIES
This chapter presents the model plants used in the analysis to
represent the actual facilities in the estimation of the impacts, the
projections used to estimate gasoline consumption, facilities, and
vehicles, and the regulatory strategies examined. Section 4.1 presents
the model plants for bulk terminals, bulk plants, service stations and
associated facilities. Section 4.2 outlines the projections used for
the analysis. Section 4.3 discusses the regulatory strategies and
their component control options for each source category.
4.1 MODEL PLANTS
4.1.1 Bulk Terminal Model Plants
This section defines the methodology used to select model plant
parameters to represent the range of new and existing bulk gasoline
terminals and provide a basis for comparison of environmental and
economic impacts of proposed regulatory strategies. A bulk gasoline
terminal is typically any wholesale marketing facility that receives
gasoline from refineries by pipeline, ship, or barge, stores it in
aboveground tanks, and dispenses i^t into tank trucks for delivery to
customers.1 The data base for determination of the model plant
parameters was derived primarily from operating data on 40 terminals of
various ages. Data presented in reports of EPA-sponsored terminal
source tests, data from plant visits, and data from information requests
submitted under authority of Section 114 of the Clean Air Act were used
as further input for the selection of model plant parameters.2
Since terminal gasoline throughputs are distributed over a wide
range, several model plant sizes (given on Table 4-1) were considered
in order to best represent the industry. A recent EPA-sponsored report2
discussed the distribution of gasoline terminals by throughput within the
industry:
1. Almost half of the existing terminals (48 percent) are less
than 757,000 liters per day so a model plant size of 380,000
liters per day was selected to represent the subset.
4-1
-------
i
i
i
i
ii
it
ik
II
•i:
&s
j 1 1 p
[— l:
DC|
Ou|
3 »
•M* II
If
} 1
*** i
§
S* P
**.
1 H
UJ*
^_m^
""""^ If
H -
s^ **
-J g
co 1
N
. M
1 1
*^* N
LU ''
•~ n
S? ^s
£E »
il
^
H
M
»
X
'•
f" o
li j.^
;! <->
« ij-
: o
I*
»»'
t^
lui
il
O
o
o
o
0
o
T-l
0
0
o
0
o
in
o
o
o
cT
LO
CM
0
o
c_^
o
o
-^
03
"O
f
cu
CL
c
o
t £1
S ^
i .
* J_s
Throughpul
«'
O *d-
o
0
o
o
CO
CO
O CO
0
o
o"
g
n
O CO
o
o
o
LO
o
O CM
o
<_^
o
oo
co
^
•ra to
c
t. o
(U •!""
CL 4-*
»f—
t/1 t/>
S- O
g ^
jQ O
oo a.
o
~
o
T3 -P
CU -P
cn o
S- CQ
CT> 3
^•^
•a 4->
O> -P
O) O
^O t- CQ
CD
1°
to CL
O
u—
i^
in
j-
cn
•5 -a
o -g
CU
*[ ^^
o
i_ =
CU •!-
.a T3
£E 'w
3 O
:z — J
T3 CU
i— D.
CD
O
r- o
CM O
CM —
0.1=
r— O.
cn
o
r-- o
CM O
CM -— •
HH
r— Q.
O
P**" ^D
CM O
•» ^o
CM ^—
E <•—
O.E
i— CL
CT1
0
r^ o
CM O
CM •—
£=
S_
cn
•1—
T3
03
O
_J
CO
03
C£
cn
Q.
E
a.
'•±-
&. e:
cu o
*~ cn
0
0 0
CM O
"LO
CM «
CO CO
to to
cu o
•p r—
•B— »—
r~ cn
o
0 0
CM O
CM "
CO OO
S- C
cu o
•p r—
r^ "tO
• cn
o
0 0
CM O
"LO
CM "
CO CO
*•"•*
to to
S- C
OJ o
•r- r—
t— to
cn
O
0 0
CM O
OJ "
CO CO
— •
.JJ
o
to
Q.
o
o
3
I—
c
1—
cn
to
as
•P
3
•r*
E
O
CM
O)
•P
3
'E
o
CM
to
CU
•P
3
•i—
E
O
CM
to
CU
•p
3
'i
O
CM
CU
E
£
cn
c
•o
03
O
CJ
3
S-
1—
c:
03
^
E E
CL CL
r- cn
0 0
0 0
CO CM
n *\
CM —•
s"s
CL CL
i— cn
0 0
0 0
•3- «*
O LO
*f~*
E E
CL CL
r— CD
0 0
0 0
«\ «
O LO
CM —
«•»•••»
E E
Cu CL
i— cn
0 0
0 0
CO O
fO co
I— 1 •—»
to
3
o
cu
c.
03
•P
C
03
to cn
1 1 •!—
-o
E 03
3 O
X
03
s:
•r-
5_
03
CU
to
-o
O
CO
cu
to
•a
o
co
i-
(O
cu
>^
to
•a
o
^st°
CO
S-
03
CU
>^
*^^.
to
•o
0
^"
CO
cu
1—
3
-o
cu
3=.
CJ
CO
en
c:
•p
03
O)
CL
O
•i-2
r— °
C*O jD
O
O O
0 0^
"O
LO O
co _o
o
o o
r- o
CO **
co r~-
«* CM
P
S
co _a
E
O
o o
oo o
CO "
"O
CO LO
CM.t— H
^— ••
.^•^
r—
co .a
E J2
CD O
«* o
CO O
« n
O LO
r-< CO
^^^*
•P^
•p-
CJ
(O
a.
03
o
cu
cn
to
i.
o
oo
a>
c
"o
to
03
CJ3
^
U3
^
co
cu
c
•r-
*O
to
03
CIS
0
M-
to
-^
03
(—
4-
0
S-
cu
1
z
,_
o
CM
CT)
-
-
co
to
CJ
i-
H-
TD
O)
C
3
O
r~~
03
C
•i^
E
cu
1—
tt-
o
S-
cu
3
Z
E
"*
r— 1
CM
1 —
CM
CO
to
cu
•r~
-P
•r—
1^
o
03
U_
1o
O
1—
4-
Percent o
c
o
CO
LO
t— <
CO
LO
o
0
CM
r—
cu
o to
2H t—
s_ e:
a> T-
CL E
to o
CU -P
•f^«
r^ -P
•t- C3
CJ -P
03
U- 0
O
4- LO
O r-H
cu -P
.a £=
E o3
3 r—
Z 0-
o
LO
t— 1
*
o
CO
CM
CO
i— 1
{i ;! ^ii i '
r_
to
•P^
(E
4-
O
•P
d
CU
o
S-
cu
a.
O)
•P -P
Approxima
Throughpu
CL
1
1
|
1
1!
H
i
\\
1
1
i
I
1
|l
1
t
||
II
li
ii
il
1!
11
1
li
• H
n
ii
i
1!
i
1
i
ii
i
i
i
4-2
-------
2. Approximately 27 percent of the gasoline terminals have a
throughput between 757,000 and 1,514,000 liters/day. A model
plant with a throughput of 950,000 liters/day was selected to
represent these plants.
3. An additional 21 percent of the terminals have a throughput
between 1,514,000 and 2,270,000 liters/day. A model plant size
of 1,900,000 liters/day was selected to represent these facilities,
4. A model plant size of 3,800,000 liters/day was selected to
represent the 4 percent of existing terminals greater than
2,270,000 liters/day
4.1.2 Storage Tank Model Plant
As disussed in Section 2.2.2, a typical terminal has four or five
above-ground storage tanks for gasoline, each with a capacity ranging
from 1,500 to 15,000 m3 (9,400 to 94,000 barrels). Most tanks in
gasoline service have a floating roof to prevent the loss of product
from breathing and working losses. The fixed-roof tank is the
least expensive to construct and is generally considered as the minimum
acceptable tank for the storage of petroleum products. Emissions from
existing fixed-roof tank are most readily controlled by the installation
of an internal floating roof. A set of model plant parameters were
developed to describe the physical characteristics of a typical fixed-roof
tank at a bulk terminal. This typical storage tank has a volume of
2,680 m3 (16,750 bbl), based on available EPA data on fixed-roof tanks
at terminals. A diameter of 15.2 meters (50 feet), and a height of
14.6 meters (48 feet), were assumed as typical values for a tank of this
capacity.3 These parameters were then used in emission equations to
estimate emission rates from fixed-roof tanks (see Appendix B) and to
calculate capital costs in Section 7.2.2.
4.1.3 Bulk Plant Model Plants
As described in Section 2.2.3, bulk gasoline plants are secondary
distribution facilities within the gasoline marketing network. The
typical bulk plant facilities include tanks for storage of gasoline;
loading racks; and incoming and outgoing tank trucks. Regardless of
throughput, typical bulk plants have the same numbers of tanks,
loading racks, and account trucks, as follows.4 The typical bulk
4-3
-------
plant utilizes two relatively small above-ground storage tanks ranging
in capacity between 50,000 to 75,000 liters for gasoline storage.
Usually a plant will have one loading rack using top filling by either
a top-splash method or a top-submerged fill pipe. Transport trucks
supply bulk plants with gasoline while account trucks deliver gasoline
to bulk plant customers. Bulk plants typically average two account
trucks each.
Bulk plant operations are based upon a large number of annual tank
turnovers, with the result that most facilities tend to be small. An
EPA-sponsored report 4>5 discusses the distribution of bulk gasoline
plants by throughput within the industry. Over 90 percent of all bulk
plants have an average daily gasoline throughput that is less than
30,280 liters per day (8,000 gal/day). Of these almost half (42 percent)
are less than 15,140 liters per day (4,000 gal/day) so a model plant
size of 11,350 liters per day (3,000 gal/day) was selected to-represent
this subset. Another 50 percent of the bulk plants have a throughput
between 15,140 and 30,280 liters per day (4,000 and 8,000 gal/day). A
model plant with a throughput of 24,600 liters per day (6,500 gal/day)
was selected to represent these plants. Only 7 percent of bulk plants
have a gasoline throughput between 30,280 and 64,350 liters per day
(8,000 and 17,000 gal/day). A model plant with a throughput of 47,300
liters per day (12,500 gal/day) was selected to represent these plants.
An additional 1 percent of plants have throughputs between 64,350 and
75,700 liters per day (17,000 and 20,000 gal/day). A model plant size
of 64,350 liters per day (17,000 gal/day) was selected "to represent
these facilities. A summary of the bulk plant model plants is shown in
Table 4-2.
4.1.4 For-Hire Tank Truck Population
The trucking industry generally consists of two major groups, for-
hire and private. Private carriers are firms which transport their own
goods in their own trucks. Examples of private carriers are the oil
companies which use their own tank trucks to deliver gasoline from
their terminals or bulk plants. For-hire carriers transport freight
which belongs to others, renting out the hauling services of their
trucks.
4-4
-------
41 4->
4J O 4J
IC1
x o o>
^1
41
a.
OO
UJ
C£
a.
•=C
Q-
o
CQ
CM
I
CQ
o
O tn
*>
+•» C
c a
41i—
u o.
41
a.
IS
a.
41
1
•o
4)
41 0)
C in
T- 4)
'o o. >>
in 41 IB
O>f—
C 3
IQ O
CO
CM
OO
in
o
o
ro
§
O
CO
CVJ
g 8
»T O ^
I-H ro 10
- i i
in o O
^H *f oo
V i-< CM
in o
5-
41 3 _
C O. 01
in
IO — 3
r- 4) "O O
O. C 41
41 O 41
*O t/1 g O)
g O) 3 i-
«> t.
4J -O 41
C = V. >
UJ 3 O -^
r- O
.^ 41 fO "
"3 o "o. <
4J 41 O +J
en r— O) C
41 =3
N O i- O
•>- o
-------
It is estimated that 85,000 tank trucks are used for the delivery
of gasoline.6 About 31 percent of the gasoline tank trucks or 26,300
vehicles are used at bulk terminals. This total includes only tank
trucks of greater than 15,100 liter (4,000 gallon) capacity in order to
avoid the inclusion of small tank trucks operating from bulk plants.
The remainder, or 58,700 vehicles, are smaller tank trucks used pri-
marily to transport gasoline from bulk plants.
For purposes of determining the number of uncontrolled tank trucks
at terminals it was assumed that if a terminal is uncontrolled, the tank
trucks loading at these terminals are also uncontrolled. Since it was
estimated that 33 percent of terminals are uncontrolled, then 33
percent of 26,300 trucks results in 8,700 uncontrolled tank trucks.
Given that 500 terminals would be regulated, then the number of
uncontrolled, terminal-owned tank trucks was calculated using the
method shown in Table 4-3. Of the total 8,700 uncontrolled tank trucks
at terminals, it was estimated that 2,900 tank trucks are terminal-owned
and the remaining 5,800 tank trucks are owned by for-hire tank truck
companies.
Table 4-3. METHOD OF CALCULATING THE NUMBER OF
UNCONTROLLED TERMINAL-OWNED TRUCKS
No. of Affected No. of Trucks owned
Terminals Percent of Terminals in ,by terminals in
Model Plant 1 thru 4a Model Plant 1 thru 4a
500 x 48
500 x 27
500 x 21
500 x 4
Total uncontrol 1 ed
x 3
x 6
x 9
x 20
terminal -owned tank trucks =
or =
Total
Trucks
720
810
945
400
2,875
2,900
aSee Table 4-1 (from Reference 2).
4-6
-------
At bulk plants It was estimated that 45 percent, or 26,400, of the
58,700 tank trucks will be controlled by a vapor balance system. [The
45 percent was obtained by dividing the number of controlled bulk plants
(7,000) estimated at baseline (Table 4-7) by the total number of bulk
plants (15,000)]. Thus, the number of tank trucks without provisions
for vapor balance was estimated as 32,300 vehicles. About 8,000 bulk
plants were assumed to be uncontrolled (see Tables 4-7 and 4-10). The
typical number of tank trucks owned by a bulk plant (determined from an
earlier study6) is two vehicles. Thus about 16,000 tank trucks (the
product of 8,000 x 2 account trucks) are owned by bulk plants and the
remaining 16,300 tank trucks are owned by for-hire tank truck companies.
All cost estimates for the for-hire tank trucks in Chapter 7.0
were calculated based on these vehicle population figures rather than
on a model plant approach. The costs for the private, facility-owned
trucks are already included in the cost estimates for bulk plants and
termi nal s.
4.1.5 Service Station Model Plants
Service stations, as defined in this document, include all motor
vehicle refueling operations that receive revenue from sales of gasoline
(public retail outlets) and that service governmental, commercial, and
industrial fleet operations (private outlets). For this report the
category does not include agricultural outlets. Miscellaneous retail
outlets that were considered service stations for this study include
convenience stores, mass merchandisers, marinas, parking garages and
others which obtain less than 50 percent of revenue from gasoline
sales. In contrast, the U.S. Census Bureau counts as service stations
only those outlets that do 50 percent or more of their dollar business
in petroleum products.
In the development of a representative estimate of the total
service station population in the United States, Census Bureau data and
the Lundberg Survey served as primary references. An accurate count is
difficult because convenience stores (C-stores), and others who sell
motor fuel do not necessarily fit the "service station" definition used
by the Census Bureau.
The Lundberg Survey, based on refiner and marketer reports as
well as grass roots data, was discussed in the September 1983 issue of
National Petroleum News7 (NPN) for 1977, 1980 and 1982. The Lundberg
4-7
-------
1977 total is 263,348 - 50 percent more than the Census Bureau figure
of 176,465.7 For 1982, the Lundberg total is 210,875, compared to the
Census Bureau's 144,690.10 Based on the Census Bureau definition of
service station and Lundberg's emphasis on recording all public outlets
which sell gasoline, an approximate ratio of the two figures, or 2/3 was
determined as the fraction of total public outlets thought to be defined
as marketers of gasoline that would be counted as service stations by
the Census Bureau. This fraction was used in the percentage determination
of outlets estimated to be independent marketers of gasoline under
Section 324 of the Clean Air Act (see discussion on page 4-12).
Estimates of 1982 service station population are presented in
Table 4-4. The number of public outlets was based on the Lundberg
estimate. Surveys confirm that the spread of C-store/gasoline combina-
tions will continue at the expense of the traditional service station
population9 and the Lundberg estimate more accurately reflects- the
number of C-stores, mass merchandisers and others which obtain less
than 50 percent of revenue from gasoline sales.7 In addition to
"public" outlets, there are a significant number of "private" facilities.
These outlets are maintained by governmental, commercial, and industrial
consumers for their own fleet operations. Government agencies with
central garages are typically regional locations for the postal service,
Federal government agencies, and state and county agencies. Other
miscellaneous facilities include utility companies, taxi fleets, rental
car fleets, school buses and corporate fleets. The agricultural sector
of private outlets, including farms, nurseries and landscaping firms,
was not considered. In general, agricultural outlets would have
throughputs of less than 37,850 liters per month (10,000 gal/mo) and,
collectively, a gasoline throughput representing approximately 3 percent
of the nationwide total.8
Service station model plant categories and the approximate distribution
of facilities within each model plant category were derived from size
ranges used by the Bureau of Census, total facilities reported for
19777 and 198210 and the total consumption of gasoline (excluding
agricultural) for each year.11 The total population of retail outlets
in the U.S. continues to decline, and as many of these stations disappear
they are replaced by units which pump more gasoline. The number of
4-8
-------
Table 4-4. ESTIMATES OF 1982 SERVICE STATION POPULATION3
Public Outlets (i.e, service stations and
convenience stores)
"Private" Outlets0
Government (Federal, military, State, local)
Miscellaneous (auto rental, utilities, others)
Trucking and Local Service
Taxi s
School Buses
Total
210,875b
85,450
94,530
21,900
5,380
3,070
421,125
Not including the about 2.5 million agricultural outlets.
Source: Lundberg Estimates. National Petroleum News,
September 1983 (Reference 7).
'Source: Arthur D. Little, inc. The Economic Impact of Vapor
Recovery Regulations on the Service Station Industry^
(Reference 8).
4-9
-------
service station closures which occurred from 1977 to 1982 was assumed
to be distributed evenly across all model plant sizes. Service station
openings were assumed to occur primarily within the two largest model
plant sizes. Data on the number of facilities and gasoline consumption
reported for 1977 and 1982 established an average annual throughput per
station for those years. As expected, the average annual throughput
for all stations increased during this five-year period. Therefore,
the Bureau of Census size distribution available for 1977 was adjusted
to reflect the increase in average annual throughput.
The five model plants and the estimated size distribution of model
plants within each of the five categories are presented in Table 4-5.
The number of nozzles per station representative of each category, as
well as corresponding average throughput values and throughput ranges,
were determined. As a check, the total number of facilities per category
were multiplied by the average monthly throughput value determined as
representative of each category and then by 12 months per year. The
total nationwide annual throughput calculated in this manner was within
3 percent of the throughput figure used as baseline.
Model plants 1 and 2 were originally one category, however, a fifth
model plant representing a throughput range of 0-37,850 1/mo (0-10,000
gal/mo.) was included (based on a graph of the size distribution for
the other four model plants) in order that the effect of a 37,850 1/mo
(10,000 gal/mo.) exemption could be determined. Public and private
facilities are distributed among the five model plant categories of
Table 4-5. The distribution of "public" facilities across the first
three model plant sizes is approximately uniform (i.e., 25 to 30 per-
cent) declining to 14 percent and 3.5 percent of facilities for model
plants 4 and 5, respectively. Based on information from Arthur D.
Little, Inc., and the U.S. Census Bureau, it was estimated that
approximately 90 percent of "private" outlets have throughputs of less
than 37,850 1/mo (10,000 gal/mo.).12»13 The remaining 10 percent of
private facilities were distributed among model plants 2 through 5 in
proportions representative of the public service station distribution.
Additional information within Table 4-5 includes percent of total
4-10
-------
o
1— H
i |
1 "'
03
2
I—
OO
3
LU
£!
00
•^
*-—
j
tA
19
0
LO
1
1
,j.
O i
1— H 1
1— 1
<
i— '
j
LU II
CJ 1
> 1
LU
CM
03
cn
Q
LU
g
•™^
1 —
OO
LU
•
LT>
|
"*
LU
— 1
CO
1—
CM
r- (
m
O
z
c
CO
a.
•o
o
i
O LO
o cc
^•*
LO LO
«»• LO
CM ^.*
CM LO
cn cn
^H*-»
LO O
r~ ""*
g?
CO
CC
•W
2
a.
f
cn
s.
.c
1—
1
1
VO
1
CM
•— 1
CM
i— 1
1
cn
o>
i
to
i
CM
CU
Is4
M
O
z
g°
cn o
o"2
f— 4
CM
«s- cn
*
^
§2
LO
Ol
CM
O LO
O •
cn -o
"CM
LO
LO
0 0
o cn
cn
en
O CM
CO
Id
o'2
«— t
CM
g^o
,— i o
t— 4
0 °*
O CM
**•
^D Lrt
^* CO
^
0 -H
o •
to «^
co"
0 0
^J Ol
A
f" cn
LO
CO
»— 1
1
1
c
0
•f—
CO
oo
CU
o
2 1
CU •>-> +J
CO CO C
tw
O
S.
CU
= "3 u
V)
c
o
13
(13
U
> c
S- 0
CU 1-
CO •*-* +J
= f— 0»
CU 3 0
CJ O. S- -J-> 0- e_
••- 0 CU
r- a. a.
-9 -9
I -^
IQ O CU
> 0. Q. ]
a.
g °
CM O
"O
t— f T-*
CM
0 CM
LO
CO
O CO
o
LO
cn
cn
O LO
cn
cn
§i —
»-*
cn
A
r-H
O CO
O LO
O
^f.
^p
CM
^
O
O
rH
r-4
CO
f^
LO
**:
if*
CM
.
•*
i
CM
2
! §
a.
1 U) i
C 1
"J i o
• c 1 ^
i -2 ^ ! -g
•?- o
(J r— CJ
Ll_ Q. &_ '
2 °" °~
h-
o. i
C- Ij j
O) o 1
o »*
h-
4-n
-------
facilities (i.e., public and private) per model plant and the percent
of total gasoline throughput generated per model plant. Several
proposed service station Stage II regulatory strategies include options
with and without site exemptions or cutoffs (the terms size exemption
and size cutoff are used interchangeably in this document). Examined
size exemptions include any service stations generating less than
37,850 liters/month (10,000 gal/mo) throughput (based on existing State
and local regulations and expected relative economic impacts) and
independent dealers below 189,300 liter/month (50,000 gal/mo.) sales
volume (in accordance with Section 324 of the Clean Air Act and assuming
that Section 324 applies to Section 112 regulations). These stations,
for which size exemptions are examined, represent the segment of the
retail industry that may be most vulnerable to changes in marketing
economics as well as external costs such as vapor recovery costs.
Service stations were classified into several operational- groups^
and those stations which, by definition, qualified as independents were
grouped into a total percentage by model plant. This percentage was
then reduced by one-third as only two-thirds of the total "public"
outlets are estimated to fall within the definition of marketers of
gasoline, according to the Census Bureau and Clean Air Act definitions
(see discussion on page 4-8). Independents as a percentage of total
"public" facilities were estimated at 18 percent, 31 percent, and 45
percent for model plants 1 through 3. It was not necessary to calculate
the percentage of independents within model plants 4 and 5, as the
proposed size exemption applies only to facilities of less than 189,300
liters/month (50,000 gal/mo) throughput. These percentages were used
together with the number of public facilities and the annual throughput
representative of each model plant to determine the throughput attribu-
table to independents and thus, excluded from the impacts analysis.
Independents with throughputs less than 189,300 liters/month (50,000
gal/mo) were estimated to pump approximately 15 percent of the total
gasoline throughput (10 percent in model plant 3, 5 percent in model
plant 2 and less than one percent in model plant 1). If the size
exemption for all facilities less than 37,850 liters/month (10,000
gal/mo), representing 14.2 percent of total consumption (Table 4-5),
4-12
-------
were considered together with the size exemption for independents,
total throughput excluded from regulation would represent approximately
29 percent of nationwide consumption.
4.2 GASOLINE, FACILITY, AND VEHICLE PROJECTIONS
Because the comparison of regulatory strategies, which will be
discussed in Section 4.3, requires the evaluation of emissions,
cost, and risk impacts far into the future, projections had to'be made
for gasoline throughput, gasoline marketing facilities, and automobile
fleet characteristics. This analysis assumed that onboard controls
would be installed only on new light-duty vehicles (passenger cars) and
light-duty trucks (<8500 Ibs. i.e., vans, pickup trucks, etc.). Thus,
the effectiveness of onboard controls increases with time. For this
reason, the analysis was extended through the year 2020 so that
essentially the entire light-duty vehicle fleet would be controlled.
In this way, onboard could be compared with Stage II in both the short
and long term. Therefore, to fully analyze the regulatory strategies,
estimates for the base year of 1982 and projections to the year 2020 '
were performed.
4.2.1 Gasoline Consumption
Total consumption of gasoline in the base year of 1982 was
389 billion liters (102.7 billion gallons).^ F1gure 4.! presents
estimates of how this consumption or throughput was distributed throughout
the gasoline marketing chain.
Several sources were contacted to determine the extent of data
available on gasoline consumption projections. An EPA Federal Register
notice (47 FR 49329)16 dealing with phasing down the lead contemTi^
gasoline contained projections of total gasoline consumption and the
decrease in leaded gasoline usage through the year 1990. The Department
of Energy was contacted and both short-term (through 1985) and mid-term
(through 1990) projections were obtained.17 However, no long-term
projections were available. In Figure 4-2, the available consumption
projections were plotted. In addition, actual consumption for
1976-1982 was plotted from data available in several issues of the
National Petroleum News Factbook.H The original analyses were to be
4-13
-------
c
o
O
UJ
o
00
00
>-
00
S2
O)
O
CO
c
o
_J 00
u. *-« c
O T-
O
UJ UJ •—•
»— -z. o-
»— oo
OO rf
UJ c3
CO
o
* 00 i—
UJ
4-14
-------
Footnotes for Gasoline Distribution System
Figure 4-1
1982 estimates.
Obtained from National Emissions Data System (NEDS), 1982 calendar year.
•»
'A number of bulk plants in Southeastern Alaska and some islands off
Puget Sound receive gasoline reportedly by barge from marine terminals.
At this time, data have not been received to support estimates of
either facility population or total throughput for these bulk plants.
Weighted average based on 1977 Bureau of Census data was determined to
be 25 percent to bulk plants, 75 percent to service stations.
a
"Based on U.S. Department of Agriculture data, Arthur D. Little estimates
that 3 percent of the total U.S. gasoline volume is consumed by the
agricultural sector (Reference 8).
4-15
-------
'f—s
u
5
p
O)
!J
C
o — J
o
a.
+J
Q.
O
o
OJ
c:
o
CO
fd
CD
C\J
<3-
O)
CD
-s
en
o
O
CO
O
OJ
I I
i I
I f I
CM O
o in
o
CM
•s-
o o
en -3
to n
o
CM
o
<3-
CVJ
uoLq.duinsuoo
4-16
-------
based on projections to the year 2000, so a graphical estimation
based on EPA projections was made. The EPA projections were used since
they were in the mid-range of DOE projections. It is recognized that
there is a degree of uncertainty within the assumptions and estimations
employed to project gasoline consumption far into the future. For this
reason and because no additional information was available, projected
consumption of gasoline was assumed to be constant from the year 2000
to the year 2020. The graphical extrapolation that was used as a best
estimate is depicted in Figure 4-2 as a dashed line.
A similar approach was used to project leaded gasoline consumption.
This was necessary to estimate projected EDB and EDC emissions. The
rapid decrease in leaded gasoline corresponds,to. the phase out of
leaded gasoline production. Again, projections were assumed constant
from the year 2000 to 2020. The estimated leaded gasoline con-sumption
is also shown on Figure 4-2 as a dashed line.
After these projections were developed and used as the basis for
the analysis, three other projections were obtained and evaluated,
one by the Department of Energy (DOE), one by the American Petroleum
Institute (API), and one in the Oil & Gas Journal. Table 4-6 compares
the three projections at selected years. Because of the uncertainties
inherent in all such projections, the original assumptions and
extrapolations were retained.
4.2.2 Gasoline Marketing Facilities
Data on the number of gasoline marketing facilities are necessary
since nationwide cost information is generated from costs per facility.
Several information sources were used to estimate the number of facilities
in the base year of 1982. Table 4-7 summarizes the number of facilities
in each industry sector. To determine the number of facilities in
controlled areas at baseline, the throughputs for all the controlled
areas were summed from the tables in Appendix B and the percentage of
total nationwide consumption was determined. This percentage was then
applied to the total number of facilities. For example, 67 percent of
the throughput for terminals was determined to be in areas that were
already controlled. Applying this percentage to the total estimated
number of terminals (1,500) resulted in 1,000 controlled terminals.
4-17
-------
TABLE 4-6. ALTERNATE GASOLINE
CONSUMPTION PROJECTIONS (billion liters [gallons])
:
>
:
I
I
«
e
R
!
i
i
C
•
fc
r
«
Ik
%
»
t>
f
t
I
*
•
p
(
i
i
*
,;,
V
f
f'
I"
»'
V
»>
»
J
y
*
»
f
*>
#
r
Tnt»i (Unleaded and Leaded) Gasoline . ,. _. , ,__ Leaded Gasoline
11
>> c
S-2
.= **
si
VI
JO
11
o. «
eu
*t
•oto
IA!C
vt •*-•
•C
|
St.
CD */* ^™
«^
a^
IS
01 CJ
C± VI
o co r^ to m
t
§^MCT\r— mroeivi«-**-*oooooo ,
i
to o" I •
• • I
Ch O i
co co
2 ., - 1 , .
R S i ^
Is
I^.H.^.it. U- >*• U— **- l^tfc-»*-U- ^H
1—1 *~1 r^ *"""' f~1 *~1 ^ " ^ ^ ^! " ^ ^ ^ 'o^
r^T i o" icuimt^-i^Ti«— '^ w °? Ci^ « .^^ . ^— • — • — • «-» • ••— • • ^ • ^— • — • • S ^
OJ Cft
t a: c^
J trt OJ
*-4 r*» cn «-H ^-« in O
^ U3 r* o ***
co co co r»* i^. £
.is-sss si
j « n rt tn CM >, -
1 : , ", : : S B
j r-i 1 -
?"eou>~i-.e3C3e9oe30CSC3Cse; "^S1
:!«"«o^»^o"^^-^^o s1!
JIPJ^^SSSSSSSSSS^CM • 3 -g
I , -D U_
3
CFrom Personal Communication with Debra Paxton, Department of Energy (DOE), September 30, 1983 (Reference 17).
dFrom "Motor-gasoline forecasting: an embryo science," OJJ «_Gas_ Jour_nal_, November 14, 1983 (Reference 18).
6From Department of Energy (DOE/PE/70045-1), Appendix B, February 1983 (Reference 19).
!£ S
*«— O
s-
•§
4-18
-------
Table 4-7. ESTIMATED NUMBER OF FACILITIES
IN THE BASE YEAR 1982
Facility Type
Bulk Terminals
Bulk Plants
Service Stations
Total
l,500a
15,000a
421,100b
Controlled At
Baseline
1,000
7,000
Uncontrolled
Baseline
500
8,000
At
- Underground 223,200 197,900
storage tanks
- Automobile 37,900 383,200
refueli ng
Storage Tanks 5S470C 4,840d 630e
For-Hire Tank Trucks
- Terminals 5,800^
- Bulk plants 16,300f
Based on 1983 NPN Factbook estimates (Reference 15).
b
See Table 4-4.
c
References 22 and 23.
Controlled by floating roof with primary or secondary seals (see Appendix B).
e
Fixed-roof tanks.
Independent trucks only. Other uncontrolled trucks at bulk terminals (2,870)
and bulk plants (16,000) included in bulk terminal and bulk plant costs.
4-19
-------
The remainder (500 terminals) were assumed to be uncontrolled (i.e. no
vapor processors).
Attempts were made to project the number of gasoline marketing
facilities into the future. Several references have documented the
fact that the bulk plant and service station populations are declining.11
However, no quantifiable data were available to project facility popul-
ation into the future. For this reason, no projections for facilities
were attempted. Instead, for costing purposes, the number of facilities
was kept constant through the year 2020 while the throughputs and cor-
responding recovery credits were decreased proportional to the decrease
in gasoline consumption. This probably biased the cost estimates for
service stations and bulk plants to the high side, but the degree of
the bias is unknown.
4.2.3 Light Duty Vehicles and Light Duty Trucks
A projection of the number of new light duty vehicles (LDV) and
light duty gasoline trucks (LDGT) was necessary because one of the
control technologies evaluated for automobile refueling would require
control equipment installed on all new vehicles after a certain model
year. Estimating costs for this approach required knowledge of the
number of controlled vehicles in each year. Because of the length of
the projections, the analysis had to take into account the scrappage or
retirement rates for controlled vehicles as they age. With the projec-
tions of new vehicles and the projections of retirement or scrappage
rates, the number of controlled vehicles in each year up to 2020 could
be estimated.
The projected number of new LDVs and LDGTs, retirement or scrappage
rates, fuel economies and mileage accrual rates are all provided within
Appendix C. Given the fuel economy and mileage accrual rates, the con-
sumption of gasoline by controlled vehicles in each year could be
calculated (see Appendix C, Section VIII). The gasoline consumption by
controlled vehicles (with and without tampering) is presented in Table
4-8. Emission reduction estimates attributable to onboard controls were
based upon the percent of the total gasoline consumed by controlled
vehicles. Figure 4-3 presents graphically the gasoline consumed by
4-20
-------
Table 4-8. ONBOARD CONSUMPTION PROJECTIONS
Year I
Gasoline Consumption
by Controlled Vehicles
(No Tampering)
(109 liters)
% Consumption
by Controlled Vehicles
(No Tampering)
Gasoline Consumption
by Controlled Vehicles.
(With Tampering)
- (1Q9 liters) I
% Consumption
by Controlled Vehicles
(With Tampering)
1988
89
90
91
92
93
94
95
96
97
98
99
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
33
62
87
110
131
148
164
177
189
198
206
213
218
223
225
227
229
230
231
231
231
231
231
231
231
231
231
231
231
231
231
231
231
11.0
21.3
30.5
39.4
47.6
54.4
61.9
67.8
73.5
78.3
82.7
86.6
89.7
91.8
92.6
93.4
94.2
94.7
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
32
60
84
105
124
140
153
165
175
182
189
195
199
203
207
210
211
212
212
212
212
212
212
212
212
212
212
212
212
212
212
212
212
10.7
20.6
29.5
37.6
45.1
51.5
57.7
63.2
68.1
71.9
75.9
79.3
81.9
" 83.5
85.2
86.4
86.8
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
-------
Figure 4-3. TOTAL GASOLINE CONSUMPTION VS. GASOLINE CONSUMED
BY ONBOARD CONTROLLED VEHICLES (1988-2020)
-A—A—A—A— Total gasoline consumption (see Figure 4-2)
Consumption by all LDVs and LDTs
Consumption by onboard controlled vehicles (no tampering)
Consumption by onboard controlled vehicles (with tampering)
109 1
350
300
250 —
200 —
150 —
100 —
50
109 gal,
90 —
80 —
70 —
60
50 —
40 —
30
20 —
10 —
^\ A A A A
1990
1995
•' i ! ; ! ' i
2000 2005
2010
2015
2020
YEAR
4-22
-------
controlled vehicles, with and without tampering, as well as total gasoline
consumption for 1988 to 2020.
4.3 REGULATORY STRATEGIES
Twenty-six regulatory strategies were evaluated for the gasoline
marketing industry. These regulatory strategies were chosen to delineate
the range of available strategies. The estimated relative costs of
control and associated reductions in emissions and health risk were
then assessed in order to bound the expected costs and impacts for all
strategies and to compare the relative costs and impacts among the
strategies. Thirteen of the strategies examined control of all facilities
while eleven of the strategies examined control of all but certain size
exempted facilities. The size exemptions were based on statutory
requirements or throughput levels below which control costs might be
expected to be excessive. The other strategies (benzene reduction in
gasoline) examine two alternate levels of control while two other
strategies (baseline or current controls and onboard controls for
vehicle refueling) do not lend themselves to exempting facilities.
Each regulatory strategy is comprised of a particular control
option for each source category. The source categories are bulk
terminals, including storage tanks, terminal-owned tank trucks and
for-hire tank trucks; bulk plants, including plant-owned and for-hire
tank trucks; and service stations, including filling or inloading of
the underground storage tank, all vehicle refueling, and as a subset,
self-service refueling. Section 4.4 describes the control options and
size exemptions considered for each source category.
The costs and impacts of each regulatory strategy were evaluated
for the time period beginning in 1986 through 2020. The initial year
of 1986 was chosen because at the time of the analysis it was the first
year that implementation of any control option could be begun. The
analysis was extended through 2020 so that all of the light-duty
vehicle fleet would be equipped with onboard controls and so that
Stage II and onboard controls could be evaluated with both options fully
implemented.
4-23
-------
TABLE 4-9. GASOLINE MARKETING REGULATORY STRATEGIES
STANDARD NUMBERS AND TITLES
Number
Title
Abbreviated
Ti tl e
I Baseline (No Additional Controls)
II A & Ba Stage II - Selected Nonattainment
Areas
III A & B Stage II - All Nonattainment Areas
IV A & B Stage I - Nationwide
V A & B Stage II - Nationwide
VI A & B Stage I and Stage II - Nationwide
VII Onboard - Nationwide
VIII A & B Stage II - Selected Nonattainment Areas
& Onboard - Nationwide
IX A & B Stage II - All Nonattainment Areas
& Onboard - Nationwide
X A & B Stage I & Onboard - Nationwide
XI A & B Stage II - All Nonattainment Areas and
Stage I & Onboard - Nationwide
XII A & B Stage II & Onboard - Nationwide
XIII A & B Stage I & Stage II & Onboard - Nationwide
XIV A & Bb Benzene Reduction in Gasoline
Baseline
Stage II - NA*
Stage II - NA
Stage I
Stage II
Stage I & Stage II
Onbd
Stage II-- NA*
& Onbd
Stage II - NA
& Onbd
Stage I & Onbd
Stage II - NA &
Stage I & Onbd
Stage II & Onbd
Stage I & Stage II
& Onbd
Gas Bz Reduction
dA-with size exemptions; B-no size exemptions
Stage I size exemptions:
(1) bulk plants with throughputs <4000 gal/d from balance controls on
outgoing loads; and
(2) service stations with throughputs <10,000 gal/mon.
Stage II size exemptions:
(1) all service stations with throughputs <10,000 gal/mon; and
(2) all independent service stations with throughputs <50,000 gal/mon.
bBenzene reduction:
A. removal of 94.5 percent of Bz from reformate fraction for total
reduction of 62.4 percent;
B. removal of 94.5 percent of Bz from reformate and FCC fractions for total
reduction of 81.3 percent.
4-24
-------
TABLE 4-10.
COMPOSITION OF REGULATORY STRATEGIES
BY SOURCE CATEGORY
Control Option3 for Each Source Category
Regulatory
Strategy
Bulk
Terminals
Bulk
Plants
Service Stations
GasoTi ne
Inloading
Vehicle
Refueling
I . Basel i ne B/L
II. Stage II - NA*
A. with size exemptions B/L
8. no exemptions B/L
III. Stage II - NA
A. with size exemptions B/L
B. no exemptions B/L
IV. Stage I
A. with size exemptions C
B. no exemptions C
V. Stage II
A. with size exemptions B/L
B. no exemptions B/L
VI. Stage I & Stage II
A. with size exemptions C
B. no exemptions C
VII. Onboard
MA*
VIII. Stage II -
Onboard
A. with size exemptions
3. no exemptions
IX. Stage II - MA a
Onboard
A. with size exemptions
3. no exemptions
3/L
3/L
B/L
B/L
B/L
X. Stage I & Onboard
A. with size exemptions C
B. no exemptions C
XI. Stage II - NA &
Stage I & Onboard
A. with size exemptions C
B. no exemptions C
B/L
B/L
B/L
B/L
B/L
C(E)
C
B/L
B/L
C(E)
C
B/L
B/L
B/L
B/L
B/L
C(E)
C
C(E)
C
B/L
B/L
B/L
B/L
B/L
St. I(E)
St. I
B/L
8/L
St.I(E)
St. I
B/L
B/L
B/L
B/L
B/L
St. I(E)
St. I
St. I (E)
St. I
B/L
St.II - NA*(E)
St.II - NA*
St.II - NA(E)
St.II - NA
B/L
B/L
St.II'(E)
St.II
St.IKE)
St. 11
Onbd
St.II - NA*(E) & Onbd
St.II - NA* & Onbd
St.II - NA(E) & Onbd
St.II - NA & Onbd
Onbd
Onbd
St.II - NA(E) & Onbd
St.11 - MA & Onbd
TABLE 4-10 CONTINUED
4-25
-------
TABLE 4-10. COMPOSITION OF REGULATORY STRATEGIES
BY SOURCE CATEGORY (concluded)
XM.
•
XIII.
XIV.
Regulatory
Strategy
Stage II 5 Onboard
A. with size exemptions
B. no exemptions
Stage I 5 Stage II & Onboard
A. with size exemptions
8. no exemptions
Gas Bz Reduction
A. 62.4% benzene Bz:
reduction
3. 81.3% benzene Bz:
reduction
Bulk
Termi nal s
B/L
B/L
C
C
0.376 x B/L
0.187 x B/L
Control Optiona for
Bulk
Plants
B/L
B/L
C(E)
C
Bzi 0.376 x B/L
Bz: 0.187 x 8/L
Each Source Category
Service
Gasoli ne
Inloading
B/L
3/L
St. I(E)
St. I .
Bz: 0.376 x 3/L
Bz: 0.187 x 3/L
Stations
Vehicle
Refuel ing
St. IKE
Onbd
St. II S
St. IKE
Onbd
St. II 5
Bz: 0:376
3z: 0.187
) 5
Onbd
) &
Onbd
x B/L
x B/L
aKey to control option abbreviations:
3/L - Baseline (Mo Additional Controls)
C - Controlled
St. II - HA* - Stage II in Selected Nonattainment Areas
St. II - HA - Stage II in All Monattainment Areas
St. I - Stage I Nationwide
St. II - Stage II nationwide
Onbd - Onboard Nationwide
St. II - HA* & Onbd - Stage II in Selected Nonattainment Areas and Onboard Nationwide
St. II - MA & Onbd - Stage II in All Nonattainment Areas and Onboard Nationwide
St. II i Onbd - Stage II & Onboard, Nationwide
(£) - with size exemptions
8z: 0.376 x B/L - benzene emissions equal to 0.376 times baseline level; other pollutants
unaffected
3z: 0.187 x 8/L - benzene emissions equal to 0.187 times baseline level; other pollutants
unaffected
4-26
-------
The industry-wide regulatory strategies that were evaluated in this
study are presented in Table 4-9. These regulatory strategies combine
the control options for each source category as shown in Table 4-10.
Baseline (Regulatory Strategy I) reflects current control under
promulgated and proposed Federal, State and local regulations. Bulk
terminals and their storage tanks are controlled under the bulk terminal
and volatile organic liquid storage new source performance standards
(NSPS). Bulk terminals, bulk plants, tank trucks, and underground
storage tank filling at service stations (Stage I) are covered by
control technique guidelines (CTG's) for ozone nonattainment areas and,
thus, many of the recommended controls have been incorporated into
State implementation plans (SIP's) and subsequent regulations. Also, a
few localities have implemented Stage II vehicle refueling controls in
order to attain the ambient ozone standards.
The nonattainment-area-only strategies (II and III) affect only
vehicle refueling emissions by recommending Stage II service station
controls in those areas while leaving all "Stage I sources" (bulk
terminals, bulk plants, and Stage I - gasoline inloading - at service
stations) at baseline control levels. All Stage II nonattainment area
controls were assumed to be installed during 1986, except for those on
independent service stations, which were phased in over a 3-year
period through 1988, as required by Section 324 of the Clean Air Act.
The nationwide Stage I and/or Stage II strategies (IV, V, and VI)
evaluate the implementation of Stage I (at all source categories
throughout the gasoline marketing system) and Stage II controls (at
service stations), either singly or in combination through a NESHAP.
Nationwide controls under a NESHAP were assumed to be installed equally
within the 2 years allowed under Section 112 of the Clean Air Act
beginning in 1987, except that Stage II controls at independent service
stations were assumed to be installed equally over a 3-year period in
accordance with Section 324 of the Act.
The impacts of onboard controls alone through a motor vehicle
requirement under Section 202(a)(6) of the Clean Air Act are assessed
under Regulatory Strategy VII. Onboard controls were assumed to be
installed on new light-duty vehicles and light-duty trucks beginning in
4-27
-------
1988 (see Appendix C). Onboard-controlled vehicles gradually comprise
more of the total fleet as additional new cars are built and as older
cars are scrapped. Onboard-controlled cars are projected to consume
over one-half of the on-highway gasoline consumption by 1993 (see Table
4-8). Essentially all of the light-duty vehicle fleet is expected to
be onboard-controlled by about 2006. Thereafter, only about 5 percent
of the on-highway gasoli.ne consumption (used by heavy-duty gasoline
trucks, motorcycles, etc.) would not be controlled by onboard.
Most of the remaining regulatory strategies (VIII through XIV)
evaluate the impacts of combinations of the previous strategies and,
in particular, combinations with onboard controls. One strategy (X)
combines nationwide Stage I controls throughout the system with onboard.
Several strategies (XVII, IX and XI) assess combinations of the
Stage II in nonattainment area options with onboard. The rationale
behind these strategies is to implement controls quickly in metropol-
itan areas to attain the ambient ozone standard and reduce hazardous
emissions, while relying on onboard controls (with Stage I controls in
Strategy XI) to reduce nationwide emissions more gradually. Two
other strategies (XII and XIII) similarly evaluate the combination
of nationwide Stage II controls with onboard (and with Stage I in
Strategy XIII). For the strategies combining onboard with
Stage II in nonattainment areas or nationwide, the Stage II controls
were assumed to be phased out, i.e., not replaced, upon completion of
one useful equipment life for balance and hybrid systems (15 years)
and two useful equipment lives for vacuum assist systems (16 years
at 8 years per life) because at that time, onboard controls would be
virtually fully implemented.
The final regulatory strategy (XIV) evaluates the effects of
reducing the benzene content of gasoline during refining. Two levels
of reduction were considered, based on 94.5 percent removal of benzene
from either the reformate fraction of gasoline alone or both the
reformate and fluid catalytic cracked (FCC) gasoline. The refinery
modifications for the benzene reduction were assumed to be installed
over a 3-year period, beginning in 1986, and to be completely effective
beginning in 1989. The removal of benzene from gasoline proportionately
4-28
-------
reduces evaporative emissions of benzene throughout the gasoline market-
ing system, but does not affect the emissions of other pollutants.
Although the reduction of benzene in gasoline apparently would have the
added benefit of reducing automotive emissions of benzene, one study
by the U.S. EPA Office of Mobile Sources (QMS)* suggests that only
evaporative automotive emissions would be reduced. The benzene exhaust
emissions were found to be approximately constant, regardless of gasoline
benzene content. Evidently, benzene is formed in the exhaust by crack-
ing of larger hydrocarbon molecules or reforming from smaller molecules.
4.4 SOURCE CATEGORY CONTROL OPTIONS
This section describes the control options evaluated for each of
the gasoline marketing industry source categories: bulk terminals, bulk
plants, and service stations (including self-service). The considered
levels of control are noted as well as any size exempted facilities.
The model plant sizes are summarized in Table 4-11. The numbers of
affected facilities for each source category and facility type are
presented in Table 4-12. Baseline controls reflect the implementation
of Federal, State, and local regulations as of 1982. The baseline
control option would require no additional controls.
4.4.1 Bulk Terminals
Two control options were examined for bulk terminals: no addi-
tional control measures (baseline controls) and control of all
facilities with no exemptions. The option requiring control without
size exemptions would limit VOC emissions from terminals currently without
controls to 35 mg/liter. This option would apply only to terminals
without controls. Terminals that are already controlled under CTG's
are required to emit no more than 80 mg/liter. In addition to the 35
mg/liter requirement, which would necessitate the use of some type of
vapor processor for the loading rack emissions, terminal operators
would be required to control fixed-roof storage tanks and to load only
a
Black, P.M., I.E. High, and J.M. Lang. Composition of Automotive
Evaporative and Tailpipe Hydrocarbon Emissions. Journal of the Air
Pollution Control Association. 30_: 1216-1221. November 1980.
4-29
-------
TABLE 4-11. GASOLINE MARKETING FACILITY MODEL PLANTS
Facility Type/
Model Plant Number
Bulk Terminals
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Bulk Plants
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Service Stations
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Model Plant 5
1
1
1 Model Plant Throughput
1 l 1 ters/ day
378,500
946,400
1,893,000
3,785,000
11,400
24,600
47,300
64,400
liters/month
18,900
75,700
132,500
246,100
700,300
gal Ions/ day
100,000
250,000
500,000
1,000,000
3,000
6,500
12,500
17,000
gall ons/mo nth
5,000
20,000
35,000
65,000
185,000
Throughput Range Represented by Model
Plant
11 ters/ day
0-757,100
757,100-1,514,000
1,514,000-2,271,000
>2, 271, 000
0-15,100
15,100-30,300
30,300-64,400
64,400-75,700
liters/month
0-37,900
37,900-94,600
94,600-189,300
189,300-378,500
>378,500
gallons/day
0-200,000
200,000-400,000
400,000-600,000
>600,000
0-4,000
4,000-8,000
8,000-17,000
17,000-20,000
gallons/month
0-10,000
10,000-25,000
25,000-50,000
50,000-100,000
>100,000
4-30
-------
TABLE 4-12. NUMBER OF FACILITIES AFFECTED
BY GASOLINE MARKETING CONTROL OPTIONS
FACILITY TYPE
BULK TERMINALS
BULK PLANTS
- No Exemptions
TOTAL FACILITIES
AFFECTED
500a
8,040t>
Number
MP 1
240
3,400
of Facilities
MP 2
135
4,000
in Each Model
MP 3
105
560
Plant (MP)"
MP 4
20
80
Size
MP
NA
NA
5
- Exempt Bulk Plants 8,040C.d
< 15,140 Ipd
SERVICE STATIONS
NATIONWIDE ALTERNATIVES
Stage I
- No Exemptions 197,900
- Exempt Stations
< 37,850 I/mo 83,100
Stage II
- No Exemptions 383,200
- Exempt Stations < 37,850 1/mo
and Independent
Stations < 189,300 1/mo 120,600
(Independents Only)
NONATTAINMENT ALTERNATIVES
Stage I
-•No Exemption Option 6,895
- Exempt Stations < 37,850 1/mo
2,895
Stage II
-No Exemption Option 114,900
- Exempt Stations < 37,850 1/mo
gal/mo and Independent
Stations <189,300 1/mo 36,200
(Independents Only)
FIXED ROOF STORAGE TANKS6 630
TANK TRUCKS
- For-hire Terminal Trucks 5,800 a
- For-hire Bulk Plant Trucks
(No Exemptions) 16,300b
- For-hire Bulk Plant Trucks
(<15,140 Ipd BP Exempt) 9,40QC,d
3,400
114,800
222,100
4,000
66,600
4,000
33,600
33,600
65,400
47,500
1,170
1,170
19,600
14,300
560
1,035
1,035
17,300
10,500
80
9,200
(3,150)
NA
29,700 15,800 4,000
29,700 15,800 4,000
57,500 30,700 7,500
34,900 30,700 7,500
(10,500) (2,700)
550 140
550 . 140
9,200 2,200
2,200
(800)
aBulk terminal affected facilities include about 2,870 terminal-owned trucks.
bBulk plant affected facilities with no exemptions include 16.000 plant-owned trucks.
cBulk plants with throughput <15,140 Ipd (4,000 gal/day) will oe required to use submerged fill on outgoing loads (account
trucks) .
dBulk plant affected facilities .under the with exemption option include 9300 plant-owned trucks requiring vapor
balance and 6700 plant-owned trucks requiring only submerged fill. An additional 6900 for-hire trucks would
require submerged fill.
eincludes only fixed roof. storage tanks at bulk terminals to oe retrofitted with internal floating r-oofs.
The two fixed roof tanks assumed per bulk plant cannot be retrofitted with internal floating roofs
since floating roofs are not compatible with vapor balancing.
4-31
-------
certified tank trucks. Fixed-roof storage tanks would be controlled
with internal floating roofs. Terminal operators would be required to
restrict loading of gasoline tank trucks to those trucks that had
passed an annual vapor tightness certification test. About 500 terminals
would be affected by the control option. An estimated 630 fixed-roof
storage tanks would have to be retrofitted with internal floating
roofs. In addition to the approximately 2870 terminal-owned tank
trucks, 5800 for-hire terminal trucks would have to be controlled.
4.4.2 Bulk Plants
For bulk plants, three control options were evaluated: no addi-
tional control measures (baseline controls) and controlled with and
without size exemptions. The control required would be vapor balancing of
storage tanks with transport trucks (bringing gasoline to bulk plants
from terminals), and either vapor balance or submerged filling of
account trucks (taking gasoline from bulk plants to service stations or
other accounts). Both transport and account trucks are required to be
vapor-tight. The bulk plant control option without size exemptions will
affect 8040 bulk plants, requiring vapor balance of both incoming and
outgoing loads. The control option with size exemptions affects bulk plants
loading less than 15,140 liters per day (4,000 gal/day). Under the
size exemption option, the 3400 small plants (MP1), which have a relatively
greater cost of control and are often exempted by existing State regula-
tions, are required to use only submerged fill to control emissions
from loading of outgoing (account) vapor-tight trucks, while using
vapor balance on storage tanks and incoming (transport) vapor-tight
tank trucks. The remaining 4,640 larger bulk plants (MP 2-5) would
still be fully controlled.
Plant-owned account trucks requiring vapor balance controls are
estimated to number about 16,000 with no size exemptions. The control
option with the 15,140 liters per day (4,000 gal/d) size exemption
would require vapor balancing of approximately 9,300 account trucks and
submerged filling of the remaining 6,700 trucks. Similarly, the for-hire
account trucks requiring vapor balance are estimated to number about
16,300 with no exemptions. The control option with size exemptions
4-32
-------
requires vapor balancing of approximately 9,400 account trucks and
6,900 trucks would be subject to submerged fill requirements.
4.4.3 Service Stations
Gasoline handling operations, emissions, and controls at service
stations are basically divided into two steps, which are commonly
termed Stage I and Stage II. Stage I refers to gasoline inloading at
the service station, that is, filling of the underground storage tank.
Stage II refers to vehicle refueling at the station. The control
options for service stations affect one or both of these stages to
varying degrees.
Stage I. Stage I, or inloading, controls employ vapor balance
between the tank truck and the underground storage tank at the service
station. This vapor balance step completes the gasoline marketing
Stage I system that captures and transfers potential emissions from the
service station storage tank, tank trucks, and, if applicable, bulk
plant, to the bulk terminal vapor processor for recovery or destruction.
In addition to baseline, service station inloading was evaluated for
Stage I vapor balance controls on all facilities and on all facilities
with gasoline throughputs greater than or equal to 10,000 gallons per month,
The size exemption reflects the higher relative costs of controls on
smaller facilities and existing size exemptions under State regulations.
The estimated number of affected facilities is 197,900 without exemp-
tions or 83,100 with a size exemption for stations with throughputs less
than 37,850 liters per month (10,000 gal/mo).
Stage II and Onboard. The vehicle refueling control options utilize
several different regulatory approaches and control techniques. One
control technique uses a vapor balance between the service station
underground storage tank and the vehicle gasoline tank and is commonly
termed Stage II controls. The other control technique uses a vapor
seal in the vehicle fill neck to force the vapors being displaced from
the tank into a carbon canister on the vehicle where they are adsorbed,
or onboard control. Implementation of Stage II controls, as well as
Stage I control, for service stations, bulk plants, and bulk terminals,
was evaluated on a nationwide basis. Stage II controls were also
4-33
-------
evaluated for ozone nonattainment areas only. Implementation of onboard
controls was assumed to occur nationwide. It should be noted that vehicle
refueling controls not only reduce VOC and hazardous emissions dispersed
from the service station, but also reduce exposure to hazardous pollutants
during self-service refueling. Reduction of self-service exposure
results in a significant reduction in estimated incidences of cancer (as
shown in Chapter 6) because about 90 percent of the incidences are
attributable to self-service refueling.
All Stage II options were evaluated for both with and without size
exemptions. Examined size exemptions include any service stations
generating less than 37,850 liters/month (10,000 gal/mo.) throughput
(based on existing State and local regulations and expected relative
economic impacts) and independent dealers below 189,300 liters/month
(50,000 gal/mo.) sales volume (in accordance with Section 324 of the
Clean Air Act and assuming that Section 324 applies to Section 112
regulations). These stations, for which size exemptions are examined,
represent the segment of the retail industry that may be most vulnerable
to changes in marketing economics as well as external costs such as
vapor recovery costs. The basic vehicle refueling control options
(excluding baseline controls and combinations of options) of Stage II
in selected nonattainment areas, Stage II in all nonattainment areas,
Stage II nationwide, and Onboard nationwide, are discussed in the
following paragraphs.
Stage II-NA. Stage II in all nonattainment areas (NA) was chosen
to reflect the development of a CTG recommending Stage II controls as
RACT for vehicle refueling VOC emissions. These areas are all ozone
nonattainment areas that were granted an extension (by EPA) up to 1987
for achieving the standard plus those ozone nonattainment areas that
EPA identified in Appendix D of the February 3, 1983, Federal Register
(48 FR 5005) as "unlikely to attain the NAAQS" by December 31, 1982.
It should be noted that Table III on page 5026 of the February 3, 1983,
Federal Register provides a listing of the "extension areas" for ozone
4-34
-------
) -« r^ CO
f)
in o '•Q
O"i ^^ •— »
t/1 S- O
rtJ JC *-< •
Of-—.
l CO C\J
*» O CT>
^3- CO
3 1-
1. O T3 «-
«C K ,— 1_ -o O
— o o s-
QJ 4-> i- V- fO >,
C ^- t- J- 3 4J
^ CO O T T* tJ
o
UJ
•=>
(S)
o! 7.
(O
z
o
<—G£.\—
< O <
z: u_ rs
•—• _i
CD «t
oo oo i— >
i— «C O UJ
I UJ
•sf Cti < C3
< Z
S- i—i
UJ h- CQ h-
—1 Z UJ
CQ UJ Q i^
< ^Z UJ Ql
1— 3 1— <
o
3
0; O
re £
U. UJ
) —. —t C3
o r-> — • CM ^
-
o
O
oo
cor-COCO^H r-cor")
cr»ovo«3-^r CM ro
CM — i •— " •— " — —i
5-1
. (I, ^—
i&^:
&_ eg re »—
O) = V O O
.^•DCJUJOJ'— — i-
OO_C^ = .— 'CC
5.1
i5 -^
C CJ r
4-35
-------
*-, ._;
O Cr
. ,_ ,— w. 3 .,_ COIO
= ^ CO
S_ rt3 C 1- -t— -^- C ^— oj —
•"
i -S
r*. f*» m co
LO o .— 4 o ^
O co co CO •"*
». vo c
a co r
O W f"> CO L
en —^ Cft —^ L
CO •
Cl C
ro —* «sf r-.
: o >•> (
( 4J C I
_ J
>V O =3 )— (d J
•? _ « .P p- !
cj o = t
OJ Crt > r— j
cj —i -J E
=J JC OJ O
<9c 2
«; o
4-36
-------
IO
en
o
2 I
O3
«C
oo
2
Q
CQ
a
UJ
i—
o
• u_
«*• =C res
•— oo
i oo _i
"d- eC o
LU CC
CQ O
VI 3
fl3 a.
O 3
e*»
H-
c
•— r- O
J^ M [^
£ 3
O Z O
a.
o
£
o
a.
C3 cn LA *f?
cn ^H cn cn
i— 4 CM cn
en co CM o
?"^ CO CD
o
^f CO P** *-^
CO «-« CO CO
O i-H CO
' i-H IO U*> **•
CM co r*«. in
•^ CO ^O *
CM" CM cn CM
CM CO P> , CM
0)
, c
V)
fQ
OJ
•o
3
O
f"
V)
4)
^
cn
u-
(U
•a
3
O
] *
IO
c
CD
c
o
VI
s
V
s
o
4->
U
4_>
VI
a
01
.c
VI
ra
VI OO
0)
4-J 4J
| |
l_4
IO
i— CO
c c
0 0
&- +•> VI
o — c
* 4->
c a.
cu c E
o o
rf 4J a o
a.' 4J
3 Qt C 41
O X OJ -C
ja LU o +^
0) 14-
>, • cu 0
.am
C CM 4->
4-» t— • C
3 re- -3- 4>
a. cu — . u
3 VI S-
t. 10 >» tu
-C J3 J2 O.
4->
•a 10 3 •
01 CLCO
4-> »-" 3 CM
o >— s-
01 J= >,
01 3
Ol ) 4-> CX
5^S2
O t- -C
•a
Ol
VI 1—
i— O
o s-
i. 4->
4-> C
c o.
o u
^ 2
Ol >»
4-> Ol
CO i-
5 "
•r- f—
•0 "*
i2
3
i— X
o
O CO
3
vi Q.
51
Ol 4-1
•O T3
C C
IO CM
VI
'o
^5
c
o
u
01
cn
4.9
CO
Ol
VI
•2
e
|
P—
0
• M
a?
tn
VI
0 0
CJ 4->
CM IO
•-4
•o
4-> C
3 IO
CO
•— O
•— O^
en
VI «3-
4-> *
Ol
VI <4-
3 O
"5 C
to O
VI •»—
VI 4->
y ,^
3
C O.
i- O
a.
VI
Ol r- •
•^ IO VI
4-> 4J I—
COO
3 4J S.
O 4J
o ia c
IO *i— i—^
3
•a .01
c. — cn
IO X IO
- t- 4J
>• C CO
i? 01
VI O Ol
•O Ol f- •—*
CM CO CM LO
i = = = =
01
10
Ol
s_
3
Ol
IO
4-" U
C I
O OJ
•a i—
c cn
01 C
CL-I—
Ol VI
-a
C 10
&. O 4-»
s.
VI Ol Ol c
^- CL O *
O 3
i_ m -a
4JO
01
3
O
O
o
z
O VI
o ••-
3 U
O 3
3 TJ
01
3 O i—
o .a o "O
o -a
O CM
T- CO VI
O VI
VI
Ol C Ol
•T- O U
C 4-> '>
3 rtj i-
O i— Ol
O 3 I/I
CL
IO O C
CL O
CM
-r- VI IO
3 S
O i—
C O
c: o
o 3
o
VI
fQ
•S:
>f_ -r- 1-4 •—I r— (fl 3 4-> -^
LU CO OO
-o
4. 1/1 3
Q. « J= r-
O
-------
TABLE 4-15
NONATTAINMENT AREAS COMMITTED TO
OR SCHEDULING STAGE II
VEHICLE REFUELING CONTROLS (NA*)
Area
Connecticut
Fairfield
Illinois
Cook
DuPage
Kane
Lake
McHenry
HUT
Maryland
Anne Arundel
Baltimore
Carrol
Harford
Howard
Montgomery
Prince George's
Baltimore City
New Jersey (Statewide)
a
Gasoline throughput at
Gasoline
Throughput3
1000 gal/yr
319,800
1,664,900
298,600
107,400
158,500
48,800
141,700
196,400
82,000 •
84,900
77,500
69,300
228,700
198,100
453,400
3,133,100
Baseline for National
Area
Mew York
Bronx
Kings
Nassau
New York
Queens
Ri chmond
Rockl and
Suffolk
Westchester
Virginia
Arlington
Fai rf ax
Loudon
Prince William
Pennsylvania
Bucks
Chester
Del eware
Montgomery
Philadelphia
TOTAL
Emissions Data Systems
Gasoline
Throughput3
1000 gal/yr
140,900
276,800
548,600
120,700
382,700
100,200
98,700
483,600
321, 30Q
67,900
197,300
35,800
81,600
176,600
163,200
191,600
254,200
397.900
11,302,700
(NEDS).
4-38
-------
with the exception that Tennessee-Nashvilie-Davidson County was
inadvertently omitted from that table under Region IV, and subsequently
was omitted from the list of nonattainment areas for the analysis. The
nonattainment areas that were assumed to be affected by a CTG
recommending Stage II controls as RACT are listed with their baseline
gasoline throughputs in Table 4-13. This list includes all the areas
given as "extension areas" or "unlikely to attain the NAAQS" in the
February 3, 1983, Federal Register, except for those areas already
using Stage II controls (26 counties in the Bay Area, San Diego, South
Coast, and Sacramento AQCR's of California plus the District of Columbia,
with a total throughput of about 34 billion liters [9 billion gallons]).
Stage II-NA*. Stage II in selected nonattainment areas (NA*)
was chosen to reflect the lowest feasible level of vehicle refueling
control through a control techniques document (CTD). The selected
nonattainment areas (NA*) are a subs.et of the ozone nonattainment areas
discussed earlier that were granted extensions (by EPA) up to 1987 for
achieving the ambient standard for ozone. Table 4-14 presents a com-
parison of the relative sizes of the selected (NA*) and all (NA)
nonattainment areas. The NA* areas are listed in Table 4-15 along with
their baseline gasoline throughputs. These areas have determined they
will need more than the reasonably available control technology
(RACT) specified in control techniques guidelines (CTG) documents for
various sources as well as control of other 100-ton stationary sources
(plus vehicle inspection and maintenance programs [I/M], Federal motor
vehicle control programs [FMCP], and transportation control measures
[TCM's]) to attain by the statutory deadline. Accordingly, the SIP's
for these areas include a commitment/schedule to evaluate and adopt
Stage II as an additional means of achieving necessary reductions in
VOC emissions to provide for attainment by 1987.
National Stage II. Stage II nationwide was selected as.a control
option to assess the effect of a NESHAP to reduce benzene and other
hazardous emissions. The estimated number of affected facilities for
this option is 383,200 with no exemptions or 120,600 with size exemptions
compared with 114,900 for a CTG affecting all nonattainment areas with
no exemptions or 36,200 with size exemptions. In contrast, a CTD affecting
4-39
-------
selected nonattainment areas would affect an estimated 43,700 facilities
with no exemptions or 13,800 with size exemptions.
Onboard. Onboard controls nationwide is the other basic vehicle
refueling .control option. The analysis assumed that onboard controls
would be installed only on new light-duty vehicles (passenger cars) and
light-duty trucks (< 8500 Ib, i.e., vans, pickup trucks, etc.). Thus,
the effectiveness of onboard controls increases with time. For this
reason, the analysis was extended through the year 2020 so that the
analysis could continue after the entire light-duty vehicle fleet would
be controlled. In this way, onboard could be compared with Stage II in
both the short and long term.
4-40
-------
4.5 REFERENCES
1. Bulk Gasoline Terminals - Background Information for Proposed
Standards. U.S. Environmental Protection Agency. Office of Air
Quality Planning and Standards. Research Triangle Park, N.C.
Publication No. EPA-450/3-80-038a. December 1980. p. 3-1.
2. Reference 1, Section 6.2.4.
3. Graver Tanks. Commodity Storage Tank Product Literature. No date
available.
4. Pacific Environmental Services, Inc. Study of Gasoline Vapor
Emission Controls at Small Bulk Plants. Report to U.S. Environmental
Protection Agency. Region VIII, Denver, Colorado. Contract No. 68-
01-3156, Task Order No. 15. October 1976. p. 3-8 through 3-14.
5. Arthur D. Little, Incorporated. The Economic Impact of Vapor Control
Regulations on the Bulk Storage Industry. Report to U.S. Environ-
mental Protection Agency, Research Triangle Park, N.C. EPA Publication
No. EPA-450/5-80-001. June 1979. p. III-9.
6. Reference 4, p. 3-14.
7. Lundberg Estimates. National Petroleum News, September 1983.
p. 12.
8. Arthur D. Little, Inc., The Economic Impact of Vapor Recovery
Regulations on the Service Station Industry. U.S. Occupational Safety
and Health Administration, Washington, D.C. and U.S.. Environmental
Protection Agency. Research Triangle Park, N.C. Publication
No. EPA-450/3-78-029. July 1978. p. 47.
9. National Petroleum News, February 1983. p. 9.
10. "Franchising in the Economy, 1981-1983", U.S. Department of
Commerce, January 1983.
11. National Petroleum News. Factbook Issues. Mid-June 1978-1983.
12. Reference 8, p. 43.
13. U.S. Department of Commerce. 1977 Census of Retail Trade.
14. Reference 8, p. 32.
15. National Petroleum News. 1983 Factbook Issue. Mid-June 1983,
Volume 75, No. 7A. p. 80.
16. U.S. Environmental Protection Agency. Federal Register, Vol. 47,
Number 210, October 19, 1982. p. 49329.
17. Telecon. Scott Osbourn, Pacific Environmental Services, Inc., with
Debra Paxton and Mark Rodekar, U.S. Department of Energy Mid-term
Forecasting Branch. September 30, 1983. Department of Energy
gasoline projections.
4-41
-------
18. long, Y.P. and D.W. Houser.
Science. Oil & Gas Journal
Motor-gasoline Forecasting: an Embryo
November 14, 1983. p. 114-118.
19. Energy and Environmental Analysis, Inc. The Highway Fuel Consumption
Model Ninth Quarterly Report. U.S. Department of Energy. Washington,
D.C. Publication No. DOE/PE/70045-1. February 1983. Appendix B.
20. Exxon Research and Engineering Company Cost Comparison for Stage II
and On-board Control of Refueling Emissions. In: On-board Control
of Vehicle Refueling Emissions Demonstration of Feasibility.
American Petroleum Institute. API Publication No. 4306. Washington,
D.C. October 1978. p. 9.
21. American Petroleum Institute. Cost Comparison for Stage II and On-
board Control of Refueling Emissions. Washington, D.C. January
1984. p. 12.
22. Peterson, P.R. et _aJL Evaluation of Hydrocarbon Emissions from
Petroleum LiquiT~Storage. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Publication No.. EPA-450/3-78-012.
March 1978.
23. Pacific Environmental Services, Inc. Estimated Nationwide Petroleum
Storage Tank VOC Emissions for the Years 1983 and 1988. Report to
TRW Environmental Engineering Division. Research Triangle Park,
N.C. Contract No. M23399JL3M. April 5, 1983.
4-42
-------
5.0 ENVIRONMENTAL AND ENERGY IMPACTS
The purpose of this section is to discuss the environmental
and energy impacts associated with the gasoline marketing control
options and regulatory strategies. The majority of the discussion will
be spent on the methodology used to generate the quantitative analysis
on air pollution emission impacts. A quantitative analysis of the
energy impacts of the regulatory strategies is also included in this
section. The scope of the overall analysis did not allow for an in-
depth evaluation of the other environmental impacts (i.e., water,
solid waste); however, a qualitative discussion of these impacts is
included.
5.1 AIR POLLUTION EMISSION IMPACTS
Estimates of the emission reductions which could be achieved under
each of the control options and the regulatory strategies were analyzed
and are discussed here. The analysis for each industry sector was
basically the same. The potential emission reductions achievable in
the base year (1982) were calculated for each industry sector. Given
no other changes, the emission reductions would be the same in each
subsequent year of the analysis. However, there were two factors which
affected the emission reductions in each year: 1) phase-in of control
equipment installations, and 2) change in gasoline consumption with
time.
Section 5.1.1 discusses the initiation date of the control options
and the phase-in schedules assumed. Emission reductions were presented
for the study period, both as cumulative and discounted totals.
Section 5.1.2 discusses the rationale for discounting emissions.
Section 5.1.3 discusses the assumptions and methodologies used to
generate the emission reduction estimates, including how gasoline
consumption changes with time were incorporated into the analysis.
5.1.1 Phase-in Schedules for Control Options
It is unreasonable to assume that all equipment required by the
regulatory strategies would be installed immediately at the time the
regulatory strategy takes effect. It was assumed for the analysis that
5-1
-------
some period of time or phase-in of controls would take place before
100 percent of the facilities had controls installed. This phase-in,
therefore, would affect both the cost and emission reduction analyses.
In all cases a linear phase-in was assumed which resulted in an even
distribution of installations with time. In addition, statutory time
frames were used to consider complete phase-in of equipment.
Under Mational Emission Standards for Hazardous Air Pollutants
(NESHAP) programs, the Clean Air Act requires compliance with regulations
within 180 days; however, variances can sometimes be obtained for up to
2 years. Therefore, a linear phase-in period of 2 years was selected
for all nationwide options (i.e., terminals, bulk plants, storage tanks,
tank trucks, service station Stage I, and non-independent service station
Stage II). Further, Section 325 of the Clean Air Act specifically allows
a different phase-in rate (3 years) for independent service stations.
EPA has not determined whether this applies to a Section 112 standard;
however this was assumed for the analysis. For all options which
required Stage II controls in nonattainment areas, a phase-in rate of 1
year was used for all sources with the exception of independent service
stations where a 3-year phase-in was used. Figure 5-1 illustrates the
three linear phase-in rates used for the control options and strategies.
For purposes of determining capital cost estimates, any costs
spent in a year were attributed to that year (100 percent for 1-year
phase-in, 50 percent per year for 2-year phase-in, and 33 percent per
year for 3-year phase-in). For emissions estimations and for annualized
cost estimates the number of facilities which achieved the emission reduc-
tion or incurred the total annualized costs for the entire year had to be
determined. This number of facilities was estimated by determining the
area under each curve. For example, the following equation was used to
determine the values for the 3-year independent service station phase-in:
1/3 X = (1/6 XJ; - 1/6 X?)
where
A = Area under the curve of Y = 1/3X
5-2
-------
re
01
-CM
CO
OJ
O)
J=
O
O)
CO
re
re
0)
0)
O
CD
O
CO
O
CM
ijq.LM saj.q.LL
5-3
-------
Using this formula, the percentage of total facilities to be considered
for annualized cost and emission reduction in each year of the 3-year
phase-in was determined (year 1-17 percent, year 2-50 percent, year
3-83 percent, year 4 and thereafter - 100 percent). This same approach
was used to determine the percentages for a 2-year phase-in (year 1 -
25 percent, year 2-75 percent, year 3 and thereafter - 100 percent)
and a 1-year phase-in (year 1-50 percent, year 2 and thereafter -
100 percent).
The onboard control option was based upon the installation of
controls on new vehicles. Therefore, the number of vehicles controlled
in each year of the analysis is based upon the number of new vehicles
and the scrappage rates projected in Appendix C.
The start date from which the phase-in would begin for the
regulatory strategies was estimated based upon assumptions of when
EPA would decide on the control approach to pursue. At the time of the
analysis, it was assumed that EPA would decide on an approach in 1984.
Even if this dates slips uniformally for all control strategies,
the relative nature of the impacts between strategies do not change,
only the actual start dates. If EPA decided to pursue nationwide
regulatory strategies, it was estimated that the data gathering and
review processes involved with standards development would take 3 years.
Therefore, 1987 was selected as the date when nationwide options (with
the exception of onboard) would become effective and would be the
initial date of facility phase-in. If EPA decided to proceed with
nonattainment area regulatory strategies, it was estimated that a
guideline document could be developed in approximately 1 year and that
it would take another year for the regulatory strategies to be incorporated
into State regulations. Therefore, 1986 was selected as the effective
date of nonattainment area regulatory strategies. If EPA were to
pursue onboard regulatory strategies, it was estimated that a vehicle
standard could be prepared by 1985; however, the vehicle manufacturers
would need 3 years to incorporate the controls into the vehicle design.
Therefore, 1988 was selected as the effective date for onboard controls.
Table 5-1 summarizes the effective dates and the phase-in schedules for
onboard, nonattainment area, and nationwide control options and regulatory
strategies.
5-4
-------
GO
LU
=D OO
Q UJ
LU >-H
3= ts
O LU
5
•Z. 01
I-H h-
1 GO
LU
GO >•
< 01
Z O
rt I
LJ_ r~~
a _i
21 Z3
< CD
LU
oo o;
LU
1— CS
< z:
a HH
lit lit
LLJ t_LJ
^^ **"^
•-i cc
l— - tO
-p a;
•i- CU >-
3 S.
00 -P
a; c co
•r- LU
•P i.
•t- S- to
r— O <1)
•!-<*->-
O
U_ 0)
i— CM
•P i—
C (O S-
a> -P as
o oo a>
i- C >-
CD HH
00 •— 1
CU r~~
C7> O
(0 S- S.
i- -P (0
a> c cu
> o >-
< 0
c
1— 1
1 CU
CU -P
oo as
as a:
_g^
Q-
00
c
o
01- •!-
O -P
(O
s_ ,—
0) 'tO
>™ ^J
to
i— C
ro HH
•1—
^_j r_
••- o
£= S-
t— 4 -|_>
c
o
0
>^
di
CU
•p
(C
s-
^J
oo
o o
o o
T— t .— i
CO O
CO O
t— 4
O UO
LO r~-
l~- LO
t— 1 CM
S- S-
(O C^
CU CU
1 1
CO CM
r*"-» r*"*
CO OO
er> en
i — * » — i
-P
*— oo c:
cu ~o c cu oo
-OS- O £ CU
•i- > (O 4-> 'f C -i-
3 CU O E 4J -r- CD
£— *^~ *^ CU ^3 'O CU
OC7>£= -O-P S-00 4-»-P
•i-CUO COO CUO) -P(O
•P -P CU -C -I— fO i-
rt3 (0 ^_> r^ CU 4-* 4~* C -p
CS-Q. CUO O-^ OOO
-P CU "O -i- r— 2:
i — OOO C> i — T- a}
i — X i— i i- i — O i — CU
•^ 1 1 1 cu *3C to i — i-
— oo u_ «=C •<
o o
0 O
<— 1 1—1
co o
CO O
1— 1
O 0
LO . O
T— (
r-. o
r-H LO
1_ S-
, >> re
1 I CTi O^
i— 1 i— < i-H
OO
c:
o
-p •.-
c: -P
CU as oo
-a -p oo cu
£= GO s- cu .-a -i-
CU CU •!— S- CT>
a. cu x: -P re cu
cu o -P •!- o -P
•a T- o i— .a «
c: > -^- e: s_
HH t- ^ 0 ' 0 -P
CU i — as OO
OO
o
O -E
•P
M- •!-
O S
C CD
l— i C
1 '!—
CU -P
00 S-
fO fO
.c P
Q_ 00
n3
5-5
-------
5.1.2 Discounting of Emission Reductions
Emission reductions for each year through 2020 were discounted
at 10 percent and then added to obtain the present "value" of
emission reductions, a single number encapsulating both the magnitude
and the timing of the reductions. Dollar costs and savings were •
handled the same way. The discounting of dollars is common practice.
However, the discounting of benefits that are not expressed in monetary
terms, such as lives saved, emissions reduced, and visibility improved,
is not done as commonly. Typically, these other benefits flow more-or-
less evenly over time through the life of the control equipment.
However, the dollar costs almost never flow evenly over time, and
therefore are discounted to the present and, if costs are to be compared
with the annual flows of, say, emission reductions, then the costs are
reannualized into constant dollar cost flow over time. Then the two
flows, money and emission reductions, for various regulatory strategies,
can be compared.
The question of whether to discount emission reductions does not
arise in this typical example because there is no need for discounting.
The discounted and reannualized value of any constant stream, such as
annual emission reductions, is exactly equal to the original stream,
regardless of what interest rate was used to compute both the present
value of the stream and the reannualized stream.
The regulatory strategies in this analysis are not typical, because
the emission reductions are not smooth over time in every situation.
For example, the emission reductions from control expenditures made for
Stage II controls for a single service station will decline signifi-
cantly year-by-year if onboard controls are implemented simultaneously.
It is not logical to ignore this effect of time on emission reductions.
One way to summarize the situation for decision makers is to graph the
flows of money and emission reductions over the years. Such graphs can
be confusing if a decision is to be made between two close alternatives.
If more concise numerical comparison is to be made, some form of
discounting is necessary, even if the procedure is to be implicit.
5-6
-------
Simply adding up the emission reductions without regard to their
flow over time amounts to the use of a zero interest rate in .
discounting. A zero rate implies that society is indifferent between
a unit of'emission reductions today and a unit sometime in the future.
For example, if a person were given the option of sniffing benzene
today or sniffing the same amount 10 years hence, it is highly likely
that such a person would opt for the 10-year delay. This reasoning
indicates that a positive, non zero interest rate should be used to
discount emission reductions. Unfortunately, there is little theoretical
guidance available for selecting the appropriate rate. This analysis
uses 10 percent for all discounting (the effects of alternative discount
rates on costs are discussed in Chapter 8.0).
In summary, given a choice between sniffing hazardous emissions
now and sniffing it later, most people would prefer to wait. Clean
air today is more important than clean air tomorrow. This "time
preference" is expressed mathematically by discounting future emission
reductions with a positive interest rate. The analysis sought to
determine how much emission reduction will result from controlling
service stations and cars. The answer for service stations depends
on what is done to control cars, and vice versa. This means that a
dollar a year spent to control pollution at a service station will
not necessarily yield the same emission reduction in one year as it
does in another year. Regulatory strategies may be compared with each
other using emission reductions, cost/effectiveness, and a variety
of other measures of costs and benefits. If the costs or benefits
change over time, the best way, theoretically, of comparing alternatives
is to discount the measures before comparison is made. Then either
present values or reannualized flows may be used.
5.1.3 Emission Reduction Methodology
5.1.3.1 Base Year Emission Reductions. The methodology for
calculating emission reductions for bulk terminals, bulk plants, tank
trucks, storage tanks, and service stations was the same. The baseline
tables (Appendix B) were used in estimating emission reductions. These
5-7
-------
tables supplied the baseline emissions for the 1982 base year. Emission
factors associated with the regulatory strategies discussed in Section 4.3
were then assigned to the throughputs in any of the areas in the baseline
table not controlled in 1982. A controlled emission rate was calculated
and the difference between this rate and the baseline emission rate was
the emission reductions achievable by the regulatory strategies in the
base year of 1982.
For example, the baseline emission rate of gasoline vapors for
bulk terminals was estimated as approximately 140,200 Mg/yr. This
baseline included controlled and uncontrolled facilities. The regulatory
strategies which affect terminals require: 1) all terminals, not
previously controlled, to install control equipment to meet a limit of
35 nig/liter, and 2) all tank trucks loading at terminals should be
leak-tight (leakage emission factor - 96 mg/liter). The baseline
emission table for bulk terminals was then adjusted to reflect-these
control measures. The controlled emission rate in 1982 for bulk
terminals was then calculated to be approximately 55,800 Mg/yr. The
potential emission reduction in 1982 for the bulk terminal regulatory
strategies was estimated as 84,400 Mg/yr.
Table 5-2 summarizes the control requirements for each of the
control options, and Table 5-3 summarizes the emission reductions
achievable in 1982. The emission factors associated with the control
requirements are discussed in Chapter 3.0, Control Technology. All the
emission factors, with the exception of storage tanks and evaporative
emissions, are in the format of mg/liter and emissions estimates are
generated by multiplying the factor times the gasoline throughput.
This is also true of storage tank working losses. However, the storage or
seal losses are determined on a per tank basis, so an estimate of the
number of tanks requiring controls had to be made (see Appendix 3, Section
B.2.2). Excess evaporative emissions are estimated from vehicle miles
traveled by vehicles with onboard controls.
The vehicle refueling emission reductions for service stations
in nonattainment areas were calculated as discussed for terminals. This
control option for service stations further required underground tank
controls (Stage I at service stations) for those areas which did not
have service station Stage I regulations. The only areas in the "all
nonattainment area" option where no service station Stage I controls
5-8
-------
TABLE 5-2. SUMMARY OF CONTROL REQUIREMENTS AND AFFECTED EMISSION
FACTORS FOR GASOLINE MARKETING CONTROL OPTIONS
Control Option
Control Requirements
Affected
Emission Factor3
Assigned For
Controls, ing/liter
Bulk Terminals
- Loading Racks
- Storage Tanks
Bulk Plants
- No Exemptions
- Size Exemptions
Service Stations
(Stage I)
- Nationwide
o No Exemptions
o Size Exemptions
- All Nonattainment
Areas
1) , Vapor processors required on all 35
uncontrolled terminals
2) Leak-tight tank trucks at all 96
terminals
1} Internal floating roof on all exist- b
ing fixed-roof tanks at terminals.
1) Vapor balance on all incoming loads
a) Storage Tank Filling 57.5
b) Storage Tank Draining 46
2) Vapor balance on all outgoing loads 96
1) Vapor balance on all incoming loads
a) Storage Tank Filling 57.5
b) Storage Tank Draining 46
2) Vapor balance on all outgoing loads 96
at bulk plants 215,100 liters/day
(72 percent of throughput)
3) Submerged fill on outgoing loads at 600
bulk plants <15,100 liters/day
(28 percent of throughput)
1) Vapor balance system for all 40
underground tanks
1) Vapor balance systems for all 40
underground tanks at service
stations 237,900 liters/month
(86 percent of throughput)
Same requirements as nationwide
See discussion of emission factors assigned for controls in Chapter 3.0,
Control Technology.
b
Calculated emission factors for internal floating-roof tanks were: 2.4 Mg/yr/tank for
storage losses and (7.3285 x 10~8 Q) Mg/yr for working losses where Q is the product
throughput in barrels per year (Ref. 7).
5-9
-------
TABLE 5-2. SUMMARY OF CONTROL REQUIREMENTS AND AFFECTED EMISSION
FACTORS FOR GASOLINE MARKETING CONTROL OPTIONS
(concluded)
Control Option
Control Requirements
Affected
Emission Factors
Assigned For
Controls, mg/Liter
Vehicle Refueling
- Stage II
o Nationwide
- No Exemptions
1) Stage II vapor recovery on all
vehicle refueling operations
at service stations
Vehicle Refueling
Underground Tank
Breathing
- Size Exemptions 1)
o All Monattainment
Areas
o Selected
Monattainment
- Onboard
Stage II vapor recovery on all
vehicle refueling operations at
non-independent service stations
37,900 liters/day and at
independent service stations
189,000 liters/day
(71 percent of throughput)
Vehicle Refueling
Underground Tank
Breathing
Same requirements as nationwide
Same requirement as nationwide
1} Carbon canister/modified fillpipe
system on all new cars after 1988.
Vehicle refueling
Excess evaporative control
54
60
54
60
21.6
0.13 g/mi
See discussion of emission factors assigned for controls in Chapter 3.0,
Control Technology.
5-10
-------
TABLE 5-3. GASOLINE VAPOR EMISSION REDUCTIONS IN 1982
ASSOCIATED WITH CONTROL OPTIONS
Control Option
1982 Baseline
Emissions, Mg/yr
1982 Controlled
Emissions, Mg/yr
a
Emission
Reduction, Mg/yr
Bulk Terminals
- Loading Racks
- Storage Tanks
Bulk Plants
- No Exemptions
- Size Exemptions
Service Stations
- Stage I
o Nationwide
- No Exemptions
- Size Exemptions
o All NAa Areas
- No Exemptions
- Size Exemptions
- Stage II
o Nationwide
- No Exemptions
- Size Exemptions
o All NAa Areas
- No Exemptions
- Size Exemptions
o Selected NAa Areas
- No Exemptions
- Size Exemptions
140,200
56,300
208,000
208,000
222,200
222 ,200
222,200
222,200
406,900
406,900
406,900
406,900
406,900
406,900
55,800
30,700
76,800
90,300
60,300
118,700
215,500
216,400
31,400
149,200
285,500
320,600
364,000
375,700
84,400
25,500
131,200
117,700
-
161,900
103,500
6,700
5,800
375,500b
257,700b
121,400b
86,300b
42,900b
31,200b
MA = Nonattainment.
Includes reductions in breathing loss emissions due to Stage II controls.
5-11
-------
are required are the 11 counties in the Atlanta, Georgia area and
Maricopa County, Arizona. All of the areas in the "selected non-
attainment area" option already have Stage I controls installed. For
the Stage I uncontrolled areas, potential emission reductions were
calculated based upon the throughput for these areas and appropriate
emission factors for Stage I controls. For those control options which
incorporated size exemptions, the percent throughput for the exempt
category was determined from the model plant size distributions (see
Section 4.1).
The onboard control option analysis differed from that of the
other control options. The onboard control option required the
installation of carbon canisters and fill pipe seals on all new light-duty
vehicles and trucks sold beginning with the 1988 model year. As more new
vehicles were manufactured and sold, more emission reductions could be
attributed to the onboard control option. Gasoline consumption by
controlled vehicles for each year was estimated by using the fuel
economy associated with each model year, the annual mileage accrual
rates, and annual scrappage rates (see Appendix C, Section VIII).
Emission reductions achievable with onboard controls were calculated by
multiplying the gasoline consumption associated with controlled vehicles
by an emission factor representing no vehicle refueling controls (1,080
mg/liter), and then multiplying the uncontrolled emissions by a factor
which represented the control effectiveness of onboard controls (98
percent control). The numbers of new automobiles, and the scrappage
rates, again discussed in Appendix C, were used to determine the number
of controlled vehicles in each year. Table 5-4 summarizes the results
for the onboard emission reduction analysis. Emission reductions are
shown for both control of vehicle refueling only and for the combination
of refueling and expected excess evaporative emission control. Emission
reductions for vehicle refueling operations are dependent on the gasoline
consumption by onboard controlled vehicles, while the evaporative
emission reductions are dependent on the vehicle miles traveled (VMT)
by onboard controlled vehicles. Aggregate emission reductions increase
steadily as more new vehicles are required to have onboard controls.
5-12
-------
Table 5-4. SUMMARY OF PROJECTED TOTAL ONBOARD EMISSION IMPACTS IN EACH
YEAR OF THE STUDY (1988-2020)
Projected
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
New
Vehicles
(million)
13.35
13.30
13.13
13.48
13.68
13.68
13.60
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
Gasoline
Consumption
by
Controlled Vehicles3
(109 Liters)
33
62
87
110
131
148
164
177
189
198
206
213
218
223
225
227
229
230
231
231
231
231
231
231
231
231
231
231
231
231
231
231
231
VOC
Emission
Reductions3'15
(103 Mg)
35
65
92
117
138
157
173
188
200
210
218
226
231
236
238
240
242
243
244
244
244
244
244
244
244
244
244
244
244
244
244
244
244
VOC
Emission
Reductions
(with E~vap.)a'c
(103 Mg)
61
117
167
214
256
294
327
356
382
404
423
-437
450
455
458
460
462
462
463
464
464
464
464
464
464
464
464
464
464
464
464
464
464
No tampering.
)
From vehicle refueling operations only.
'From vehicle refueling operations and estimated control of excess
evaporative emissions.
5-13
-------
5.1.3.2 Cumulative and Discounted Emission Reductions. The
emission reduction analysis evaluated the impacts of the control
options and regulatory strategies for each year from 1986 through the
year 2020. Emission reductions for each option were calculated for
each year and cumulative and discounted emission reductions were
generated. The emission reductions in each year of the study period
can be found in Appendix E.
As discussed in Section 4.2, gasoline consumption was projected to
decrease substantially during the years 1986 to 2000. Leaded gasoline
was projected to decrease even more rapidly due to lead phase-down
regulations. Tables 5-5 and 5-6 present the percentage decreases in
total gasoline consumption and leaded gasoline consumption, respectively,
which were projected to the year 2000 from Figure 4-2. The analysis
further assumed that the gasoline consumption would remain constant
after the year 2000.
The emission reduction calculations are directly proportional
to gasoline throughput since the reductions are based on consumption
and emission factors. To develop emission reductions over time,
therefore, the base year emission reductions for gasoline vapors and
benzene were decreased each year at the same rate as the total gasoline
consumption was projected to decrease. For example, if gasoline
consumption did not change, the gasoline vapor emission reductions from
bulk terminals in 1990 would be the same as for 1982, or 84,400 Mg/yr.
However, the 1990 consumption is only 75.7 percent of the 1982 consumption.
The gasoline vapor emission reduction achievable in 1990 was, therefore,
estimated as only 75.7 percent of the 84,400 Mg/yr, or 63,900 Mg/yr.
This type of calculation was repeated for each control option for each
year of the analysis.
EDB and EDC emission reductions can only be achieved from
leaded gasoline, therefore the leaded gasoline projections were
used to estimate EDB and EDC emission reductions in a given year.
The EDB and EDC emission reductions were obtained by multiplying
the gasoline vapor emission reduction for a particular year by the
EDB or EDC emission ratio (see Section 2.2) and by the percent of
total consumption in that year which was leaded gasoline.
5-14
-------
Table 5-5. PROJECTED TOTAL GASOLINE CONSUMPTION CHANGES
FROM THE BASE YEAR
Consumption3
109 Liters 109 Gal
% of 1982 Consumption % Decrease
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000-2020
389
364
349
338
326
317
309
300
294
288
284
280
273
269
265
261
257
254
250
102. 7a
96. lb
92.3
89.2
86.1
83.8
81.5
79.2
77.7
76C
75
74
72
71
70
69
68
67
66
100
93.6
89.7
86.9
83.8
81.6
79.4
77.1
75.7
74.0
73.0
72.1
70.1
69.1
68.2
67.2
66.2
65.2
64.3
0
6.4
10.3
'13.1
16.2
18.4
20.6
22.9
24.3
26.0
27.0
27.9
29.9
30.9
31.8
32.8
33.8
34.8
35.7
1982 consumption from Reference 1.
)
1983-1990 consumption from Reference 2.
'1991-2020 consumption estimated from Figure 4-2
5-15
-------
Table 5-6. PROJECTED LEADED GASOLINE CONSUMPTION
Total Gasoline
Consumption
109 Liters 109
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000-2020
364
349
338
326
317
309
300
294
288
284
280
273
269
265
261
257
254
250
96.
92.
89.
86.
83.
81.
79.
77.
76^
75
74
72
71
70
69
68
67
66
Gal.
1*
3
2
1
8
5
2
7
i
Leaded Gasoline
Consumption
109 Liters 109 GAL.
158
134
112
96
84
74
64
56
49
42
34
28
23
19
15
11
8
4
41.79
35.4
29.7
25.3
22.1
19.5
17.0
14.7
13t>
11
9
7.5
6
5
4
3
2
1
% of Total
Gasoline
Consumption
43.4
38.4
33.3
29.4
26.4
23.9
21.5
18.9
17.1
14.7
12.2
10.4
8.5
7.1
5.8
4.4
3.0
1.5
1983-1990 consumption values from Reference 2.
1991-2020 consumption values estimated from Figure 4-2.
5-16
-------
Another factor which affected the emission reduction in the
initial years of the analysis was the phase-in of control equipment.
Obviously, the total emission reduction achievable in a year is
dependent upon the number of facilities that had control equipment
installed. Therefore, the emission reductions for the initial years of
the analysis took into account the percent of facilities which had
equipment installed in that year (see Section 5.1.1 on facility phase-in
rates).
Table 5-7 summarizes the cumulative emission reductions (simple
sum of all reductions in all years) and the 1986 net present value
(NPV), or discounted, emission reductions (discounted at 10 percent).
All discounting was brought back to a NPV in 1986 since this was
the earliest year any of the regulatory strategies took effect.
The onboard cumulative and discounted emission reductions were
developed from the data presented in Table 5-4. After phase-in,
all of the other control options are affected only by the reduction in
gasoline consumption. The onboard control option, on the other
hand, does not cover the entire vehicle fleet until about the year
2002 or 2003 due to the phase-in of new controlled vehicles and the
retirement of existing uncontrolled vehicles. However, once onboard
controls are in place for the entire fleet, greater emissions
reductions can be obtained with this option when compared with Stage
II options because of the higher control efficiencies involved.
The nationwide emission reductions based on theoretical -
efficiencies associated with the regulatory strategies, discussed in
Section 4.3, are, in most cases, simply a combination of the emission
reductions for the appropriate control options. This is true for all
cases except those which combine Stage II and onboard controls. When
Stage II and onboard coincide at a vehicle refueling operation, the
onboard fill pipe seal does not allow vapors to enter the Stage II
system. The emission reductions for this operation are then associated
only with the onboard controls. Therefore, a separate analysis was
performed for Stage II emission reductions when combined with onboard
controls. Emission reductions for Stage II were attributed only to
that percent of the throughput (and therefore that percent of the
5-17
-------
I
» '
fe-
O-
1 =
to
00
as
o
HH
o.
o
1
o
o
m OCJ cor*. 0*0 oo iP^- 2^ . vm oo
PH CM »-* *—«
_
So oo oo oo oo oo S2 S 52
o oo oo entn oo S oo oo
"^ **^ ^,e** » *» •* •• * * »•
fr> ««< Vm U>O tf> ^ CM •-< cn r**
"
g g || |g || ' II §s §| II
VH m r-» ui v^co *r r«- coco io<«f' ID ID
I I §1 II §1 II §| §§ I I
cy\m torn u>to cr* CM couo ou) r* CM ,— « o
Sew oca cnm *r«a- co«5 r^-^ ocw ~S
tr» CM oo r-» a* to CM tn cou> mcsi ««- to^
§o oo oo oo oo oo oo oo
o oo oo oo oo oo oo oo
^> (\j vO CO W ^O ^^ f ^"* ^ *^ *"** *" ^' *" ®
coco com i«csj «-««-i ^ ^L ^^ cnr-». ^^
*-T CM* CM* m^CsT CO U) CM -H VD m
« , l l , ra CT
cu 4> o.
•** "c oS c Si'«*^iU3 •a^iL**^i"Z^i" Ui-5
J^ nj *^CLO o Q.V o o. a» **- •** o. S << o. £ ^ a. § o> QJ uj
o= ^"^ cxw *32 ouj < OJLU a: <-• c - o •«— o T- •*- cTifl 0*1—^- o *t— a) o<*» o :^ >•
« *o o >— Nat*JZcOf— 3e(/i^a«zzco o
I 1 f- 1 1 t. 1 1 3 ' '
)— GQ 3 •<
, , ;.: ,
CU
w
fo
tf»
0»
. ""
OD
•o
A3
0)
5
T3
«
ej
3
VI
0)
*O
• O
(A
1 8
u u-
HJ
H -g
« I i
cu *»
o j* •
1 *' I
HI 1 w
«*.. k. at
c: 4J
C3 O *
" -i S
t~ 01
«3 k C
U O
— e
** o *n
o
u
c: (/)
o « -o
T3 3 W
0 r- O
10 U .a
-------
emission reduction) in each year which was not controlled by onboard
systems. Furthermore, it was assumed that Stage II systems were
not replaced after one useful life of 15 years (two useful lives of
8 years for vacuum assist systems) because by the time this would
be required, the entire vehicle fleet would be controlled by onboard
systems. The combination of Stage II and onboard systems allows
for the more immediate emission reduction achievable by installed
Stage II systems while the onboard systems are gradually controlling
the vehicle fleet. Table 5-8 summarizes the cumulative and discounted
emission reductions associated with Stage II controls when combined
with onboard control options.
Using Tables 5-7 and 5'-8, the emission reductions associated
with the regulatory strategies can be determined. Table 5-9 summarizes
these emission reduction impacts. Emission reductions are presented
for gasoline vapors, benzene, EDB, and EDC.
As a further comparison between Stage II and onboard control
options, an analysis was performed to evaluate the emission reductions
which could be achieved considering in-use efficiencies rather than
theoretical efficiencies. In-use efficiencies take into account
tampering and deterioration which occur during actual use of the
equipment. Section 3.7.3 and Appendix D discusses in-use efficien-
cies in greater detail. Table 5-10 summarizes the emission reductions
for nationwide Stage II control options based upon in-use efficiencies
associated with annual inspections (86 percent) and minimal" enforce-
ment (56 percent). These emission reductions can be compared to the
nationwide emission reductions obtained from onboard controls,
considering in-use efficiencies (87 percent).
5.2 OTHER ENVIRONMENTAL IMPACTS
5.2.1 Hater Pollution Impacts
The overall impact on water resources is negligible. Only
refrigeration systems for bulk terminal control, which cool and
condense the vapors from the loading operation for liquid recovery,
create a potential water pollution impact. As the vapor-air mixture
collected at the loading rack is cooled, a small amount of gasoline-water
mixture is generated. The amount of water generated is dependent
5-19
-------
*
o
eC
O
ca
o
z
H- <
o
UJ
"Si
CO
y^
O
o
5E
UJ
CO
^^
t—»
S
^^J
H»4
UJ
•c
CO
El
o
U.
CO
ac
0
l_
0
o
UJ
Q£
_
Mb
O
I— 1
CO
CO
Sf?
UJ
*
CO
if)
UJ
_i
CO
h-
••^
.X
f
• tn
C
o
*J
1
I/I
c
o
I/I
1
UJ
1
vo
co
en
•-4
g
UJ
0
g
O)
LU
i
a.
z
vo
CO
en
1
N
C
CU
CO
£
U
^^
O-
z
CO
en
cu
5 u
— o
o o. ,
tn IB
IB >•
•°
i
0)
o.
'-
£>
1"^
,—
lfr
o
0 O O O
en vo to PO
o o o o
PO en to «s-
en vo mm
to en 10 ^-
^^
o o o o
CM CM CO O
W CM CM
§o o o
o o o
oi~T vr CM"
o o o o
- 0 0 0 O
P. P. P. P.
co o en —<
PO «r CM o
en vo «a- PO
o o o o
0 § gg
•> * * •»
CM CO «3" CO
vo vo po «r
in o vo «»•
•M 1-4
in en
tn o tn o
C •»- C 1-
O 4-> O 4-> U)
•^ O. •»- O. IB
*J B *• E CU
0.0) Cu CU U
0^ UJ tn oi UJ
X IB X «C
oi oi cu 01 eu 01 z
-O N U N
4- o-- «e 01- -o
3: z 1*0 z c-o oi
O 1 1 Z II U
— cu
4J r— i—
1C r» CU
=e «C
VO CM
*H *•*
S2
CM >4
CM CM
,
O CM
*•(
§o
u>S
«- T^i
o o
o o
°.°«
CM en
U) O
o o
o o
0 0
•t A
•» CM
CM VO
CM -H
in
in o
O •!->
T- a.
•U S
o. o>
O) UJ
X
Ol Ol
N
o ^
1 1
cu
en
IB
in
cu
IS
n
§•
0.
o.
IB
cu
*»
en
c
•5
to
.0
cu
IB
*->
0
cu
L.
IB
tn
CU
*-> •
IB r^
C 1
ULO
4-> CU
15 3
IB
IB e
O-t-
"c-o
O 1-
tn o
p2"e
o. o
i— i o
V4»
CU
en in
IB CU
<-> 3
IB
E >
o
L. CU
*4" «C
e o
0 *>
u ^«
T3 IB
0)
*
5-20
-------
.
UJ
as —
i— i o
_1 CM
00
CO CM
< 1
CD ^D
CO
SE: en
O fH
a.*""
CO
CO UJ
O C9
t-H UJ
|— (_•
3 OH
0 1—
UJ CO
^^
Z Of
00
1— 1 ^
lo<
2 =
UJ UJ
UJ
O CD
•-i Z
3 i— i
ZZ J"~"
O UJ
p ee
z=:
•
UJ
«J
CO
H-
L.
^
en
Z
tA
o
*>
u
3
T3
Q.
J
""-"••
£-8
O •*•
* s*
=3 l_
0> 4-»
X
X
»
7*
i
LS
1.
o
-i
3
U
»
L>
to
CO
rj
"
3
«i
o
SO
Cft
P*
i
=1
J
i— —
tOO OO OO OC3 OOO CM IO OO OO OO OO
••* CM en TV. ~* *r en «a- Tl ^* *•* ^i ^ en — • «o- men toen
»-«—* ••* CM CM *•* *•« «•* «•«
§*"*
oo oo oo oo oo o u> o oo oo oo oo
r*. co ^- co ~H T* !£* r-» in co -H »-* CM unr-*- ^N tn IOCM en co
^ »-< CM CM -H CM CO in CM CM CM CO »-H
CM CM tn to <-*co en co o r- o -M CM com — 4 ro ^co toen
-^-N — CM CM —...4 SS
'S ' " '
to co«a- i^^ om r^un r^oo CMCM «>to Oin tn— « enco
— * -H CMCM -*CM COLO CMCM CM CO *H
§o oo oo oo oo oo oo oo oo oo
O OO OO OO OOO OO OO OO OO OO
«>o coo oo oo ooo oo oo oo oo oo
• » •» *• mm **(«•) • * •.» •* mm mm
•HCM *f \o coin om coo * oo CMCM cMin ^"co vio
_« ^N *-4*^ CMcoen »MVI ~-t^« CMCM CMCM »H«H
o oo oo oo oo oo .00 oo CD o oo oo
o oo oo oo oo oo o oo oo oo oo oo
O f^ W O O O O O O O O O O O O O O O • O O O O
to vto «^ot tom r+* ^ co en » *o r^. en^n *HO into CM vo
r^ ^^*4 «»• in co m coo in vv ^ in en o en o in in
^* — « *er — < ^«
OO OO OO OO OO OO OO OO OO OO
oo oo oo oo oo o oo oo oo oo oo
O O O O m^ OO OOO OO OO OO OO OO
SO OO OO OO OO - OO OO OO OO OO
— * tOCM OO OO OOO OO OO OO OO OO
CM co toen —i in toco r* co o into co en in o en «r *-» *r
CMCM *-HCM CO«r * »**-4 *-4«M CO W CO ^^ *^cn om
co ^^ CMCO r*»en inco cor*. * ^.r*. r*^ r^ ^vo into coco
CM -^ .— 1 to -M «-t i— t ^H
*™ •
tO IA 4A I/I i— « to «O (A (A I/) M^tA
c c c c •— ' e= c*ec-oc«*j-oc: s- c
* O to O tn O(A OtA O in * O tn OtAUOtA t_ O CA rtj o co
.^-C • ^- C -^C f-C tt)-^EZ • ^E • —C 10-^-C »«0«*-C 0-^C
< *^ O < 4-» O •*-» O 4->O C04-» O *C 4J O «C 4J O O4^O CX 03 £X VI Qj CL (U CX QJ CL O tt) CX O (I) CX O) &
1XE 1XE X E x E x E t x E i X E X E > x E •« X E
3)01 4> d> CJ QJ A> 4> *O 4J CJ V (U O fl) *0 O O) ^) O> 4J CJ S
«^- O O*i—O fQ O) to f— O 0><0-^O O) ••— O 0) cn co c cntnc cntAC cntACZ cnvic o en o to c cnotAC OIAC cncncnc en en cz
CO CO«CCO CO. * .. *
0-t 1— * >~« > «— 1 1—4 I—I >< I_H
^
«
tn
IA C>
m •*-»
C (0
*° o
0) Q.
C (O
•— o>
0)
tA ^»
jo e
o> •»-
£ £
C T7
•^ (Q
•0 Ol
0) 7
fe =
^3 O
•^ cz
IA •»-
c
O •*->
0 0
O)
tA in
o o
5 a
c *
JZJ O
(A tA
r— C
S 5
i 2
u o»
CX
o o
= »
X d)
o «*-
CM O)
0 I-
CM
o>
S 0
i- .CZ
O) >
(A IA U
tA •*- Z3
•*» E ^
E eg cu
O> 1.
IA
at i- c
C 0) O
*^ ? *tn
O tA Vt
(A C —
ns03 js° *
5-21
-------
UJ
o
to
§
ex
U_ (/)
UJ
OO »-4
=C O
O Ul
a co
UJ
ce: >-^-
o:-a
s: o a
o H- *a
•-< «c s
CO _lr-
00 SP O
J-J O C
ac uj o
UJ OS O
UJ CO*"*
Q =
t-* 1—4
=£ UJ
O *£
IT)
UJ
i
i
i
*
-
IA
|
•b
cc
c
IA
' £
!
!
1
i
M
C
CD
i!
n 10
in
13
I
fc.
4-1
(O
o
10
1
tX
L
§
2
B
§
71
i
i
|
i
»
1
1
O O O O
•-H CM
gg 00
ro m
r- Cvj OO
— CM
cO£> °°
§1 i§
cu m csi eg
O o oo
o o o o
cb*-^ ir>*-^
Ol ^M CO ^^
§§ °°
o
S8
co en
O O o o
il
*-• e c
e o o
O) -^ ^ •*-
en IA 4-* *»
•0 C 0) (JO
4J O V) C 33
CO •*- C Cl -O ^3
•*•» O N O O
•Q Q.-*- C i- t_
^ E ** 0*
i-t t. O) O. COO NN
« X E C CO CO
O OJ ^~
cncotuj =so *r c*-)
• >
4j iv o QJ re oj *-H
V) «O W Z Di U VO CO
~ -C co
•-* X
X
5-22
-------
o
oo
h— I
rf
«C OO
Q. UJ
o a: o
f ^ a^^ *T?>
O UJ
•Z. CD i-<
O 2 O
•— 1 O I-H
H- U-
0 Q U_
Q <
UJ UJ
ce H-. oo
H^ 1H3
Z 1
O UJ 2
i— < <_S i— i
OO — i oo z
S 1-1
uj 32 a;
UJ UJ
• UJ Q
O 3 t-i
~-< t— oo
1 UJ =Z
LO CO O
0
UJ
CO
r— . un o L^) c\j-., «* -
.*-4- «-r —4 ou evj «-«
•
o o o o
csi «~H PO ^r ^3 r*^>
CTt CO Lf> CO CM f**-
*-4 t-l
.
O O O O
'r-. ' u> o uo oo ^tf-
-H ^H —4 OJ CSJ — <
§00 ooo oooo
oo ooo oooo
ro ro ,o , , o ro co PO p** r-. LO
ocn^fi LOCOCO air»-coio
. . ; . - . ,
_
OOO OO-O OOOO
ooo ooo oooo
cst COO co I-HO ^r-x *— icr»
*»« *.* ».*«*
f*1- OOCM «^ C^OJ. tn\jO CMCft
""CM LO^-PO <*CO^ r«fc
O'OO OOO OOOO
o oo o oo oo oo
ooo ooo oooo
o o o ooo •00*0*0'
o o -o o oo " ocr» oo
*J3^CTi CO *— 4 rO ^"VOfO LO
^^ *~* CM CM *— l T— t CM »-4 * CM
ooo ooo oooo
ooo ooo oooo
O OO O_OO O'O O"O
ooo ooo oooo
ooo ooo oooo
f"» . i— 4 PO ro vocTi cr> i— i ^-^-1
LOLOPO COr-"** ^OPOVOCM
• '
U^liO^O LO(JO^ COCOCMCNJ
01 COtT) cn'COLO S S . S! - Si
(tj
Ol CTi
y «^- .,_
i— cn i— a>
• £ ^ t^ ^~
"oc ccr ccz ooj O)O)
•^ O i— OOJtrtr— OQJ i— Q£3 rv* -s
C4-*o ^-> S QJ O C_) J-> E d> UOOJ (UOJ
0 °"£ r-U3(_>-^'^ ^030 -^-i— C£ »— Ctf
*J <1J flJ OJ =S CX-r- O CL (U OJ =3 CX-^ O OJt-.CX flj.f-.cL
s: UJ o^ cc-^-c: c^ cc-r-c OGJOJ> 3Cj
•— t a; j= c "^ UJJTC >•* ui ^>->uj^>.>uj
— 'Ml— »-H| 1 H- >— . i | -^j—,1 _,,
•i- o S-
OJOOOO 2:OO TOO O
GO 0
cn
^
^
OJ
E
*o
o
' W)
*c s-
o
•a
cn
cn • u
g OJ
S- U
ll ,^
u-
C OJ
o
•r- C
•P re
= cn
•a o
OJ 5-
i- O.
CZ Ul
O 4->
to OJ
li ^
-------
upon the relative humidity of the atmosphere. The mixture passes
through a gasoline-water separator, with the gasoline returning to
storage and the water being discharged. It is estimated that this will
produce only a negligible impact on water quality.
5.2.2 Solid Waste Impacts
None of the control techniques evaluated generate a solid waste as
a by-product of its operation. The only solid waste that may be gener-
ated under a worst case assumption would be the carbon used in the bulk
terminal vapor recovery systems and the onboard canister systems. EPA
has determined in a previous study that, even under the worst case
assumption that the carbon cannot be reused, the resultant solid waste
impact from bulk terminal controls would be negligible.3 If it is
assumed that the carbon in the onboard systems must be discarded after
its useful life, it is estimated that 3 pounds of carbon would be
discarded with every vehicle.4 This is negligible when compared to the
mass of solid waste generated by discarding the vehicles themselves.
5.2.3 Other Environmental Impacts
Other potential environmental impacts include noise, space
requirements, and availability of resources. The relative impacts of
the regulatory strategies on these environmental concerns is expected to
be insignificant. An EPA test showed that the noise level from terminal
vapor processing devices, which created significantly more noise to the
unprotected ear than any other system considered, was less than 70 db
at 7 meters from the noise source.5 The strategies cause only minimum
impacts due to space requirements and resources availability.
5.3 ENERGY IMPACTS
Energy impacts for the strategies are estimated in the form of
liters of gasoline saved. Energy savings were estimated from the
recovery credits assigned to the different industry sectors. Table 5-11
indicates the gasoline savings ratios for bulk terminals, bulk plants,
and service stations. Gasoline is recovered at terminals where carbon
absorption or refrigeration systems are used. The ratio of return for
terminals (1.1 liters of gasoline recovered/1000 liters of gasoline
transferred) takes into account the energy consumed by the vapor
processing systems. Gasoline is recovered, or not lost to evaporation,
at bulk plants where vapor recovery is used on outgoing loads. When
5-24
-------
Control
Option
TABLE 5-11. GASOLINE RECOVERY RATIOS
Net Energy Savings (Liters of Gasoline
Recovered/1000 Liters Transferred)
Bulk Terminalsa
Bulk Plants'3
Service Station (Stage II)C
Storage Tanks
1.1
0.6
1.5
d
Reference 6.
Table 7-6 and 7-7.
'Table 7-15.
Net Energy Savings = 60,250 liters/year/tank (Table 7-5)
5-25
-------
gasoline is pumped from storage to fill the trucks, vapors
are returned to the tank, thereby reducing evaporation and saving
gasoline. This same operation occurs at service stations when
Stage II vapor recovery returns vapors to the service station
underground storage tank.
Table 5-12 estimates the cumulative energy credits associated
with the regulatory strategies, in units of liters of gasoline
saved, from 1986-2020. No gasoline is recovered or saved under the
onboard control scenario, even though substantial emission reductions
are achieved. The gasoline vapors are absorbed in the carbon
canister and burned in the engine. No energy or recovery credit is
given to this operation because the added weight of the canister is
assumed to offset any added fuel economy (see Section 7.2.5.3 and
Appendix C).
5-26
-------
TABLE 5-12. ENERGY SAVINGS ASSOCIATED WITH
GASOLINE MARKETING REGULATORY STRATEGIES
Control Alternative
I. Baseline
II. Stage II-NA*
- Size Exemptions
- No Exemptions
III. Stage II-NA
- Size Exemptions
- No Exemptions
IV. Stage I
- Size Exemptions
- No Exemptions
V. Stage II
- Size Exemptions
- No Exemptions
VI. Stage I & Stage II
- Size Exemptions
- No Exemptions
VII. Onboard
VIII. Stage II-NA* & Onboard
- Size Exemptions
- No Exemptions
IX. Stage II-NA & Onboard
- Size Exemptions
- No Exemptions
X. Stage I & Onboard
- Size Exemptions
- No Exemptions
XI. Stage II-NA & Stage I & Onboard
- Size Exemptions
- No Exemptions
XII. Stage II & Onboard
- Size Exemptions
- No Exemptions
XIII. Stage I & Stage II & Onboard
- Size Exemptions
- No Exemptions
Gasoline Savings (1986-2020)
(109 Liters)
1.0
1.4
2.8
3.9
4.9
5.1
8.1
11.4
13.0
16.5
0
0.2
0.3
0.6
0.9
4.9
5.1
5.5
6.0
1.5
2.1
6.4
7.2
5-27
-------
5.4 REFERENCES
1. National Petroleum News. 1983 Factbook Issue. Mid-June 1983,
Volume 75, No. 7A. p. 80.
2. U.S. Environmental Protection Agency. Federal Register, Vol.-47,
.Number 210, October 19, 1982. p. 49329":
3. Bulk Gasoline Terminals - Background Information for Proposed
Standards. U.S. Environmental Protection Agency. Office of
Air Quality Planning and Standards. Research Triangle Park,
N.C. Publication No. EPA-450/3-80-038a. December 1980.
4. Cost Comparison for Stage II and On-Board Control of Refueling
Emissions. American Petroleum Institute. Washington, D.C.
January 1984. Appendix III, p. 2.
5. Betz Environmental Engineers, Incorporated. Gasoline Vapor
Recovery Efficiency Testing at Bulk Transfer Terminals Performed
at Pasco-Denver Products Termtnal. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. Contract No. 68-02-1407.
Project No. 76-GAS-17. September 1976. 97 p.
6. Reference 3, p. 7-8.
7. Memorandum from Gschwandtner, K., Pacific Environmental Services,
Inc. to Shedd, S.A., Environmental Protection Agency. March 16,
1984. Internal Floating - Roof Emission Factor.
5-28
-------
6.0 EXPOSURE/HEALTH-RISK ANALYSIS
This chapter outlines the methodology, assumptions, and results of
the health risk analysis for the gasoline marketing industry. Exposure
and risk estimates were developed for four pollutants emitted by the
industry: benzene (Bz), ethylene dibromide (EDB), ethylene dichloride
(EDO, and gasoline vapors (GV). These substances were chosen for the
analysis based on the availability of quantitative unit risk factors
for carcinogenicity. The availability of unit risk factors, derived
from human and animal research studies, permits the quantitative
evaluation of health risks.
Section 6.1 discusses the unit risk factors and types of health
hazards associated with each of the four pollutants. Section 6.2
outlines the methodology and assumptions used to estimate the annual
nationwide incidence (occurrences) of cancer expected to result from
the projected exposure and to estimate the lifetime risk (probability
of an incidence over a 70-year lifetime) that would result from a
realistically high exposure scenario. The incidence and lifetime risk
attributable to each of the four pollutants were estimated for each of
the following four source categories: bulk terminals, bulk plants,
service stations, and self-service refueling and then were projected for
the years from 1986-2020. Section 6.3 presents the cumulative incidence
over 35 years and the lifetime risk from 70 years exposure to expected
emissions from each source category under the implementation of each
regulatory strategy examined.
6.1 UNIT RISK FACTORS
The unit risk factor for an air pollutant is defined as the proba-
bility of getting cancer as a result of continuous exposure for a
lifetime (70 years) to a unit concentration of the agent. Derivation
of a unit risk factor permits quantitative estimates of the health
risks for exposed populations. Table 6-1 lists the four pollutants of
interest with their respective unit risk factors, and a brief health
effect summary.
6-1
-------
TABLE 6-1. UNIT RISK FACTOR SUMMARY
Compound
Unit
(probability of cancer
given lifetime exposure
to 1 ppm)
Health Effects
Summary
Comments
Gasoline Vapors
Plausible Upper Limit:b
Rat Studies
Mice Studies
3.5 x 10-3
2.1 x lO-3
Maximum Likelihood Estimates;b
Rat Studies
Mice Studies
Benzene
Ethylene
01bromide
Ethylene
01 chloride
2.0 x 10-3
1.4 x 10-3
2.2 x 10-2
4.2 x 10-1
2.8 x 10-2
Kidney tumors in
rats, liver tumors
in mice.
Human evidence of
leukeraogenicity.
Zymbal gland
tumors in rats;
lymphoid and other
cancers in mice.
Evidence of carci-
nogenicity in
animals by inhalation
and gavage. Rats:
nasal tumors; Mice:
liver tumors.
Evidence of carci-
nogenicity in animals.
Rats: ci rculatory
system, forestomach,
skin, and mammary
glands. Mice: liver,
lung, mammary gland
and uterus.
Gasoline test samples in
the animal studies were
completely volatilized,
therefore may not be
representative of ambient
exposures.
EPA listed and regulated
as a hazardous air
pollutant. IARCC:
Sufficient evidence to
support a causal associ-
ation between exposure
and cancer in humans.
EPA suspect human carci-
nogen; recent restrictions
on pesticidal uses.
EPA: Suspect human carcinogen.
Draft health assessment document
released for external review
March 1984.
aUn1t Risk Factor is the probability of a cancer incidence (occurrence) per 70-yr lifetime exposure
to 1 PPH.
In the case of gas vapors, the point estimate unit risk factor represents the slope of best fit
fit linear regression of the data through the origin. The plausible upper limit unit risk' factor
represents the maximizing of the slope (wwhich is the unit risk factor) while remaining within
the 95 percent confidence interval.
CIARC: International Agency for Research on Cancer.
6-2
-------
As noted in the table, benzene is the only pollutant of interest
for which there is sufficient evidence derived from human epidemiolog-
ical studies to support a causal association between exposure and
cancer. Although there is limited health evidence from studies of
occupational populations exposed to gasoline vapors in the gasoline
marketing system, the evidence is not sufficient for use in risk estimation,
The unit risk factors for gasoline vapors, ethylene dibromide, and
ethylene dichloride are based entirely upon animal studies.
6.1.1 Credibility of Risk Estimates
As with all unit risk estimates, these values were derived using
several standard assumptions in the absence of information to the
contrary. The major assumptions are:
1) The agent is a human carcinogen.
2) The linear dose-response relationship is plausible as a means
of estimating the risks associated with the small doses
typically occurring in the environment.
3) In the absence of human data on carcinogenicity:
a) Animal bioassays are appropriate for human risk estimation.
b) Humans are as sensitive as the most sensitive animal
species.
c) If the route of administration is not appropriate to the
human exposure situation (i.e., gavage), equal fractional
uptakes are assumed via the two exposure routes (i.e.,
inhalation or gavage).
For each of the agents, these general assumptions are discussed in the
following sections.
6.1.1.1 Benzene. An association between benzene exposure and
leukemia has been documented in several human studies of occupationally
exposed populations. Benzene has also been found to be carcinogenic in
both rats and mice by gavage and inhalation routes of exposure. The
benzene unit risk factor (the risk of cancer resulting from a 70-year
lifetime of exposure to a unit concentration) was derived from the
average of three occupational studies, assuming a linear dose-response
function. A unit ractor risk derived from the animal data is very
close to the value derived from the human studies thereby indicating a
similar dose-response relationship.
6-3
-------
6.1.1.2 Unleaded Gasoline. The evidence of carcinogenicity
comes primarily from the American Petroleum Institute chronic inhalation
study of unleaded gasoline vapor in rats and mice. The unit risk
estimates for each species based on a linear nonthreshold dose
extrapolation were derived from this study. Although API studied"
unleaded gasoline, other gasoline grades (e.g., leaded gasoline) are
expected to have as much carcinogenic potency.
6.1.1.3 Ethylene Dichloride (EDC). No human evidence of
carcinogenicity is available. The animal evidence consists of positive
responses at several sites in male rats and mice via gavage. The unit
risk for EDC inhalation was estimated by two separate methods: (1) a
direct estimation based on the EDC gavage study, assuming that the
absorption rate by inhalation is one-third that by the oral route; and
(2) an indirect estimation from the EDB inhalation study. The potencies
calculated from both approaches are similar.
6.1.1.4 Ethylene Pi bromide. No human evidence of carcinogenicity
is available. The animal evidence consists of positive responses in
mice, in both inhalation and gavage bioassays, as well as nasal cavity
tumors in rats following inhalation exposure. The unit risk was obtained
from the rat inhalation experiment using the linear dose-response
extrapolation procedure.
6.2 EXPOSURE AND RISK METHODOLOGY AND ASSUMPTIONS
This section briefly outlines the methodology and assumptions used
to estimate the concentrations of benzene, EDB, EDC, and gasoline vapors,
as well as their associated health risks, from each source category
to which the nation's population as a whole and to which selected
individuals subject to high exposures would be expected to be exposed.
Estimates were made and projected to 1986 through 2020 for exposures
due to emissions from each of the three source categories of bulk
terminals, bulk plants, and service stations. Both ambient exposures
to the public around service stations from all vehicle refueling and
individual exposures during self-service vehicle refueling were estimated.
(Although the emissions appear to be double counted, the exposures and
risks are not double counted because the service station estimate
considers the emissions dispersing to area residents off the station
premises while the self-service estimate considers the emissions near
6-4
-------
the source when the public comes to the station.) The emission sources
considered for each of the exposure categories are listed in Table 6-2.
Further details of the methodology and assumptions used to estimate the
various risks are given in the docket (Docket No. A-84-07).
The estimates of risk, in terms of individual lifetime risk from
high exposure and aggregate incidence, are applicable to the public in
the vicinity of gasoline marketing sources and those persons who refuel
their vehicles at self-service pumps. This analysis did not examine
the risk to workers from occupational exposure (e.g., terminal operators
and service station attendants). The lifetime risk from high exposure
for these workers is probably substantially higher than for the general
public. In addition, the estimates of aggregate incidence would be
higher if such worker populations were included in the analysis. Of
course, any controls to reduce gasoline marketing emissions would reduce
exposure for workers as well as for the general public.
6.2.1 General Assumptions
Several basic assumptions underlie the estimation and projection of
exposure and risk for the gasoline marketing industry. The same
assumptions were made as in the emissions and cost analyses, regarding
baseline control levels, gasoline throughputs for each source category
and gasoline type (decreasing until 2000 and constant thereafter),
exempted throughputs, emission factors, onboard control start-up and
phase-in, phase-in of Stage I and Stage II controls, and phase-out of
Stage II controls when in combination with onboard controls. Generally,
the risk estimates due to benzene were calculated for a base year. The
risk due to the other pollutants was then calculated by multiplying by
a factor incorporating the ratio of emissions (from either a storage
tank or a vehicle tank, since the vapor temperature and, thus, emissions
differ) and the ratio of unit risk factor to that of benzene (see
Table F-l in Appendix F). This ratioing technique is valid since the
other pollutants are emitted from the same sources only at different
emission rates and different associated unit risk factors. (The unit
risk factors resulting from a unit volume concentration (ppm) were
used for ratioing as they are thought to better represent the health
response.) In the calculation of EDB and EDC risks, the fraction of
gasoline which is leaded must also be considered, because EDB and EDC
6-5
-------
TABLE 6-2.
EMISSION SOURCES CONSIDERED
IN RISK ANALYSIS
Bulk Terminals
1. loading racks (including tank truck leakage)
2. storage tanks
3. vapor processor (controlled terminals only)
Bulk Plants
1. loading racks
2. storage tanks
Service Stations
1. underground storage tank vents
2. automobile refueling (tank displacement and spillage in vicinity
of gasoline pumps)
Self-Service
1. automobile refueling (tank displacement emitted from fill neck)
6-6
-------
are contained in and emitted by leaded gasoline only. The risk due to
each pollutant was then projected through the year 2020 in proportion
to the total (for Bz and 6V) or leaded (for EDB and EDO gasoline
consumption.
The gasoline vapor to which people are exposed in the environment
contains primarily the most volatile components in liquid gasoline,
whereas in the API animal experiments, an aerosol formed from all
components in liquid gasoline was inhaled. The identity of all the
carcinogenic components in gasoline has not been determined. Thus,
there is no way of knowing whether the potency of the gasoline vapor
to which humans are exposed, is equal to, more than, or less than
that of the gasoline aerosol used in animal studies. In the absence
of other data, it was assumed in this analysis that the unit risk
estimate based on total vaporization applies to all exposures resulting
from emissions from the several gasoline marketing source categories.
6.2.2 Incidence Analysis
Annual incidences (occurrences) of cancer expected to result
(after some unknown latency period) from estimated exposures to benzene,
EDB, EDC, and gasoline vapors were calculated for the various control
options examined in each industry segment. The estimation procedures
considered exposure levels and populations, existing controls, exempted
facilities, and phase-in and phase-out of additional control measures.
Estimates of incidence due to EDB, EDC, and gasoline vapors generally
were calculated from the benzene base year incidence numbers for all or
part of the source category (industry segment). Base year estimates of
incidences due to self-service exposures to EDB, EDC, and gasoline
vapors, however, were based on measured hydrocarbon concentrations.
incidences were projected from the base year estimates to the years
1986 through 2020 in proportion to the total or leaded (for EDB and
EDC) gasoline throughput for the source category.
6.2.2.1 Bulk Terminals and Bulk Plants. Since there are about
1,500 bulk terminals and 15,000 bulk plants in the United States
handling gasoline, limited resources would not allow modeling each
plant individually, even if data were available regarding exact location
and throughput. Therefore, a method was developed to assess the
nationwide incidence due to either bulk terminals or bulk plants.
6-7
-------
This method took into consideration the geographical distribution of
gasoline throughput and current controls and the varying population
densities around various plant types and sizes. This method is
outlined in Table 6-3 and the following discussion.
In general, several cities across the country were selected to
determine an average incidence for each bulk terminal or bulk plant
model plant. Bulk terminal and bulk plant model plant locations
were selected such that larger facilities were more often assumed to
be placed in larger cities. An exposure model was used to determine the
incidence due to emissions from each model plant in each of the cities.
The total nationwide incidence due to bulk terminals and bulk plants was
then estimated by: (1) calculating an average incidence for each model
plant, determined from the incidence in the selected cities, and (2)
multiplying this average incidence by the number of facilities in
each model plant category.
Ten specific localities were selected for each model plant by
considering:
(1) the proportion of bulk terminal throughput or bulk plant
throughput through each of the ten EPA/DOE Regions;
(2) the proportion of the regional and nationwide throughput for
each facility type that was controlled or uncontrolled at
baseline (see Tables B-7 and B-9);
(3) an assumed locality size distribution;
(4) as widespread as possible a geographical distribution of
cities representing each model plant size (in order to achieve
a composite representative of nationwide climatological
conditions).
For each State and region, the bulk terminal throughput was assumed to
be equivalent to the total throughput. The bulk plant throughput for
each State was estimated based on the percentage of total throughput
that passed through bulk plants according to the 1977 Census (see
Sections B.2.1 and B.2.3). In most instances the localities were
selected from cities where bulk terminals were known to be located or
where bulk terminals or bulk plants were likely to be located. Because
many of the selected localities reasonably can be assumed to have
bulk terminals and/or or bulk plants of varying sizes, it was only
6-8
-------
For
TABLE 6-3
BULK TERMINAL AND BULK PLANT
ANNUAL INCIDENCE ANALYSIS
Assumed a distribution of localities by size for each of the 4 bulk
terminal and 4 bulk plant model plants. For example, the largest model
plant of (1,000,000 gal/day) was assumed to be located only in cities
with populations greater than 500,000.
Ten specific localities were selected to represent each model plant by
considering:
1. the proportion of total bulk terminal or bulk plant throughput
through each of the ten EPA/DOE Regions.
2. the proportion of the regional and nationwide bulk terminal or
bulk plant throughput that was controlled or uncontrolled at
baseline.
the assumed distribution of locality size for each model plant.
known or likely locations of bulk terminals and bulk plants.
3
4
example
•i
if 10% of the national gasoline throughput is in Region V, then
10% of the localities were chosen from Region V;
2. if 80% of the gasoline throughput is within controlled areas,
then 80% of the localities were selected in controlled areas;
3. if 30% of the plants represented by bulk terminal model plant #2
are assumed to be in localities with populations greater than
500,000, then 30% of the localities for model plant #2 had
populations greater than 500,000.
4. if possible, a locality known to have a bulk terminal or plant
and that met the criteria of region, control status, and
population range, was selected.
Specific coordinates for a bulk terminal or plant in a locality were
chosen using features on U.S.G.S. topographical maps, in the following
order of preference:
1. addresses of known facilities,
2. marked gasoline tanks,
3. unmarked tanks,
4. pipelines,
5. transport (major highway or railroad) routes,
6. commercial/industrial areas.
Human exposure model (HEM) was run for each locality and for each emission
source (see Table 6-2), assuming different release characteristics and a
surrogate benzene emission rate (so the same HEM run could be used for
several model plants with differing emission rates).
Average benzene incidence calculated for control 1ed and uncontrolled
plants under .each control option for each model plant. Used proportion
of throughput controlled, distribution of bulk terminals or bulk plants
among model plant sizes, and assumed total number of bulk terminals and
bulk plants to calculate total nationwide incidence due to benzene
emitted from bulk terminals or bulk plants in a base year.
Calculated incidence due to other pollutants using appropriate ratios
of emissions and risk.
Projected base year incidence into future based on gasoline consumption
(except for one type of bulk terminal emissions proportional to the
number of tanks).
6-9
-------
necessary to use a total of 41 actual localities across the country to
provide 10 locations for the analysis for each of the 8 model plants.
Specific coordinates for the facility in a locality were chosen
using features on U.S. Geological Survey topographical maps. In order
of preference the features were: (1) addresses of known facilities;
(2) marked gasoline tanks; (3) unmarked tanks; (4) pipelines; (5)
transport (major highway or railroad) routes; and (6) commercial/
industrial areas.
The System Applications Human Exposure and Risk (SHEAR) version of
EPA's Human Exposure Model (HEM)1 was used to estimate annual incidence
from benzene (the model incorporates the unit risk factor) for a surro-
gate emission rate of 100 g/s from loading racks and storage tanks at
both bulk terminals and plants and from vapor processors at controlled
bulk terminals. (The surrogate emission rate was used so that a single
HEM run could be used readily for different model plants by rat'ioing
the actual emission rate to the surrogate emission rate.) Each source
and facility type combination was assumed to have different release
heights, stack (or release) diameters, and initial dispersion parameters.
Uncontrolled and controlled emission rates were calculated for each
emission source at each model plant. (For the bulk plant size exemption
control option, the smallest bulk plant model plant was left uncontrolled.)
The average incidence due to benzene from each controlled or
uncontrolled model plant was then calculated based on the control option
and whether the locality is in an area that is already controlled. The
number of facilities represented by each model plant was determined
from facility size distributions given in Sections 4.1.1 and 4.1.3.
Thus, the base year nationwide incidence due to benzene could be
calculated (by multiplying the average incidence by the emission
rate and the number of facilities for uncontrolled and controlled
facilities of each model plant size, and summing for each facility type)
for each of the bulk terminal and bulk plant model plants.
The incidences due to the other three pollutants (EDB, EDC, and GV)
were than calculated using their respective ratios (see Table F-l)
of emissions and risk relative to those of benzene and using the leaded
gas fraction for EDB and EDC. Incidences were projected to 2000 and
assumed constant thereafter based on the total (for Bz and GV) or
6-10
-------
leaded (for EDB and EDO gasoline throughput for bulk plants (see
Tables 5-5 and 5-6). The same projection method was used for all
emission sources at bulk terminals except for some types of storage tank
emissions. The AP-42 emission factors for these types of emissions are
proportional to the number of tanks (containing all types of gasoline
for Bz and GV or containing only leaded gasoline for EDB and EDO,
rather than the gasoline throughput.
6.2.2.2 Service Stations. There are about 400,000 service stations
in the country. Because even a single metropolitan area can have a
large number of stations, only a fraction of which could be located and
for which throughput data could be found, even a case study analysis
using actual service stations is infeasible. Therefore, an area source
approach was used to estimate service station incidence. It was assumed
that the population in an area was exposed to a uniform concentration
of emissions from service stations in the area. Service statio'n
emissions were determined by: (1) assuming all gasoline consumption
(less 3 percent for agricultural use) in an area passed through a
service station, and (2) using service station emission factors
to estimate the emission rates. The uniform concentration for the
selected area was then computed by the exposure model, assuming that
all service station emissions were uniformly distributed across the
land area.
The country was divided into seven ranges of population within
multiple or single county areas (primarily standard metropolitan
statistical areas (SMSA's) or counties outside of SMSA's. Four to
seven sample areas in each population range were randomly selected to
represent the population in all the areas in the range. (The population
densities of the selected areas as well as the populations were checked
to ensure that the range was well represented.) A total of 35 areas
were chosen to represent the entire nation using the method presented
in Table 6-4.
The HEM-SHEAR model area source routine4 was run to provide the
ambient benzene concentration that would result in each sample area from
a uniform area-wide emission rate. Uncontrolled emissions from service
stations in each sample area were calculated for both inloading
gasoline to underground storage tanks and vehicle refueling (full
6-11
-------
TABLE 6-4.
SERVICE STATION
ANNUAL INCIDENCE ANALYSIS
A. Divided all areas (primarily standard metropolitan statistical areas
(SMSA's) or counties outside of SMSA's) in nation into 7 population
ranges.
B. Randomly selected (from 4 to 7) areas to represent the population io all
the areas in each range (a total of 35 areas to represent the nation).
C. Used gasoline throughput and land area for component counties of each
area selected in B above to develop area source emission rates for
inloading and vehicle refueling, assuming emissions are released
uniformly over the sample area.
D. Ran Human Exposure Model for a unit area source emission rate (1.0 kg/
km2-yr) from each sample locality to estimate resulting ambient
concentration.
E. Calculated estimate of incidence due to uncontrolled inloading and
uncontrolled vehicle refueling for each sample area using the area-
specific ambient concentration, emission rate, and population.
F. Assigned baseline control levels, appropriate population range, and
current attainment status to each county in the country. Then estimated
fraction of population in each population range at a given control
(emission) level under each control option.
G. Estimated base year benzene incidence from both inloading and vehicle
refueling under each control option for each population range considering;
1. total incidence due to uncontrolled emissions from all sample
areas in the population range;
2. the fraction of total population in the population range that
is within the sample areas;
3. the population-weighted emission level summed over all control
levels under the particular control option.
H. Estimated national incidence due to benzene in a base year by taking the
sum of the incidences for each population range.
I. Calculated incidence due to other pollutants using appropriate ratios of
emissions and risks.
J. Projected base year incidence into future based on gasoline consumption.
6-12
-------
service and self-service). The base year on-highway gasoline through-
put from the National Emissions Data System (NEDS) for the applicable
counties was used to estimate emissions. Emissions from inloading and
refueling were segregated because the release parameters and the ratios
of emissions of other pollutants to benzene emissions vary with the
differing tank temperatures. Annual incidence from service-stations for
each sample area in a base year was then calculated by: (1) assuming
that all service station uncontrolled emissions from inloading and
vehicle refueling were uniformly spread over the total land area given
in the 1980 census^; (2) using the unit risk factor for benzene;
(3) using the 1980 census population for the counties comprising the
area^; and (4) using the uniform ambient concentration resulting from
an emission rate of 1.0 kg/km2-yr in the area estimated by the HEM-SHEAR
model. A unit emission rate was used, as with terminals and bulk plants,
to allow the flexibility of ratioing the model results for controlled
and uncontrolled emission rates.
,The current control level, and the current attainment status (for
assessment of nonattainroent area options) was assigned to each county
in the country and the population of each county. The fraction of the
population in each population range that was at a given control level
under each service station control option was then estimated. For the
control options with size exemptions, the percentage of the population
that would be exposed to uncontrolled service stations was assumed to
be equivalent to the percentage of throughput through size-exempted
facilities calculated from the data on service station number of facili-
ties, throughput, and model plants discussed in Section 4.1.5.
The base year incidences due to public exposures to ambient benzene
emitted from both gasoline inloading and vehicle refueling under each
control option were then calculated. The incidence for each population
range was calculated by the procedures noted in Table 6-4. The incidences
for each population range were then summed to calculate the total
national incidence. The resulting base year benzene incidences for
both gasoline inloading and vehicle refueling under each control option,
when fully implemented, were multiplied by appropriate gasoline throughput
fractions, emission ratios, and risk ratios to estimate risk due to
EDB, EDC, and gasoline vapors and to project risks from 1986 to 2020,
considering phase-in and phase-out of controls with time.
6-13
-------
6.2.2.3 Self-service Vehicle Refueling. The self-service vehicle
refueling analysis assesses the risks to the general public resulting
from exposure of individuals to high concentrations near the emission
point (tank fill neck) within the service station during self-service
refueling. These individual self-service risks are in addition to the
risks resulting from exposure of area residents to much lower ambient
concentrations from dispersion of emissions during both self-service
and full-service refueling, which were assessed by the service station
incidence analysis. Because of the much higher exposure concentrations
involved, the self-service refueling incidence was found to contribute
about 80 percent of the total nationwide incidence.
The self-service vehicle refueling incidence analysis was based on
actual concentrations measured in the region of the face of the person
filling the tank. Exposure data were taken from an API study6 that
measured personnel exposures during multiple (4 to 5) fillings-of either
unleaded, leaded, or premium gasoline at 13 service stations in 6 cities.
Samples for the study were collected in charcoal tubes (MSA-type) using
battery-operated pumps and analyzed using gas chromatography and flame
ionization. Benzene, total hydrocarbons (measured as n-hexane), and
eight other compounds were measured.
For the incidence analysis, an average benzene exposure calculated
from the API results was used for each of two fuel types: leaded and
unleaded (a weighted exposure for premium and regular unleaded was
calculated to represent the entire study period). The average benzene
exposures during refueling were calculated to be 0.96 ppm for unleaded
and 1.46 ppm for leaded. Gasoline vapor exposure concentrations in
parts per million were also calculated based on the measured value of
mg/m3 for total hydrocarbons, the individual station ambient temperatures
from the API study, and the value for the molecular weight of gasoline
vapors of 66, used in AP-42.7 The calculated gasoline vapor exposures
are 58.8 ppm for unleaded and 72.4 ppm for leaded gasoline. The assumptions
and calculations for incidence from self-service refueling are outlined
in Table 6-5.
Several assumptions had to be made regarding self-service operations
so that the incidence could be computed. First, the time duration of
exposure had to be assumed. The duration of exposure is dependent upon
6-14
-------
TABLE 6-5
SELF-SERVICE
INCIDENCE ANALYSIS
A. Calculated average benzene and gasoline vapor exposure concentrations during
vehicle refueling for unleaded (0.96 and 58.8 ppm, respectively) and leaded
(1.46 and 72.4 ppm, respectively) gasoline from results of monitoring study of
attendents during multiple fillings at a number of service stations.
B. Calculated average annual risk due to exposure, e.g., for benzene:
(2.2 x 10-2 incidences/ppm - 70-yr. lifetime) / [70 yr/(70 yr. lifetime)]
= 3.14 x 10~4 incidences/(ppm - yr) •
C. Assumed 8 gpm average pumping rate and one person pumping; therefore, one
minute of exposure resulting from pumping each 8 gallon unleaded or leaded
gasoline, e.g.,
minutes of exposure to 0.96 ppm = (national annual unleaoea gasoline
througnput)/ 8 gal/mi n.
D. 'Converted to years of exposure resulting from leaded or unleaded gasoline:
years of exposure = (minutes of exposure) / (60 min/hr x 24 hr/d x 365 d/yr)
E. Assumed 70% of all service station throughput was self-service, so that
using equations developed in C and D above,
years of self service exposure = 0.70 (national annual gasoline throughput)
8 gal/mi n x 60 min hr x 24 hr/d x 365 d/yr
(national annual gasoline throughput)
= 6,006,857.1
F. Calculated annual incidence resulting from self-service refueling during a
base year, assuming a linear dose-response relationship, e.g., if all gasoline
were unleaded:
annual incidence = 3.14 x 10*4 incidences x 0.96 ppm x
(ppm-yr)
(national base year unleaded gasoline throughput)
6,006,857.1
= 5.02 x 10"11 (national base year unleaded gasoline throughput)
G. Calculated annual incidence due to benzene and gasoline vapors emitted during
all self-service refueling based on:
1. projections of onboard-controlled (unleaded),unleaded gasoline not
controlled with onboard, and leaded gasoline (not controlled with onboard)
in a given year.
2. fraction of throughput controlled with Stage I! under each control option
in a given year;
3. estimated fraction of uncontrolled refueling emissions and resultant
exposures still emitted out of the tank fill neck with Stage II
(e.g., 90 percent of 5 percent theoretical emissions after control)
and Onboard (50 percent of 2 percent emissions after control) controls.
H. Calculated annual incidences due to other pollutants using appropriate ratios
of emissions and risk (see Table F-l), assuming EDB and EDC exposures to be
proportional to total gasoline vapor exposures.
Note: The impact of self-service refueling emissions on the user population is
independent of population under the assumptions made.
6-15
-------
the pumping rate assumed for each self-service fillup. Gasoline" pumps
operate typically within the range of 8 gallons to 12 gallons per
minute. The lower value of the range (8 gal/min) was assumed as the
pumping rate because a typical fill up is not conducted at the maximum
pumping rate for the entire fillup period (e.g., "topping off" to get
an even dollar figure or reduced pumping rates toward the end for
prepaid self-service fillups). Therefore, it was assumed that 8 gal/min
represented a better basis for estimating total exposure duration.
It was further assumed that only one person would be exposed to the
self-service concentrations at each fill up. Finally, the proportion of
self-service fillups was assumed constant at the level occurring in the
base year (70 percent of consumption was through self-service operations
in 19828).
Since a nonthreshold, linear dose-response model is the basis for
the unit risk factor, any exposure (no matter how small) is assumed to
result in some risk of cancer. The risks across the exposed population
are summed to determine the total cancer incidence expected. For self-
service, wherein some person is always pumping fuel, the total annual
incidence is directly proportional to annual self-service gasoline
throughput. Thus, knowing the leaded and unleaded gasoline throughput,
pumping rate, and pollutant concentrations, the incidences due to
benzene and gasoline vapor uncontrolled emissions in the base (1982)
and subsequent years were calculated.
The incidences for the study years (1986-2020) and control options
were then calculated. These incidences were based on the projected
amounts of gasoline pumped to onboard-controlled vehicles (all of which
will use unleaded gasoline) and of leaded or unleaded gasoline pumped
to vehicles not controlled by onboard, either through stations with or
without Stage II controls. For onboard, one-half of the theoretical
emissions after control were assumed to be emitted from the fill neck.
The other half is assumed to be emitted from the carbon canister. For
Stage II, nine-tenths of the theoretical emissions after control were
assumed to be emitted from the fill neck and one-tenth from the underground
tank vent. For in-use incidence calculations, all vehicles with tampered
onboard controls are considered totally uncontrolled and assumed to use
one-half unleaded and one-half leaded gasoline. For Stage II in-use,
6-16
-------
the. emissions from the underground tank vent are considered the same
as for the theoretical case. All of the additional in-use emissions
are assumed to be emitted from the fill neck. Onboard and Stage II
controls were phased in and out according to the control option in
proportion the gasoline throughput.
6.2.2.4 Regulatory Strategies. The total incidence due to any
pollutant under any given regulatory strategy was calculated by summing
the annual or cumulative incidences for the appropriate control options
for each source category. (This is the same approach as used for
emission calculations in Chapter 5.0). For example, the total incidence
under the baseline regulatory strategy is comprised of the baseline
incidence for each of the source categories. In contrast, the total
incidence under the Stage II - All Nonattaiment Areas (Stage II - NA)
strategy is comprised of the baseline incidence for bulk terminals and
bulk plants and the Stage II - NA incidence for service stations and
self-service.
6.2.2.5 Automobi1e Operations. The incidence due to benzene
emitted during the operation of automobiles was also estimated as a
point of reference for the gasoline marketing evaluation and to assess
the total effect on incidence of reducing the benzene content in
gasoline and, when onboard controls are used, of controlling
evaporative emissions thought to be escaping current evaporative controls.
Benzene emission rates and vehicle miles traveled (VMT) for the various
types of gasoline-powered on-highway vehicles based on the Mobile 2
model were obtained.1-0 Composite emission rates were calculated that
included both evaporative (from fuel tank and carburetor) and exhaust
(tailpipe) emissions or only evaporative emissions from all gasoline-
powered vehicles (light-duty vehicles, light-duty trucks 1 and 2,
heavy-duty gasoline trucks, and motorcycles). An incidence estimation
approach similar to that for service stations was employed, i.e., an
approach using uniform area source emission rates in each of. a number
of randomly selected sample areas. The same sample areas, SHEAR-calculated
ambient concentrations, and 1980 Census populations and land areas were
used. In addition, the on-highway gasoline-powered vehicle miles
traveled from the National Emissions Data. System (NEDS) were used to
calculate the uniform emission rates, and subsequently, the incidence
6-17
-------
for each sample area. Annual benzene incidences for 1986 through 2020
were projected from the base year incidences in proportion to the
projected vehicle miles traveled. The VMT projections were taken or
extrapolated from DOE's Highway Fuel Consumption Model 10,11 The Highway
Fuel Consumption model output contained VMT values for 1986 through
1990, 1995, and 2000. After 2000, YMT was assumed to be constant.
Benzene in exhaust emissions were assumed constant regardless of
the gasoline benzene content, based on a study using four carsl2. The
study results show that reductions of about 60 or 80 percent from current
levels of benzene fuel content (about 2 percent) would not be likely to
reduce benzene exhaust emissions. Presumably, benzene is formed during
and after combustion by cracking larger hydrocarbon molecules and by
reforming from smaller molecules. Evaporative emissions of benzene
were assumed to be reduced in proportion to the reduction of benzene in
gasoline.
Evaporative emissions can be further categorized as (1) emissions
that are being controlled by current evaporative controls and (2)
emissions that are not being captured by current evaporative controls.
A preliminary estimate of the average evaporative emissions that are
currently escaping capture of 0.13 grams of benzene per vehicle mile
traveled was used to estimate the total incidence attributable to
vehicle operations if these additional evaporative emissions are
considered. These additional evaporative emissions would be reduced
proportionately if the benzene content of gasoline were reduced and
would be controlled by the larger canister that would be required for
onboard control of vehicle refueling. Benzene reduction in gasoline
reduces only the benzene contained in these additional evaporative
emissions, without affecting the other pollutants. Onboard controls,
however, were assumed to control all of the additional evaporative
emissions, including other pollutants.
6.2.3 Lifetime Risk Analysis
The lifetime risk analysis estimated the probability that an
individual subjected to high exposure levels throughout a 70-year
lifetime would result in a cancer incidence. The term "lifetime risk
from high exposure" is conceptually similar to the term "maximum lifetime
risk" which has been presented in other EPA documents, including those
6-18
-------
on benzene sources regulated or considered for regulation under
Section 112 of the Clean Air Act. The term "lifetime risk from high
exposure" rather than "maximum lifetime risk" is used in presenting the
risk calculations for the gasoline marketing study because EPA is less
certain in this case that the assumptions used result in the maximum
exposure to any single person or group. The Industrial Source
Complex (ISC) dispersion model13, capable of estimating individual and
combined contributions to ambient concentrations at a number of receptor
points from multiple emission sources, was used. The ISC model calculated
annual concentrations of benzene, gasoline vapors, EDB, and EDC at receptors
in the vicinity of a bulk terminal complex, a bulk plant complex, and a
service station complex. The following three sections describe the methods
used to calculate lifetime risk from high exposures attributable to each
of these three industry segments. An additional section describes the metho
used to calculate the lifetime risk from individual self-service fillings.
6.2.3.1 Bulk Terminals. Table 6-6 outlines the lifetime risk
analyses for both bulk terminals and bulk plants. Terminals are
typically clustered together in a location either at a point along a
pipeline or river. Therefore, a complex of terminals was used to
estimate the lifetime risk instead of a single terminal facility.
Figure 6-1 shows the hypothetical layout of the bulk terminal complex
used as input to the ISC model. The layout was based on the apparent
centers of individual terminals at a known bulk terminal complex in an
attainment area, shown on a topographical map. Six bulk terminals were
assumed to be in the complex, including at least one of each of the
four model terminals of various sizes with gasoline throughputs ranging
from 100,000 gal/day to 1,000,000 gal/day (for a total of 18 loading
racks, 27 storage tanks, and 6 vapor processors in the complex). The
physical dimensions of each source (release height, location, initial
dispersion parameters, etc.) represent the dimensions of typical sources
within a bulk terminal with throughput comparable to the model terminals.
The ISC model was executed with varying emission rates (based on controls
and estimates of unleaded, leaded, and premium gasoline throughput) for
each of the years 1986, 1990, 1995, and 2000 to obtain predicted concen-
trations at each of the receptors shown in Figure 6-1. The model was
6-19
-------
TABLE 6-6.
BULK TERMINAL AND BULK PLANT
LIFETIME RISK ANALYSIS
A. Used Industrial Source Complex (ISC) dispersion model to estimate annual -
average ambient concentrations due to combined effects of the various
emission sources (see Table 6-2), assuming different dispersion parameters
for each emission source at each size model plant.
B. Ran model using sets of meteorological data from several different
locations across the country (selected because of known bulk terminal
locations or suggested high resultant concentrations and because of
proximity of likely terminal location to residential areas).
C. Modelled a complex consisting of a number of bulk terminals (6) or bulk
plants (4) of the various model plant sizes, both for controlled and
uncontrolled complexes (since the receptor with the highest total concen-
tration resulting from all emission sources could change with control).
Configuration of facilities within complexes was based on actual or
likely arrangements.
D. Used modelling results for the years 1986, 1990, 1995, and 2000 to
estimate total risk over a 70-yr. lifetime.
E. Calculated incidence due to other pollutants using appropriate ratios of
emissions and risk.
6-20
-------
A A A
. 2
A A •
A
A .A
A A A A
3
A A A A A A A
6
A A A A A
A
A 4
* A A
AAAAAAAAAA
• *'
250 METERS
SCALE
A A A A
Figure 6-1. Map of Bulk Terminal Complex
LEGEND
^ TANKS
• VAPOR RECOVERY UNIT
• LOADING RACK
* RECEPTOR
6-21
-------
also executed with three different sets of meteorological data represent-
ing various parts of the United States; however, the maximum concentra-
tion always resulted from the same set of meteorological data. The
maximum concentration predicted in each of the years 1986, 1990, 1995,
and 2000 was then used along with the benzene unit risk factor to
calculate the maximum lifetime risk for benzene (both controlled and
uncontrolled). The maximum lifetime risk for gasoline vapors was then
calculated by using the unit risk factor and emission rate ratios.
Because only storage tanks containing leaded gasoline emit
EDB and EDC, and leaded gasoline has a lower and rapidly decreasing
(with time) throughput, a different set of storage tank emission rates
was used with the ISC model to obtain predicted concentrations for EDB
and EDC. The maximum concentration predicted in each of the years
1986, 1990, 1995, and 2000 was then used along with the EDB or EDC unit
risk factor to calculate the maximum lifetime risk for EDB or EDC.
6.2.3.2 Bulk Plants. The lifetime risk analysis for bulk" plants
is outlined in Table 6-6 (p. 6-20). Figure 6-2 shows the hypothetical
layout of the bulk plant complex used as input to the ISC model. The
configuration shown was selected to represent a typical complex of bulk
plants, all located in one part of a metropolitan area. The bulk
plants shown include each of the four model plant sizes with gasoline
throughputs ranging from 3,000 gal/day to 17,000 gal/day. (Each of the
bulk plants in the complex was assumed to have one loading rack and 3
storage tanks for gasoline.) The physical dimensions of the sources
within the complex are representative of typical sources within a bulk
plant. The ISC model was executed with varying emission rates (based
on control and gasoline throughout) for each of the years 1986, 1990,
1995, and 2000 to obtain predicted concentrations at an array of
receptors. The model was also executed with four different sets of
meteorological data representing various parts of the United States.
The maximum concentration always resulted from the same set of meteoro-
logical data. The maximum concentration predicted in each of the
years 1986, 1990, 1995, and 2000 was then used along with the benzene
unit risk factor to calculate the maximum lifetime risk for benzene
(both controlled and uncontrolled). The maximum lifetime risk for
gasoline vapors, EDB, and EDC was then calculated by using the relative
6-22
-------
s 1
£ £ 3
S3 «C O
•4 4
£
UJ
§
X
-------
unit risk factors and emission,rates, and in the case of EDB and EDC
the leaded gasoline consumption rate. All bulk plants were assumed to
emit EDB and EDC from leaded gasoline at a rate proportional to the
leaded gasoline throughput.
6.2.3.3 Service Stations. The methodology of the lifetime risk
analysis for service stations is presented in Table 6-7. Figure 6-3
shows the hypothetical layout of the service station complex used as
input to the ISC model. This complex configuration was developed to
represent a grouping of service stations at an urban exit from an
interstate highway. The complex was assumed to be comprised of the
eight service stations shown, which include at least one of each of the
five model stations with gasoline throughputs ranging from 5,000 gal/month
to 185,000 gal/month. The entire complex was comprised of 14 refueling
islands and 8 underground storage tank vents. The physical dimensions
of the sources within the complex are representative of typical" sources
at service stations. The ISC model was executed with varying model
plant-specific uncontrolled emission rates (based on baseline throughput)
for each of the years 1986, 1990, 1995, and 2000. The model was also
executed with three different sets of meteorological data representing
various parts of the United States. The maximum concentrations again
always resulted from the same set of meteorological data. The maximum
baseline concentration predicted in each of the years 1986, 1990, 1995,
and 2000 was then used to calculate the maximum concentration for each
regulatory strategy in each of these years (including the phase-in of
specific alternatives). These maximum concentrations were then used
along with the unit risk factor to calculate the maximum lifetime risk
for benzene. The maximum lifetime risk for gasoline vapors, EDB, and
EDC was then calculated by using relative unit risk factors and emission
rates and in the case of EDB and EDC the leaded gasoline consumption
rate. All service stations were assumed to emit EDB and EDC from
leaded gasoline at a rate proportional to the leaded gasoline throughput.
6.2.3.4 Self-service Vehicle Refueling. The baseline maximum
lifetime risk due to individual exposure at self-service filling is a
product of the concentration to which the individual is exposed
(obtained from a report by API6), the unit risk factor (for either
benzene, gasoline vapors, EDB or EDC), and the estimated time of exposure
6-24
-------
TABLE 6-7.
SERVICE STATION
LIFETIME RISK ANALYSIS
A. Used Industrial Source Complex (ISC) dispersion model to estimate
annual-average ambient concentrations due to both underground storage
tank unloading and vehicle refueling, assuming different dispersion
parameters for each emission source at each size model plant.
B. Ran model using three sets of meteorological data from different
locations across the country (same locations as for bulk terminals and
bulk plants).
C. Modelled a complex of eight service stations of the various model
plant sizes based on a typical exit from a major highway. Uncontrolled
emissions from both inloading and vehicle refueling were modelled (since
the receptor with the highest total concentration resulting from all
emission sources could change with the control option).
D. Used modelling results for the years 1986, 1990, 1995, and 2000 to
estimate total risk over a 70-yr. lifetime.
E. Calculated incidence due to other pollutants using appropriate ratios
of emissions and risk.
6-25
-------
toll'
• •
• •
g.
§
a-
o ce.
o
S (-O »-H t_J
UJ «t =1 LU
> d ea o;
. • B -
~=\
— o
— o
a.
I
o
c
o
•r-
•!-»
1C
cu
o
-------
per individual per lifetime. For this analysis, it was assumed that
high exposure equated to a person conducting self-service fillups
totaling 40 gallons each week (e.g., a traveling salesman). The
maximum lifetime risk from individual exposure for benzene gasoline
vapors, EDB and EDC under the different regulatory strategies is obtained
by proportioning the uncontrolled lifetime risk in accordance with the
fraction emitted out of the vehicle fill neck as discussed in Section
6.2.2.3. The calculation procedure and assumptions used are outlined
in Table 6-8.
6.3 PRESENTATION OF RISK ESTIMATES FOR REGULATORY STRATEGY
This section presents results of the health-risk analyses. Both
the cumulative incidence and lifetime risk from high exposure are given
for each pollutant from each source category under each regulatory
strategy. The health-risks from EDB and EDC are reported although they
are much smaller than those from benzene and gasoline vapors. The
lower risks from EDB and EDC are due to the lower pollutant contents in
gasoline, which result in lower emission rates, and to the dependence
of EDB and EDC emissions on only the leaded gasoline throughput, which
is expected to decrease significantly. Appendix F presents annual
incidences for each source category and control option as well as risks
from high exposure in 1986, 1990, 1995, and 2000.
The cumulative incidences from each source category and the total
from all source categories are presented in Tables 6-9 and 6-10 for
benzene and gasoline vapors, in Table 6-11 for EDB and EDC, and in
Table 6-12 for benzene and gasoline vapors from vehicle operations.
The cumulative incidences are presented for all regulatory strategies
(discussed in Section 4.3.2) both with and without size exemptions.
For example, the baseline (no additional controls) estimates of cumulative
incidence due to benzene over the 35-year study period (1986-2020) are
2.4 from bulk terminals, 1.2 from bulk plants, 6.5 from service stations
(due to ambient concentrations in the area), and 113 from self-service
(due to on-site concentrations during refueling by the public, i.e.,
not by paid station attendants). The total cumulative incidence due to
benzene emitted from all source categories is 123. The reduction in
incidence from baseline levels is also given for the strategies assessing
additional controls.
6-27
-------
TABLE 6-8
. SELF-SERVICE
LIFETIME RISK ANALYSIS
A. Assumed linear dose-response function, i.e.,
lifetime risk (probability of cancer) = unit risk (probability of
cancer/lifetime exposure to 1 ppm) x average exposure concentration
(ppm x fraction of lifetime exposed to that concentration).
B. Assumed benzene unit risk of 2.2 x 10-2/0ifetime-ppm) (see
Table 6-1).
C. Used average benzene (gasoline vapors) exposure concentrations of
0.98 ppm (58.0 ppm) for unleaded (regular) gasoline and 1.46 ppm
(72.4 ppm) for leaded gasoline based on monitoring data.
D. Calculated fraction of lifetime exposed to concentration assuming
1. an average pumping rate of 8 gal/min;
2. a weekly gasoline consumption rate for an individual receiving a
high exposure under a given control option;
3. an equivalent "individual refueling life" of 50 years within a
70-yr lifetime.
So that for a weekly gasoline consumption of 40 gal/wk:
fraction of lifetime exposed = (40 gal/wk)/(8 gal/min x 60 min/hr
x 24 hr/d x 7d/wk) x (50-yr
"refueling life "/70-yr lifetime)
= 0.000354
E. Calculated lifetime risk due to benzene in uncontrolled emissions,
e.g., for unleaded gasoline:
lifetime risk (unleaded, uncontrolled) = (2.2 x 10-2/1ifetime-ppm)
x (0.98 ppm) x (0.000354
. lifetime)
= 7.63 x 10-6
F. Calculated lifetime risk to reflect the effects of various control
options by multiplying by the estimated fraction of uncontrolled
emissions emitted out of the tank fill neck after either Stage II
(0.045, theoretical) or Onboard (0.010) controls, e.g., for Onboard
alternatives:
lifetime risk (unleaded, Onboard) = 7.48 x 10~6 (uncontrolled) x (0.010)
= 7.48 x 10-8
6. Calculated lifetime risks due to EDB and EDC using appropriate
ratios of emissions and risk relative to gasoline vapors (see
Table F-l).
Notes: (1) The difference in benzene concentration in the vapor of
unleaded versus leaded gasoline changes the lifetime risk
from high exposure (refueling 40 gal/wk) with no control
from 7.63 x 10-6 for unleaded to 1.14 x 10-5 for leaded.
(2) A change in the assumed pumping rate from 8 gal/min to 10
gal/min reduces the lifetime risk from high exposure due
to uncontrolled pumping of unleaded gasoline from 7.48 x 10-6
to 6.11 x 10-6.
6-28
-------
o:
o
»_
o
i —
CO
?y
1 —
s:
Z3
>— <
S
— 1
CC. >-
LU CD
Q. LU
Q- i—
C£.
LU 1 —
— ~j CO
CS
\>O £*
*""s t""i
•a: r-
C 1
^'cD
CO LU
o _^_
«^ CJ
> <:
LU
LU
•zi a:
>-. LU
_l Q
o -z.
V7 HZ3
CD CO
LU
0 >-.
< 0
CD
LU LU
•z. i—
LU <
LU UJ
CO CJ
CZ.
0 O
OC CO
*"*" CD
LU Z
CJ *—<
z: h-
LU UJ
S ct
CJ <
2: s:
LU
LU ZZ
^* >•— <
1— C
<: co
— j <
=5 CD
SI
^> .^L.
cj o
CD Ll_
| . |
5
5-
^.
CO
LU
^
s_
O
£"
OJ
o
l/J
ID
u
c
CC
o"
CNJ
0
CVJ
1/1
a>
IT)
J_
QJ
o
o
5
."2
c
~~*
>
2
3
E
e
i"fi •
- t
. t
;, e
4
c
"
t
f
4
&
C
t
e
iH
>
c
c
c.
*
ffi
c
c
c
a
c
t-
~
s
*
- y
c c
r
p>
r
c
rs
tn
S
i
ir
=
r
0
C
n
vC
n
P1
c
c
£
' is
'-'
0.
ps.'
"\
5
CC
\£
x
T
;.
£
c
c
ffl
% £ "" £
25 S?
c c c c
r- — c* c
O IP CS C
IN (N O C1
co r. (N * C
«•« ft,m
7* ff> p* (r
0 ffl W (f
5 c 35
C CD 9 C
£%. ££
xx
(S IS N (V
r-! r-i »C rJ
r* PS ^ fs
00 00
* *
* C
0 t
£5
rwr,
fv —
5?
0) IX
ss
rt tn
co e
o c
0 C
K /%
::
~ £
0 C
•r P»
o •
- vC
f
«
«S IT*
"* "*
C C
"! ^
- ^
»n r
O C
tr> Cf
P* (S
m m
c c
- -
r^ rv
"* ""
**. ^
•
=
0 C
— e
a, —
* *E.
X f
t t 1
fie
(£
P- C
5 c
* ^
O <£
N 0
\J) CC
in r
m —
o c
IS
0 C
m S
5Z
- ^
•v
0 C
0 r-
^ (N
ifi ^
P' n
mn
en ff.'
m m
o e
CD tt
(S IN
x x
• •
NM
Pk P>
K r*.
0 C
9 C
•£ <£
x v
TT
C
5 E
& —
1 —
e c.
- X C
* b ft
?^C
tr.
fv C" O
(C — C
0 C C
*x cr rw
g£ ?.
if) 1.1 ff1
M V P*
m ts m
\a * —
p> — n
(DC •*
"^ i
c c e
m so e
fv T N
Ml". •«
zt o
O C O
rt %c *
XX X
»* S"1 (^
no m
co r' tp*
~ — n
0 C 0
r» i£ o
— *s
XX X
•8 g ^
o o *"
*!•
(S P< r*
or •*-
5 c c
— — c
IN fM t£
XX X
"t ^ T
M
C £
o- c e
« — c
i/ C. C.
X t
etc c
5- N C
O — 0 £
V, O
a>c
3 c
ts r.
(ft IT.
— fy
CD CC
ss
mr.
?s
°-
0 0
V 9)
en c
p. «
"" S
0 C
n *•
S C
S -11
•B ft i
0 X
i .a « a.
f °9 i
tr. a
1- t£
(T. r
0 C
^ it
m m
tr> e
CO (C
cr> cc
o in
IN IN
p- m
ss
0 C
HE
gs
« if
C Po
0 0
n tf
PS
X X
mn
u* e-
m r
0 0
, *> *
M IN
*^ x
N (S
•s P.
5 c
CO C
o «c
X X
* T
c
i £
v "c.
C X
« « i
?- =
B C - C
in j
* o
«
•> *i
- ™
i. K
0 C
(N vC
«• 0"
"* x
CD O
si in
o c
• •
n r:
0 C
IN rv
X X
-. P.
-
•c
M C
a =
c c e
H!
u t i.
X •
• « t
tf N
e - c
« C =
*£
0 C
in ^c
sg
n e
N r.
if) (N
CO C
0 C
— cr
CD te
^ j
P* iTi
0 C
r» -
ZC
ft «
e ri
- —
h W
C 0
fN tf
" *
x "*
4) C
in in
,o c
• •
in in
3 c
*s ri
^ x
p- P-
•o
ti e
c e
< J e
z c c
a c.
— t
t,- V 4
f «"- 0
tn tr.
— >* fi <-
So o -
3 C 0 C
tn t£ r1 IN
a1 r~ "> c
IT. £ ^ C
v C if> p*
ff- O C^ O
n p. o c
v r-, n (\
CD CO J, «
M r. ^ "
SK J;S
S £ S ?
5= 55
5^ -RS
n m •« n
-
^ ff> ,J _|
* '** 5 ?
ss ss
-o nf>
c! s°'
m « in «:
ui oi • o yi
n & — —
00 0 C
« e 's tf
Sr^ - ^
XX ^ ^
^ **! * ?
"" "" o c
tN f%
rx p> »1 m
0 C C C
o « « -
^ u; tv IN
.XX XX
** ^ P-
T
o • e
C C I1 S
? "" " -
^ S a « t
** X * X
v e li £ t
cc h 9- S N
57 -C7
W « li
o o
1. u
0 0
o o
Pv C •
c, e-
v, k.
3 C
f* r
V V
CD «
IT r
it m
CO C
o o .
0 C
ix r. •
m IP
£ w
S S l
M —
C C
o «c
X X i
nn
o (ri
ffl m -
0 C
co c
(S f^*
^ x
r is
0 C
^ •
E
23 "5
£
C tt C
aj i£* **
V V C
x X v
?» -
°° ' Z
e
"c
e e *•
0 C •
- - 0.
c - -> . **
e t f
-55 T
w -C T 13
t 4 * C
£ N N C
EC. ffl ffl S
m T n e
• • t
t
?
-
i
i
-
—
V
a
•fl S
"a^
ff
3 0
s Z
Oj J,
—
Is
:i
=
cr-.
6-29
-------
ID >-
eJ
a uj
o h-
O
CO CSS
ss
UJ
Q J—
UJ UJ
2: o
UJ OS
UJ CO
en
U- UJ
^x:
UJ O£
UJ
O H-4
2: _l
t-H O
UJ
2 8'
« ON
PN
r* rv M p, H 03
CO
§
«=C O
O OS
UJ O
uj o:
UJ
_j
ca
*/ a x ** o *• ;
! Si
e £ e c
D - 0 0 — C
(N n *.
- C c — e
*• c *. - o
a-* in &.—
* "c. a c "S.
H X • X C
• < e H < e
FI pj n n
•. - cr<
** 0 O O -
i e *) e
e c -j c c
: c *- c
c a-
zt-<°* s-S != 8
§*i **
u o
1-
6-30
-------
LU
— 1 >-
O CD
OO LU
«=C 1—
CD «SC
o;
O CO
DC.
QZ.
CJ O
Q h-
LU o
CJ OO
Q CD
| , J ^»
•=C P
2: LU
i— i i£
I— o:
oo <
LU s:
r-'
IO
LU
i
03
1—
t
9
A
CO
Q;
03
e
*
y
_Q in
uj ™*
o
LU
e
— * 0
o -
CM *
O *
1 W
CO 0
I— >
""""* g
U)
1_
V
uo
rn
t_
QJ
> e
° -s
03 -
0 a!
c
OJ ^
-O -M
•r- 3
cj a
QJ
•M
3
o 7
.•
s
b
0
t-
*3
a
Is
a o
*3 0
en b
c *»
m v
N O
o o
0 0
; v m
ss
0 0
N ON
m m N
d do
3 SS
(N N (N
O O O
•4 O ••
N \n ^
(N (N N
0 00
N MM
0 00
f (NO
3 SS
o oo
0 00
CO ON
(N N N
0 O 0
N N N
N N N
N N N
8 88
d do
8S
00
•J) PI
««
d d
o in
o o
m m
o o
Ss
0 0
N <4
SS
P P
0 0
0* *
-I
OP
P 0
N N
N (N
O 0
d d
O.O0558
O.OO330
ss
o o
do
in m
OP
d d
a
< — s
9 O.
H X C
"CO
N N MO
00 00
M M m N
N (N •« O
O 0 00
22 SS
P P O O
P P OP
OP P 0
t* \o C> v
N N — -
PP P P
PO OO
f m N N
88 §8
do do
N ff> in m
o o op
o o o o
dd do*
-- il
o o o o
dp d d
ea eo N N
o o n tn
o o o o
do do*
o e
e s
2S 5
** O *J
O.M a
* 4J •
2S- «S
« * « - «
9 a 0 9 *
0> H en H
e M 0 o -•
*j B c ** a
MO O
m co o
M fS O
do d
^ N O
MM §
do d
M Ft
So
O O
d d
o.tiosse /
o.oossa /
0 O
. o 0
0,0
r* tv
in ft
o o
d d
,
Jj
0
a a
= e
O ^0
" a
M f
HI 0
1 1
H .
O O O O
fN C* O O
I M 00
00 00
M In o o
MM 0 O
00 00
n eo N N
MM n m
do do
M N -4 M
o o OP
mo N N
o o o o
0 P OP
M 4) MM
M p (N N
do do
T N co eo
(MM m m
OP 00
00 00
c> n « • •
X X
6-31
-------
2^
§
U.
CO
D;
0
Q-
UJ
Z CO
1-4 UJ
«-J 1— «
O CD
CO UJ
CD «£
CC.
O H—
§ CO
03 CD
E o£
0
U. UJ
UJ O
O UJ
2 _J
UJ UJ
O CO
1 II j
o D;
z ui
§
UJ 3
»-« CO
1— . -Sf
i — <
i i
O UJ
a.
o o
LU
*£ j
Z 0
1 ^ -.p*
CO UJ
UJ >•
CM
rH
1
U3
•°_.
f !_
0
Ow
CO
CD
c
•^*
"o
tn
to
0
1 a>
c
0)
c
>
1 £
j=|
r- >
3: uj
•tn
c
o
re tn
c in
21
t> UJ
a a)
•a >
4J
U ro
3 i-
0 0
c a.
P CO
S LJ
t/1
C
O
V)
tn
13 "5
§"
i- tU
V)
3;
1
CL
O
a;
^«
U
*•
I
ra
o
r—
ie
«§.
4-> CO
tn
C
O
ra in
C U)
o •*-
a U
4->
"5 Z
J= 0.
4J CO
•»- >
3: uj
UJ "*
_J r
£ 8 £•
i. j- >>
Eo en
4J O
r * *;
S i— ro
i a i;
a o-
^
CM
t- O O
1 O O O
1 — ( O O
U) CM -H
co en CM
o o o
I- f_ S_
j O 00
1 O O O
O CM O
to co
J
» •
c 3E 2£ Z 2
*"-.. ***. **^^*.
\o co r»- ^*-
"•sf PO co ro
fO PO CO OO
^ < <<
z z z z
CO CO CM o
CO CO CO CO
CO CO CO CO
0 O
C •>- 1-
O 4-» 4-»
T- O O
4J 3 3
o -o -c
3 cc N N
c -o ca ca
•»- t- N
"ru o
tn .0 tn ^r co
rO C ro • •
CO O CO CM r-<
UJ CO
•-" > «S CD
"— ' S* X
1111
_—
i
,^
f— 0)
s. . "3
a> 01 «»
a. s- j=
a. o
a> £ s-
r* CD O
3 - tU
CO T3 4->
r— c re
Q. ro E
t. C 4J
o o tn
•p** QJ
cu tn
4J tn tn c
V c c
O CO
•O f— U
0 4->
o ra tn
J= S- 0)
•r- a; u
r— a. c:
(U o 1)
•^ QJ >^
1— i— -0
T- > C
X co
CO 4-> •
^ U 0) tn
— - 0) C C
t*—
> c at tn
01 o j= tn "E
i^ ^ *^ 1.^
*O r^ 0 re"'a)
V) OS- > O.
CO S- tfv •*—
Ol 4«* QJ fi
C C C •—
S- O O i- ••-
O O T- r— CO
t«- 4-> O 4J
^< O tn -^
>i « 3 re
r— -0 C3 4J
re a; tu in
o en s- 3
CO • re
ro 4-> O dj jz
4J cy> s_ ,_ X
CO U jQ CO C i- C
CD i- r— O
4-> CL.4J
cn flu , — co co
c en 3 s_
i- re in 4J aj
in 4-» aj o a.
=3 OO 1. Z O
CO JS t_>
6-32
-------
For Stage II-NA* with size exemptions (II.A.), for example, the estimated
incidences due to gasoline vapor emissions using the plausible upper
limit unit risk factor (see Table 6-9) are 46.8 or 77.2 from bulk
terminals, 23.8 or 39.3 from bulk plants, 96.5 or 159 from service
stations, and 601 or 996 from self-service, for a total of 768 or"1,271
and a reduction from baseline of 75 or 122. As outlined previously (in
Section 6.2.2) the incidence estimates consider size exemptions, phase-in
and phase-out of controls, and gasoline throughput projections. It
should be noted that self-service refueling contributes about 90 percent
of the total incidence due to benzene and about 80 percent of the total
incidence due to gasoline vapors.
The lifetime risk from high exposure from each source category is
given in Tables 6-13 and 6-14 for benzene and gasoline vapors and in
Table 6-15 for EDB and EDC. For example, the estimated lifetime risk
from high exposure to benzene at baseline (no additional controls) is
1.23 x lO-4 for bulk terminals, 6.4 x 1Q-6 for bulk plants, 2.4 x 1Q-6
for service station (ambient concentrations in the vicinity of facili-
ties), and 1.13 x 10'5 for self-service (concentrations at a station
near a vehicle tank during refueling). Thus, the highest lifetime risk
from benzene exposure estimated for any of the source categories under
the regulatory strategy is 1.23 x 10~4 for bulk terminals. In fact,
the lifetime risk resulting from high exposure to bulk terminal emissions
was found to be higher than the lifetime risk from any other source
category, under all of the regulatory strategies and for all the pollutants
examined. The lifetime risk cannot be summed for the industry as a
whole because it is unlikely that any one individual would be exposed
to high exposures from all source categories.
Lifetime risks are given only for regulatory strategies with size
exemptions because the calculated results differ only slightly for the
with-size exemption and the without-size exemption options. Thus, the
presented risks demonstrate the effect of the regulatory strategy,
rather than the risk associated with any exempted or unregulated facility.
6-33
-------
CO
»—t
a
H-
ea: >-
LU CJ
O. LU
O. H-
=5 .
co o;
=> o
O CS
Z UJ
»-( D£
CO
=3 ar
~_~ cj
CO UJ
Z CO
O D£
CO O
< as
O LU
o£
z u
LU
LU O
z o:
LU =>
I~W O
Z CO
LU
ea ca
OZ LU
U. ^
a:
i . en
— | co
VI C O
CJ I-
.cr o> *3"
014J 10
= 5 CM
«. co°
0 CM
U» ~4
4) O
u ( en
"> ! «-
1- O
CU
IA r~
I «*•
a) co
01 ! _;
*-4
2 I •*:
1 ; K
1 4J O
O to
~-l LO
X U CO
vi u • «r
e c« CM
»O
> 1
IS 81 ! •—
U = ~
U f- 1 O
3 ,— Ul CM
MO 4J
0 VI e : fc.
O. (9 Q O
as E -
•ss * -
= s - 0
p
t-S 0
ji S!
•»• ti« i— ] CO
Lifetime
(probability of e
Bulk Termlr
e 123(2364 or
? r
0 >> ea
= *
0) 4J —«
a: to
§o
0
cn cn
CO CO
123(2364 or
123(2364 or
~* «— i
«• «•
u •—
o o
in in
CM CM
*-t •**
in in
0 0
i** r*
— i ^*
•9- «•
0 O
co co
in in
CM CM
m co
—4 ~<
CM CM
O O
CM CM
t_ S-
0 0
CM CM
CM CM
•N •*
to to
0 S
§ §
co co
0 0
*r «r
to to
CO CO
CM CM
CO CO
CM CM
»— 1 *"1
*
•< «c
z z
1 1
CU OJ
01 o>
(O fi
4J -W
Ul OO
i— i *-i
— O
CNJ O
*r c^
to co
o o
a* «r
CO to
CO CO
— CM
"°. «
O CM
CM ~*
CS ^
en
° fe
^*
* IT)
^ •
fr> ^H
• in
-H O
C» r-*
9 5
0 O
LCI CO
CO LO
Si CM
to co
~-4 ^^
"m
CO CM
LO O
CM
° fe
CO
• CM
CM CM
CO —<
—4 to
CM O
«• cn
to co
o o
cn *r
CD to
CO CO
•_• CM
CO •— •
CO
O CM
CM -*
V CU
en en
Q (O
JJ 4J
to to
>• >•
— C3
CM 0
«• Cn
to co
0 0
cn w
co to
CO CO
-»• CM
CO • — •
CO
O CM
CM —«
CNJ
— a
o
s-
"! o
to o
0 0
— ^
CO
r^» ^
O 0
«• CO
2 S
CM iO
O -«
in
ro CM
LO O
CM
* *
CM
CM CM
CO ^H
—4 to
— S
CM O
Sen
CO
o o
o> *r
eo to
CO CO
*^ CM
CO —
• CO
O CM
CM -H
4)
cn
(O
•w
to
«o
u
CU (0
cn o
Q J3
JJ C
to o
> 1—
>
o
0
S
0
•a-
'XI
CO
CM
CO
CM
r-t
CM
O
o
o
to
o
0
«r
^-
«r
o
CM
to
CM
«T
•H
CM
g
t.
O
CM
CM
10
I
CT»
C")
O
^
^O
(T>
CNJ
CO
CM
«— 4
•e
-K
<
3
I
""S
0) «O
cn o
K3 J=
4-* C
? O
S
0
o
cn
CO
o
«r
<£>
CO
CM
«n
CM
CM
0
O
o
to
o
o
«r
^
^r
cn
to
CM
w
CM"
O
CM
S_
O
CM
CM
to
i
cn
CO
II - N.A. i
) 123(2364 01
i,
tl> 03
cn o
a .0
4J C
) O
X
CM
*r
to
o
cn
CO
CO
CO
O
CM
CM
0
0
o
to
o
o
o
in
••4
*H
en
LO
o
CO
CO
LO
t.
0
CO
CM
CO
«~4
CM"
^
to
a>
CO
CO
"n
o
CM
I
(t)
O
.n
c
o
•e
0)
cn
4Q
4->
to
X
CM
*T
to
en
CO
CO
CO
G
CM
CM
O
O
o
to
0
o
o
o
to
«
o
CO
CO
in
!_
O
CO
CM
CO
wH
CM
«•
to
§
(T>
S
"1
• ta
<. o
• J3
z c
o
1
•B
to a>
cn cn
«
00 01
X
cn
CO
«r
to
CO
CM
CO
CM
CM
O
O
o
to
0
o
«3"
W
«•
en
S
*r
^M
CM
S
u
o
CM
CM
**
«o
O
O
en
CO
*r
to
CO
CM
co"
CM
Z
10
o
J2
C
o
•B
ID
cn
a
4^
01
X
CM
^r
to
o
en
CO
CO
co"
0
CM
CM
O
O
o
to
o
o
0
0
o
1-4
IO
*r
CO
o
CO
CO
LO
U
o
CO
CM
CO
CM
"9-
to
O
en
CO
CO
CO
S
•e
01
cn
IO
4J
to
«•£
O
OJ J2
cn c
10 o
4J
01
=
s§
cn 01
m ro
i. u
^* ^r
^O _ ^ 4J
F- j: o
4J 0) 3 01
•F- E O c/1
3 n3 J= tj
« 4J £
•o F- a.
0) (1) 10
IO 4J U
% £-•?
VI 10 t-
« £ u
n QJ o
— . E •*-
I/I V "^
ji -a *J TS
tS— a s-
•*- J tw —
c eg
— cn a; eu
ra > cn
V) 4-> 10 10
a> 01 f 4->
— to
OTO O
0) c vi cn
4J 10 r— S.
i<-o =
v> z t. ja
i it) S
>,— 0 0
i. — J3 0
O C
4^ ai o vi
10 cn c
<— Id J= O
3 4J 4J ^«
CALO ^- 4J
Isks demonstrate effect of re
). Therefore. Stage 1I-NA*.
e II options In combination w
except for service station op
L. vi cn
4) IO *
O> •— 4J •
E +•> 01 -O
4J ^ O> IO
oj ^_ QJ o
«- U 1- J2
^- IO £ 1=
_1 U. 4-1 C
IO
CO
<—4
I
IO
LU
_J
ca
-------
cs
O LU
O I—
O «£
z a.
1-4 |_
_i oo
UJ
s«i >-
I-H o;
_i O
X UJ
< cu
us o
z <
0£ =9
O
Q_ 00
«=C LU
LU O
Z C9
O <
00 O
^C'
O LU
UJ
•Z.
UJ
to
CD
CD
o
OS,
oo _j
az oo
LU C3
I— O
LU C£
Lu U.
<=) O
LU I—
H- O
CO tO
LU i—i
Q£
I
vo
CQ s:
0)
>
4-»
*•> 4->
z 3
0
u_
u
>
s_
V
10
u-
*••
0)
00
C
o
•f—
+J
« 5
i GO
o
u
x • •—
>
— u
in
IB
CU CU
U C
3 i- Ul
>*- *>
a. in «
X «3 —
uj en a.
.ea) .*
en C t—
•r- CU 3
z N CD
B
E 01
O -O
i. •
M—— .
*»
.K U
tn cu
i— »*- in
a: u- f—
cu 10
E u_ —
•So E
*j E
u >> cu
w— 4«> h—
_J p— JL
£1 3
.0
O
u
a.
Regulatory
Strategy
0
CM
CM
CM
O
CTv
U>
<•>
CM
~*
^
»•*
ir>
s-
o
CO
IO
00
r-0
v»4
**
CO
o
TT
0
oi
CM
W
CM
to
••<
O
CM
CM
CO
w
10
o
CM
CM
CM
O
1— 1
CT>
10
ro
CM
••H
CD
"«3
t/t
to '
CO
0
CM
CM
CM
0
^H
CJ>
l«
PO
CM
t«H
oo
u
o
r»
•_<
in
o
CO
m
CM
o
r*.
ro
fH
in
•H
v^
O
CM
CM
CO
^
vo
C3
CM
CM
CM
O
~H
en
in
PI
CM
•K
<
•X.
1
cu
C3)
T5
«
OO
O — * O — * O O O • — • — O - — • CM CM
CMinCMLTlCM CM CMLT5 ITICM LTJ CMCM
CMVOCM^OCM CM CM«3 LOOM \O CMCM
CM n CM ro CM CM CM en ?n CM ro
i. S_
00000 O.OO 00 0
I-H i-^
^H C*J ^H CVJ *— I ^ —*. CM OJ*^ CM CT^CT^-
O^VOO^^OO^ O^ CT\ kO ^O O*i iO LDLO "CJQJ+J
tncNjincMuo to iricsj CMIO cv ^-**H «S3
— m •*- • co ^^ — ^ •— * m m -— n mo flt/)^
cn»ro.fn ro m« -m - •• ,__ .^ tn
CNJOCNJOOJ CNJ CNJO OCNJ O *^m 3 0> — ,c 2 x
£^ =
^- ^r^r^^-tr^r^- — '*-» .^ **
^-. • — * — * * « «» *. . *& ^ i- v> O)
eo — 4 ro ro O o OO OO O •• o-»-4-»o»
Ul ^H ^H S. C
O O O O O OO OO O OO Q.-T- (4- «-
CO romc-orofomn-) s-*J**-o
r*.«r**r*.» • •• •• . coco o» LD o o OO oo O CM^< -i-J oj rs 1)
• t— (••« • •« •. . *• *^£O1
O «-« O O O O O O O O O ^CSJ %(OJS*O
tn *J J=
TC3 .— 0.
O) 0* ,_» .. •r-.irf.'*-
• • 'C\j • • »uo r** « r^ oo u in • «m • ^^s- o>«^3
C\JC\JCS1^-CSJ CNJ CNJCO ITJCxJ. IT) i/>ir> * m m 4^ ^, ^^
i^ o «^ o «-« •— » .-HO o «— • o »-*»-H o
w* O>
OOOOO O OO OO O OO (Qn3
0> t/1 .C +J
CMP-*.C\ir**CSJ CNJ CNJf** ^.CM P*- CNJ CM *^ CO
CM^-4CMt-HCNi CM CM -H ^HCM ^H CM CM OJCtAO
CO CM CO CM CO CO CO CM CM OO CM COCO +J «3 p— &
U3 — H U3 — 4 UD 1O 10*-* *-H^0 ^ CM -H VtZuIS
1 (0 S
>,*— O O
*—*-*. . W i-^ .Q O
— » — » «—•. — - — » ^^ • O O O C
O — O — » O O O—» -^-»O — * CNJCM 4-> CU O W
CM LO CM IT) CM CNJ CMUO IOCNJ IT) CMCM «O»C:
CM ^O CM VO CM CM CM ^O ^O CM ^O CMCM r— « JS O
CNJPOCMPOCM CNJ CMPO POCM PO 3^-»+J-»-
s- i. cnoo i- *J
OOOOO O OO 0,0 O V. • 0
»-* — * -K e
»MCM»-tr> o 3 — o
U1 CM LO CM un LT> tO CM CMLT) CM — 1 -^ 14^-r-
^-* fO ^^ f*1) ^^ «_• <~~ PO CO ^^ C*0 f"O O D '•^ C *O
ro -ro >ro m ro- -m • ••• a) •—••->
CMOCMOCMCM CMO OCM O IOPO K^CU^U)
^HCM-^CM»^ "^H *-H'CM CM*-H CM' ^* CM >*— o> s
cu 0 U
CD OO -^
EC 4-> C >
oo o o «o M*T— u
i— — s. cu cu
— «e -o— *>+j+ji.S.U CUCUQ.
z *•> z z c z c o *-> u •••o j= o *J
> OO OO 3 M N ' ' 1— O.
1 1 1 log *7cacQV)i— *CU
•O «aaOa6CU J»tf <— • O
•— •-, • . !_i _ ^i»- eK»e»a 10 •• x
i. i. i- i_ sj • . s-uicn-o
cu cu cu cu 10 cut) cua cu cucu cu cu«3 cacM<-< cuia-cu
en en en en o cno eno en cncn cncno ^oco cu •— ** ^-* u
10 re re 10 .a to ,0 ia^ IQ IQIQ ia IQ^ v>- E^OO-OCU
4-> *J +J JJ C 4->C J-J C +J J-> *J 4J 4->C as • • -r= i- S- T3
00 00 00 OO O OOO OOO 00 OOOO 00 OOO UJ«SC3 J-> r— -«>-«—< x x i— i i_i«-. ^» *i-
-------
TABLE 6-15.
ESTIMATED LIFETIME RISK FROM EDB AND EDC FROM GASOLINE MARKETING
SOURCE CATEGORIES UNDER EACH REGULATORY STRATEGY
1.
II.
III.
IV.
V.
VI.
VII.
VIII.
IX.
X.
XI.
XII.
XIII.
XIV.
Regulatory
Strategy
Basel 1 ne
Stage II -
Stage II -
Stage I
Stage II
Stage I *
Onboard
Stage II •
Onboard
Stage II
Onboard
Stage. I 4
Stage II
Stage I 4
Stage II
Stage I &
Onboard
Lifetime Risk from High Exposure*
(probability of effect) to EDB/EDC x 10-*
Bulk Bulk Service Self- Highest
Terminals Plants Stations service for Alternative
• N.A.*
• N.A.
Stage II -
- N.A.* &
- N.A. I
Onboard
- N.A. 5
Onboard
& Onboard
Stage II &
Gas Bz Reduction
A. 62.4% Bz. reduction
B. B1.3Z Bz. reduction
347/448
347/448
347/448
68.8/89.0
347/448
68.8/89.0
347/448
347/448
347/448
68.8/89.0
68.8/89.0
347/448
68.8/89.0
347/448
347/448
16.8/21.7
16.8/21.7
16.8/21.7
4.4/5.7
16.8/21.7
4.4/5.7
16.8/21.7
16.8/21.7
16.8/21.7
4.4/5.7
4.4/5.7
16.8/21.7
4.4/5.7
16.8/21.7
16.8/21.7
6.4/8.2
3.5/4.5
3.5/4.5
4.4/5.7
3.5/4.5
0.67/0.86
6.4/8.2
4.3/5.5
4.3/5.5
4.4/5.7
1.6/2.1
4.3/5.5
1.6/2.1
6.4/8.2
6.4/8.2
196/253
8.8/11.4
8.8/11.4
196/253
•8.8/11.4
8.8/11.4
196/253
8.8/11.4
8.8/11.4
196/253
8.8/11.4
8.8/11.4
8.8/11.4
196/253
196/253
347/448
347/448
347/448
196/253
347/448
68.8/89.0
347/448
347/448
347/448
196/253
68.8/89
347/448
68.8/89
347/448
347/448
.0
.0
•Lifetime risks demonstrate effect of regulatory strategies (not higher risks associated
with exempted or unregulated facilities). Therefore, Stage II-NA*, Stage II-NA, and
luge ^nationwide) have the same lifetime risk. The same three Stage II «P*°»S "
combination with Onboard also have a common lifetime risk (although Afferent than without
Onboard). Since EDB and EDC are emitted only by leaded gasoline, Onboard controls do not
reduce EDB and EDC emissions because Onboard controls will be used on cars using unleaded
gasoline only. EDB and EDC lifetime risks from self-service, therefore apply only to
those individuals using leaded gasoline during their entire lifetime. Onboard controls in
combination with Stage II would increase EDB and EDC emissions after Stage II controls are
phased out.
6-36
-------
IS)
a; oo
o LU
a. >-<
«c o;
>• o
cs
LU LU
M =1
_J C_)
O
OO LU
=C O _J
CD a: uu
=5 :>
o o uj
2: co _j
LU 1— Z
h-J LU LU
LU &^ i*v
ca eC O
S U_
•Zi *Z
O LU LU
LL. i—i O
LU O i
a 1—4
o o LU
i—i U. et
LU .—-I—
> o: oo
—4 O
h- h- >-
«=C <_) a:
—i — OO
s: —i
i—t CC
t— o; LU
OO LU O
LU a. s:
Q. =3
UO LU
.—! _J
i en
oo
LU =3
CO —I
c
0
tare
<« t
c
c •*
0 i
t 01
*J tf
U f
D 03
CE
C
D
a
0
D
Cre
•H M
-H O
D a
0 1-
X
c
II
N
e
0
a
o a
82
8 >
s in
w2
01
B (0
k
C
e
m
to
i> a
> c
O 0
U «
C 4i
V Eft
•o
15
M >
<»< C
0 CO
a
2
4J
0
_4
a
JC
3
CD
n
._(
' O
c
L.
0
^
•3
CD
CP
9
jj
0
£
10
o
Jj
e
3
S1
B
O
0
in
CO
to
N
5
t-t
0
to
*
to
fH
?
"
g
fv
«^
X
VD
tf
m
to
g
03
S
N
N
1^
U
0
03
VB
N
S)
e
a
o
a
<£ CO
«o -v
0 0
to o
X X
82
co . 03
03 tr>
0 0
CO Ml
N 0*>
in in
O vC
° S
**
m in
0 0
in N
OJtfj
X X
w <<)
v to
m m
to to
0 0
CO 03
to to
(N
M U
» U b
< 01 0
z av<
6) C
1 C HI
« C
« S -i
IP C C
e c -<
*J < X
in
in o
0 O
^ 0^
» N
X X
ID 0^
CO f
N Oi
0 O
m M
to o
vv
in co
sa
0 0
si
tn M
a> m
V f»
Sn
00
tM t^
N tn
X X
O i
•v n
to n
o> «n
to m
^
03 03
M tM
N (S
N W
N r^
b b
0 0
CO CO
V V
NN
0)
N a
0 m
n ti u
•2 ffl 0
4J Of***
a B) c
C E
•p < Z
in
If »H
N O
h V<
0 0
O V
CT* \0
in m
X X
tO N
§to
O
V CO
0 0
^0 CO
fM V
(0 \0
»4 tS
IM n
0 0
O rt
v «•
rt P)
N tO
S3
* N
03
b k
0 O
t» N
8R
X X
r-
LI
0
(N
V
X
in
N
0
N
O
tn
*
S
03
0
g
X
N
£
2
d
g
03
03
10
X
CO
m
m
to
t?
CD
^
N.
«-<
w
r^
0
co
§
•s.
V
CM
C '
a i
ft C
t- 0 V
.C 0 ft!
4J "H >
3 •* 4J
T3 < 0
"c * w
O i
°*
to
U
0
UD
in
X
to
g
0
a>
M
•«•
tn
03
0
g
X
^
m
2
g .
CO
03
? S
^ 0
* w
1
o
a>
v>
0
JC
5
*.»
—
>
I
c
CJ (DO
*O" re *-
4-1 U
Vi t/1 O)
« Cft (/)
4-> C C
C "- (O
i P
JU -D 4-)
i -5s
. £--
>> o <
jD QJ CT
^Z ^CO tn
».
re
c
4->
•5
•o
re
o
s
=
^
1
'i
?
a*
^
OJ
i-
s
•^
>
CT
C
"^3
ft)
•£
1
Q;
£ S
'o o
re *->
&> re
i-l
§°
ft!
I/I f—
= JC
'o >
c
cS
*— tn
£ -^
"c §
o 01
U QJ
-tJ i-
o c
UJ e
revs
JCi C
o
O >fc
c a;
o >
tj IE
*£
S "
sf
— (J
ex—
c" ^
^ w
aJ
£ g
a; ai
T; S-
o o
£ 0)
> U
c*"a)
*£ >
Q.- C
i1^
c o
*o w
re *E
WQ,
(/) O)
"u t-
c o
.^ n.
il
O
W c
U —
"c ?
o re
u
4-> C O
O i- i-
-o^1 °
6-37
-------
t/o
O
O.
LU
LU
O H-I LU
cot— s
-< Q
LU _1 5E
rxi o «c
= to
LU «£ CO
CD O LU
LU--~t—
o ed e/>
o o o:
LU I— 4 CJ3
>• o: LU
>-» OS
<: HH o.
_i z LU
=3 LU LU
O H- _l
•a: LU
=3 I- a:
i to LU
Z LU O
LU O to
I— 3=LU
< H-4 HH
S _i ct:
•-< LU o
»— :*: CD
CO >-4 LU
LU _l I—
^H i-« O
I X C£
t
0
£"°t
c
C -4
O i
•~\ c
•a o
u e
3 £0
•o
It
c:
B
1
a
Bra
-4 -1
"o "
C 0
a t-
01
o
B
C
N
B
C
a
O
IN C
o u
M -4
' ^
£ a
cn (n
W *M
O C
a
*>-
n
n
b
a «
> E
O 0
ss
« «
•a
-4 V
U 0
"s
a
z
B
C
Q.
X
0
B
B
E
b
a
X
g
Q
X
0 >•
*> er.
B V
-4 4J
3 0
Dl b
c n
cn
0
CO
m
CO
•H
CO
CM
b
0
o
rt
to 7
•r cn
p« 0
b b
0 0
in o
KS
V P)
v r)
» »
N N
CM CM
b b
0 0
O O
VD 10
X X
CM M
cn cn
CO CO
b b
O 0
in m
CO P)
T V
CM CM
X
C
e B E
NEC
~< 0 f
"38
* 0 b
< V 0
z a n
N CM
CM V
7 f
CM CM
CM CM
b b
0 0
0 0
to to
X X
CM N
cn en
CO CO
b b
0 0
n m
PJ pi
v v
CM CM
X
t>
4J
0 B e
0 E
--< o
E *> 0
0 U b
•H O 0
4J 0.1*4
Z C UI
1 M
Stage I
Annual
Minima
""io
cn
PJ
b
0
^
CO
CM
X
ID
10
1
b
0
en
CO
CM
**
(n
in
CM
CM
0
$
-H > 0
B ^ -<
•TJ b 13
Onboard
- w/o A
Evapo
- w/ Ad
-ON ]
CO t
b 1
0 I
»H 1
CO 1
0) l
1
X 1
1
f"* 1
1
1
1
t
m i
O l
0 i
i
CO 1
M I
l
1
to i
1
£!
I
h i
. 0 I
l
$ !
•H 1
r
M i
* i
«H 1
m i
i
i
i
i
• i
^ i
*D 1
I
b I
0 l
t
t
P) l
• t
\D t
X l
l
CO t
f) t
1
1
1
l
T I
N j
o :
O I
• l
*4 1
CM i
«H 1
1
1
I
cn i
• i
m i
0 i
i
i
CO i
1
• I
N 1
I
t
1
0 I
h i
§ !
O 1
l
0 '
> 1
O I
(4 1
O t
Q, 1
0 1
> 1
U 1
'1
i
1
cn
*^
T3
1
O
QJ
•3
1A
u
(*_
QJ
(A
•£
1
_e
o
in
|
£
^
£
C
w V
c
«TS
CO OJ
Gf> in
C C
4J
in »—
*x §
in M
?"£
i<
u z
c «
— (U
OJ U>
M (C
CD to
Q
(B
J
+j
(O
**-
o
£
§
u
4->
3
•*•>
»
C
1
e
Of
u
1
I
Q)
!
*r- tf)
0§
>»fc
o °
u
V *p»
"u >
c
.£ E
o
c k.
o
1/1 o
*O (A
4-1 E
§s
u o
o 4->
«£
u o
QJ O.
t?
uj Qj
f
g
^_ j_
a* m
(/) O
ra A
ja c
o
5 >,
•o
C
"*" —
U JC
= U
~o to
£!o
t. ^
o*
IA
(A
s JT
•^ «
"3 c
& 4J
S- (D
t.
U O
J= QJ
CO.C
C 01
W
"ii
: (/)
il
§> w
II
•§£
C O
i ^
o
CA 15
(A C
cl
O (D
U
O (A
"S £
O) •*-> 4-1
t§§
LU U U
"D
6-38
-------
Therefore, the lifetime risks associated with Stage II are the same
whether applied in selected nonattainment areas, all nonattainment areas,
or nationwide, both when used alone or when used in combination with
onboard controls.
As a further analysis of Stage II and onboard nationwide control
options, the incidence reductions achievable when considering in-use
efficiencies were estimated. Tables 6-16 and 6-17 indicate the total
annual incidence and the annual incidence reduction achieved when
considering in-use efficiencies associated with no inspections (minimal
enforcement) and annual inspections for Stage II and the expected level
of tampering for onboard.
6-39
-------
6.4 REFERENCES
1. Anderson, Gerald E. and G.W. Lundberg. User's Manual for SHEAR:
A Computer Code for Modeling Human Exposure and Risk from Multiple
Hazardous Air Pollutants in Selected Regions. Systems Applications,
Inc. San Rafael, CA. SYSAPP-83/124. May 1983. 73 p.
2. Bulk Gasoline Terminals - Background Information for Proposed
Standards. U.S. Environmental Protection Agency. Office of Air
Quality Planning and Standards. Research Triangle Park, N.C.
Publication No. EPA-450/3-80-038a. December 1980. p. 6-11.
3. Arthur D. Little, Incorporated. The Economic Impact of Vapor
Control Regulations on the Bulk Storage Industry. U.S. Environmental
Protection Agency. Research Triangle Park, N.C.
EPA Publication No. EPA-450/5-80-001. June 1979. p. II1-9.
4. Reference 1, p. 53-55.
5. U.S. Department of Commerce, Bureau of Census. 1980 Census of
Population, Vol. 1 Characteristics of the Population, Chapter A
Number of Inhabitants, Part 1 United States Summary, PC 80-1-A1
April 1983. Table 17: Land Area, Population and Population Density
for Counties: 1960 to 1980.
6. Clayton Environmental Consultants, Inc. Gasoline Exposure Study
for the American Petroleum Institute. Job No. 18629-15.
Southfield, MI. August 1983.
7. Telecon. Siebert, Paul, Pacific Environmental Services, Inc. (PES)
with Coffman, Mike, Clayton Environmental Consultants, Inc.
June 14, 1984. Clayton's gas chromatography-f1ame ionization
analysis procedures.
8. How Self-Service Appeals to Motorists. National Petroleum News
1983 Factbook. 75_(7a):103. March 1984.
9. Telecon. Siebert, Paul, Pacific Environmental Services, Inc. (PES)
with Fletcher, Bob, California Air Resources Board. June 22, 1984.
Origins of residual vehicle refueling emissions after Stage II controls,
10. Office of Mobile Sources. Mobile 2 Model Output: Benzene Exhaust
and Evaporative Emission Factors for Vehicle Operations, 1975-2001.
Job No. 665964 University of Michigan Terminal System (Model CT043).
Ann Arbor, MI. November 2, 1983.
11. Energy and Environmental Analysis, Inc. The Highway Fuel Consumption
Model Ninth Quarterly Report. U.S. Department of Energy.
Washington, D.C. Publication No. DOE/PE/70045-1. February 1983.
Appendix B.
12. Energy and Environmental Analysis, Inc. Highway Fuel Consumption
Model Output: With No Diesels less than 8500 Ib. U.S. Environmental
Protection Agency. Office of Mobile Sources. Ann Arbor, MI. ca.
1983.
6-40
-------
13; Black, P.M., I.E. High, and J.M. Lang. Composition of Automotive
Evaporative and Tailpipe Hydrocarbon Emissions. Journal of the
Air Pollution Control Association. 30:1216-1221. November 1980.
14. Bowers, J.F., J.R. Bjorklund, and C.S. Cheney. Industrial Source
Complex (ISC) Dispersion Model User's Guide. U.S. Environmental
Protection Agency. Research Triangle Park, N.C. Publication No.
EPA-450/4-79-030. December 1979.
6-41
-------
-------
7.0 COST IMPACTS
7.1 INTRODUCTION
Cost data were obtained and developed on a per facility basis for
each model plant size of each source category within the gasoline
marketing network. These per facility costs were then combined with
data on the number of facilities requiring controls within each source
category in order that nationwide costs could be determined. This
section presents source category costs for bulk terminals, storage
tanks, bulk plants, tank trucks and service stations. Service station
costs were further divided into Stage I, Stage II and onboard analyses
on a per facility basis.
Nationwide capital and annualized costs were calculated for each
control option and regulatory strategy on both a cumulative and a net
present value basis. Net present value (NPV) costs were developed to
compare the strategies and to take into account the opportunity costs
associated with spending money now or in the future. Discounting costs
(using the NPV of costs) was performed because, like emissions and
incidence reductions, control cost for the regulatory strategies are
not uniform over the time period analyzed due to 1) the slower phase-in
rate of onboard controls compared to Stage I and Stage II controls, 2)
the varying equipment life of onboard, Stage I, and Stage II control
equipment, and 3) the declining gasoline consumption which directly
influences the recovery cost credits. All costs are presented as
constant 1982 dollars.
For each control option, the capital cost spent and the annualized
cost incurred in each year was determined for 1986-2020. These costs
per year (time lines) for each control option can be found in Appendix G.
Capital costs over the 35 years of the analysis were incorporated
in the initial years of the analysis and then repeated in the years in
which the economic life of the equipment ended and replacement equipment
should be required. The capital costs and annual costs time lines took
into account phase in of equipment, as well. Annualized costs were
7-1
-------
adjusted each year, as appropriate, to reflect reduced recovery credits.
Since the number of facilities were assumed constant and the gasoline
consumption declined with time, the consumption decline was reflected
as a reduced throughput at each facility and therefore reduced recovery
credits. Reduced recovery credits result in higher annualized costs.
Nationwide capital and net annualized costs are presented in
Section 7.3 for each of the gasoline marketing control options.
Enforcement costs are not included in these estimates. Net annualized
costs were combined with the emission reductions (discounted and
non-discounted) given in Section 5.0, resulting in cost-effectiveness
determinations for each of the gasoline marketing options. Similarly,
as per the gasoline marketing options of Section 7.3, costs and cost-
effectivenesses of each gasoline marketing regulatory strategy were
developed and are presented in Section 7.4.
7.2 INDIVIDUAL FACILITY COSTS
7.2.1 Bulk Terminals
Capital expenditures and net annualized operating costs for the
control of emissions from bulk gasoline terminal loading operations
have been estimated for four model plant sizes. The costs presented
for bulk terminals are updated costs from the 1981 costs presented in
the New Source Performance Standard (NSPS) background information
document for bulk terminals.! Further, capital charges reflect a
10 percent interest rate. The complete list of model plant parameters
is presented in Table 4-1.
The control strategy recommended for bulk terminals requires a
35 mg/liter control level be applied only to the approximately
500 terminals that are currently uncontrolled. Those terminals currently
with controls are operating below 80 mg/liter or 35 mg/liter (see
discussion on baseline emissions). As in a previous EPA study,2 it was
not considered cost effective to try to upgrade control systems meeting
80 mg/liter to a level of 35 mg/liter. As in estimating baseline
emissions, it was assumed that approximately 90 percent of the uncontrolled
facilities (450 terminals), regardless of model plant size, currently
practice bottom loading and 10 percent (50 terminals) practice top
splash loading. Therefore, within each model plant size, both top and
bottom load control costs for the base year were determined for three
7-2
-------
types of vapor processors (carbon adsorbers, thermal oxidizers, and
refrigeration systems) and presented in Tables 7-1 and 7-2.
Average bulk terminal control costs are summarized in Table 7-3.
Comparison of net annualized control costs for each vapor processor
type indicates that the thermal oxidizer would not be a cost-effective
alternative for the two larger model plant sizes in Tables 7-1 and 7-2.
In these cases, average costs were based upon only the carbon adsorber
and refrigeration systems to obtain representative capital and net annualized
operating costs. Average cost estimates for the two smaller model plant
sizes are based upon the average of all three vapor processor types.
Bulk terminal average weighted facility costs are presented in Table 7-4.
This breakdown by model plant and loading configuration, used in conjunction
with the average bulk terminal control costs in Table 7-3, provides a weighted
average net annualized cost in the base year of $34,600 per facility, a
weighted average capital cost of approximately $342,000 per facility and a
value of $63,700 for the weighted average recovery credit.
TABLE 7-4. BULK TERMINAL AVERAGE WEIGHTED COSTS
(THOUSANDS OF FOURTH QUARTER 1982 DOLLARS)
Model Plant % of Facilities
Size Each MP
(103 liters/day) Category
380 48
950 21
1,900 21
3,800 4
Average value
(weighted by MP)
Top/Bottom Adjustment0
Total
Net Annual 1 zed Costs3
Bottom Top
Load Load
55.9 141.2
41.7 167.6
-39.5 73.1
-139.5 0.8
24.2 128.4
21.8 ' 12.8
34.6
Capital Costs*
Bottom Top
Load Load
252 689
289 949
325 996
444 1,368
285 851
257 85.1
342
Recovery Credit3
Bottom Top
Load Load
18.8 21.9
47.0 54.8
141.2 164.7
282.4 329.4
62.7 73.1'
56.4 7.3
63.7
Average costs per model plant taken from Table 7-3.
Obtained by multiplying the average values weighted according to model plant by
the 90% of terminals assumed to practice bottom loading and 10.% assumed to be in
the top load configuration.
Gasoline consumption was projected in Section 4.2 to decrease with
time and this had a corresponding inverse effect on net annualized cost
figures. As recovery credits decrease, net annualized costs increase.
The weighted average recovery credit was decreased annually by a percentage
7-3
-------
Jl
-I
m
CO I I CD IT> Crt
^4 VO VO CM
in to .-H co
o «• o r-» i
*-- en i \o in en
• o n
J en \o
co i CM en tr> en
i u.
• .-* CT»
CO
to
^H o f*-en CM
C£ CO
o o
o o
. «^ r*. ^ «r o VD
i i UD in
> *f «H r*. CM
o cr>
r** in i in in C
O)
CO
<• t^ *^ co o *
cr
10 <•« P- 01
UJ O
«r —« to O e
- —4 CM tn
-g
CQ cn
I "inmr"
^tlOlf)<-4 O CM O CO
CO
I I
S.
g.
•s-se s.
—
o-ja c
' O U T-
^- jj trt t— M
J •*• V O "T—
ffl 3 ^- I
, O. W U - -
:o H- f-o t-o
7-4
-------
vo m *r o m r*) *
U.
u.
||
0 C
O t-
o"
CO
m
<
c_
u
u
Q
1
^•1
0 C
^
* *irt §1
SI- ~!
Or—
O O - «•
Q
O CM
C£ 05
z 2 L
O !
o fc. '
o>
Q •«-> i
•* *- x
<^ (O *°
/•y _«
c^ ; E
O JC g
^- o
"^ g
«C <*-
z o
S to '
UJ C
1— <0
t/i
-i 0 t
=> -^ t
03 t— , c
CM !
1 ^
^ -el
o - »
o
f-"*
"§
1— i
i
*j
3
Q.
•§,
3
0
w
J=
H-
i
^
O
to
CD
; r^, o>r\i *>
CO in tf> m
• ^^ —4 CO --
; 0--CM VO
~«-4 CO -^
! "^ O CM VO
CM en vo cn
CM •-* CO -^
m
f*- vo en -N
J ^^ CM PO VO
: — » ^M vo
CO
*
3 223*
pn
n^ p^ o* ••<
«! r— «-fnio
J, _,—..£>
«
! r*^ tA O> O
^ CM <*> ^
C
°I
3 «Jr- 0>O
- esj o «*> ^"
"1
51 «§S°
1 I T
L. m c^teo
J •*• CSJ CM C*
: i— v
3 tnr-Iio o
- -JOCVIKJ
•o;
i i vo vo VO O
S f> wt CM CM
J . ^4 ^4 W
MO) -^
0 4-> Ut
*rt ^ (/» U O
(U 4->
o aj u 4J > c en
u E u w» c O c
3u *3 3 C O y •—
Vt Q.^- u ^
1_ OJ Jrf «
a> -M 4-> ^£ U t-
e -^-*- o 3 «
ra — • c c m i- a
> 3 =D ai h- o
*> ^3
^- 3
^1 I
co i i co ui cn -i n* men rC
CM (*)
m o o ^- o r-* ^^
CM f-t VO O «• ' — '
CM <*•>
ro -^ eo «r o irt -^ ^ ^r
co icMcnmen vcovocn en
CM . VO CM VO CM
CM ro
.
—4t ivounvr r^. co co «r o
CM f*> tO CO VO f*
* _^ * ._ ... ^ _^ _^ .^^ j — .
•-« CM
-4 co r*- *r »- O CO r» CM
in i «M ^^. vo «r «er vo O •* vo
^* —4
vo r^^pw v -4m
.-i 1 1 VO iO CM VO VO CD CM ^*>
CM i i m u> m co *•
p*. «r i eo *• P*. o in
f^ tn in in CM r^. en vo i CM
CM v ro —
OM CM
CO^CDW^ -* ^m
•* '-^"^ sisss 5
"7 1 1 **^^^ *7 ^°1
en r < vo ir> *^ CM vo eo CM m
eo i CM er»^- o
en *— «o^7 « c o. i- u *•*
>,»- i— 6*—^- o &_ (Q fc
^ §•
trt
1 - ft
^ *m
0 Z *-
1- -x» (t)
--»**• c=
° " i
U M C
(A V
g £ '-
> 9) JX
> *-
"a w m
•»•* u
0 01
h- *»- ••
H- 41
ua u
4>* 3
0 °
tJ *
7-5
-------
Footnotes for Tables 7-1 and 7-2
i
Carbon Adsorption Unit.
Thermal Oxidation Unit
r* '
"Refrigeration Unit.
d
Cost of installing vapor collection equipment on existing bottom
loading tank trucks, $3,180 per truck.
a
"Cost of converting top loading racks to bottom loading and vapor
recovery, $200,000 per rack.
Cost of retrofitting existing top loading tank trucks with bottom
loading and vapor collection equipment, $6,784 per tank truck.
i
Electricity costs are based on average consumption rates reported by
manufacturers.
i
Propane for pilot burner estimated at 12.5 liters per hour, a.t $0.18
per liter.
Estimated activated carbon replacement period is 10 years, at $3,85
per kilogram carbon cost.
Estimated as 4 percent of unit purchase cost, plus annual rack vapor
collection maintenance of $200 per rack and $200 per terminal.
Daily system inspections at one hour per day, plus a monthly inspection
for liquid and vapor leaks in the vapor collection and processing
systems.
Cost to perform annual vapor tightness testing, including one-half
day downtime, $450 per truck.
Total capital investment x (capital recovery factor + 0.04), where interest
rate = 10 percent, equipment economic life = 10 years (0.163 capital recovery
factor).
i
Amount recovered per year, at $0.29 per liter assuming a density of
0.67 kg/liter.
Difference between uncontrolled and controlled emission level.
3
Cost effectiveness not calculated because net annualized cost is a
negative quantity (cost credit).
1
7-6
-------
n
e
X-. ' Q.
.a w» o
«• s- »-
re a
I— ^
co o
0 0
o
CM
—1 00
O Oi
f^* ^.
^b ^^
G3 Gi
_j *
«C 3
z o-
£ f
LU ^f "re
1— o
M— — *
•»! iR
rs w t>
CD T3 O
c 03
LU re
||
LU .C
<«—
,
CO
1
r*.
0>
.a
1C
^
CO
o to in CM to oo o — i
§o r«* m co to to
oj ~* CO -H ro r*.
o o to en *-4 to co »-*
o to co co to en r*» co
cn^ r-t — < to en co
en
o to«j-eno oeoo
in «9-cMPO«- intoen
en ** *-* 10 en -H — t
*^ «^
§-^ CM »o o en «-i «o
CO «^ CM CM CO ^* !"*>
CO ^^ ** ^- tO ^^
en in
i s • s s ? a a
eo^ CM -^ «• ^4 to
f* in
o| o to eo in en ic
o to co CM CM co CM
en^ — 4 _. ro i PO
•-^1
CM t^
o to ^ en en — < co
in . «; CM -< co «r to
en »-4 «^ , CM *•*
to en CM
o — < CM en CM m in
co co ^H ' in in to
CO 4-4 r* CM
^.
u
^
en
^-
in 4-> ^
O *> in >» it)
u in 0 C ^.
— 4-> o U Ol i—
>, in c u > o
re o o c o i.
•0 0 -- C O U M4->
--- 4J O ^- (U M4JC
f— 4J O) « T- U5 L. +t ino
c in >— ML. M oo
o a? re ,— i- >o o c_>
O •i3 U 4J > C Q. "O O
*»^ inL-incoos i" 01^
O13COU> re N
in > O. «- o +J ^. r-
3 i— . 4-> J-) ji O O Q.
CL -^i-U33 IO 3O
J= i— C C « S. t. o Cl—
cnio=5=3o:(—i— c
3 ** • — *
O -4— IO
S- Q. 4-> +J
f IO O O)
H- «J (- Z
^
2
_^
CO
co
in
in
oo
^
1
1
1
1
1
1
in
in
CM
to
in
CO
en
«*>
in
in
O)
^j
01
i«_
14—
LU
in
0
<8
o
CL
5
e
•o
IO
o
— ^_*
O in
4>* in
0 0
J3 U
eno
in 01
3-0
CLV
to en~"
~* i1^
1 5§
c w-
re u o
p— O
I— >. 01
o CL en
•o w
g "S I
2 M- *
"* o*»
o e:
*» o 5
en in
TS <0 Ol
0) t- u
4-* O) CL
re > 01
CSL
= €5
Oi in
• in o
4-1 Ol U
j_
Ol CL L.
U O) Ol
c ^ ^
oi i>
« 43 °
M- tn^-
Ol Or—
o 40 ja - ~
7-7
-------
equal to the percent decrease in gasoline consumption. Revised annualized
costs were then calculated for each year of reduced gasoline consumption
(see Table G-2). The projected consumption of gasoline is assumed to
be constant from the year 2000 to the year 2020, therefore recovery
credits and net annualized costs remained constant from the year 2000 on.
Data on the number of bulk terminals requiring control installa-
tions in each year were necessary since nationwide cost information was
generated from per facility costs. It was assumed (see Section 5.1.1)
that 50 percent of the facilities incurred capital costs in the initial
year of implementation (1987) and 50 percent the following year.
This 2-year phase-in of the capital costs was repeated every 10 years
(1997-1998, 2007-2008, 2017-2018) since 10 years was estimated as
the useful life of the control equipment. Top to bottom loading
conversion costs were assumed only for the initial installation on
those applicable facilities. Bottom loading costs were used for all
facilities when the equipment had to be replaced. Net annualized
operating costs were incurred by 25 percent of facilities in 1987, 75
percent in 1988, and 100 percent each year thereafter. This was
determined in the same manner as the emission reduction percentages for
bulk terminal phase in over 2 years.
Total cumulative net annualized costs were determined to be approxi-
mately $789 million. Net annual ized costs were discounted at a rate of
10 percent to obtain an NPV of $214 million. Capital costs were also
determined on both a cumulative and net present value basis and approxi-
mated $598 million and $221 million, respectively, during the period
1986-2020.
7.2.2 Storage Tanks
The cost analysis for storage tanks is based upon the control
approach of requiring the installation of an internal floating roof on
all existing fixed-roof gasoline tanks at terminals. The costs and cost
assumptions used in the storage tank analysis were obtained from a draft
Volatile Organic Liquid Storage Tank NSPS3 and a draft CTG for Control
of Volatile Organic Compound Emissions from Volatile Organic Liquid
Storage in Floating and Fixed Roof Tanks.4 These draft reports presented
costs in fourth quarter 1982 dollars so that updating the costs was not
necessary. The capital and annual ized costs to install an internal
7-8
-------
floating roof with a liquid-mounted seal on an existing fixed-roof tank
are presented in Table 7-5. Some of the capital cost estimates are a
function of the storage tank parameters which in this case are assumed
to be: 1) a volume of 2,680 m3 (16,750 b,bl), 2) a diameter of 15.2
meters (50 feet), and 3) a height of 14.6 meters (48 feet).
The largest single capital cost component involved in the installa-
tion of an internal floating roof is the basic cost of the roof at
$17,188. To make modifications needed to control emissions from existing
tanks, an estimate from two vendors was used to establish a relationship
for the costs to clean and degass the storage vessel. The "controls"
for deck fittings are gaskets for covers and sleeve seals for support
columns. To account for retrofit cost, the cost of additional work
(i..e., cutting roof vents, etc.) was estimated as 5 percent of the
installed capital cost of new construction.
The annualized cost without product recovery credits is calculated
by adding the annualized capital charges to the costs for taxes,
insurance, and administration (4 percent of the capital costs) and the
operating costs. Operating costs include the yearly maintenance charge
of 5 percent of the capital cost and an inspection charge of 1 percent
of the capital cost. Including product recovery credits, the net
annualized cost per facility was determined to represent a savings of
approximately $11,832.
The baseline analysis indicated that there were approximately 683
fixed-roof tanks at gasoline terminals (see Appendix B, Section B.2.2).
Total capital and net annualized costs were determined by using the
same method as for bulk terminals and bulk plants.
7.2.3 Bulk Plants
Capital investment as well as annualized costs are calculated for
loading operations at bulk plants for four model plant sizes. A
complete list of model plant parameters is given in Chapter 4.0.
Costs were estimated for each of the two bulk plant control options.
Capital cost estimates were developed for installation of a vapor
balance system with submerged loading of storage and truck tanks, for
both incoming and outgoing transfer of gasoline. Several past
studies5'5'7'8 were reviewed to obtain a range of costs for installing a
vapor balance system and submerged fill on incoming and outgoing loads
7-9
-------
Table 7-5. COST OF INSTALLING A BOLTED INTERNAL
FLOATING ROOF ON AN EXISTING FIXED-ROOF
(Fourth Quarter, 1982 Dollars)
Capital Cost & Installation
Degassing0
Basic Roof Costd
Liquid-Mounted Primary Seal6
Controlled Deck Fittings
Retrofit Adderf
Total Capital Cost
$ 7,515
$17,188
$ 124
$ 250
$ 859
$25,936
Annualized Cost
Maintenance (5%)
Taxes, Insurance, G&A (4%)
Inspections (1%)
Capital Charges9
Total Annualized Cost
Product Recovery Credit*1
Net Annualized Cost (Savings)
$1,297
$1,037
$ 259
$3.046
$5,640
$17,473
($11,832)
Tank parameters for fixed roof tank: Volume = 16,750 bbl = 2,680m3
Diameter = 50 ft = 15.2m
Height = 48 ft = 14.6m
References 3 & 4.
•*
"For an existing tank the first step is to clean and degass the storage
vessel, before installation of the floating-roof. Cost for this procedure
is based on the following relationship:
Total Cleaning Costs = 130.8 (tank capacity = 2,680)0.5132
Estimated from the equation:
D 3 tank diameter in meters.
Cost ($) = 1,069 D + 939 where
(References 3 & 4).
"The cost of the liquid mounted primary seal is estimated to be
$2.60 per linear meter of circumference.
Retrofit costs (installing an internal floating roof in an existing
fixed-roof tank) are the cost of additional work (i.e., cutting
roof vents) and is estimated as 5 percent of the installed capital
cost of new construction.
Capital recovery factor of 0.1176 is based on a 10 percent interest
rate and 20 year lifetime for the tank and floating roof.
Amount recovered per year, at $0.29 per liter assuming a density of
0.67 kg/liter. Based on emission factors between controlled [internal
floating-roof losses of (7.3285 x 1Q-8Q) + 2.4 Mg/yr, where Q is the
weighted throughput for all model plants] and uncontrolled (fixed-roof
breathing loss of 8.8 Mg/yr and working loss of 34.2 Mg/yr).
7-10
-------
for a typical bulk plant. Cost estimates were obtained for both top
loading and bottom loading systems for outgoing loads, A past survey
of bulk plants9 showed that an average of 91 percent of bulk plants use
top loading on outgoing transfer of gasoline and that the remaining 9
percent of bulk plants use bottom loading. This percent distribution of
top load vs. bottom load bulk plants was used to estimate a weighted
average capital cost for control of outgoing loads at a typical bulk
plant.
7.2.3.1 No Exemption Option. The "no exemption" control option
for bulk plants requires vapor balancing of storage tanks, and trans-
port and account trucks at all bulk plants. Uncontrolled plants will
incur the cost of balancing both the incoming and outgoing loads as
shown in the top half of Table 7-6. Weighted costs are presented to
account for both top and bottom loading systems for outgoing loads.
Some bulk plants already have a vapor balance system installed for
incoming loads. These plants would incur lower capital and annualized
costs in order to balance outgoing loads only as shown in the lower
half of Table 7-6.
Capital costs are assumed the same for each model plant size in
this analysis. The physical layout of the bulk plant (number of
storage tanks for gasoline, distance to loading rack, number of loading
arms for gasoline) does not vary significantly between large and small
bulk plants. Therefore, one control system and installation cost was
used to represent vapor systems installed at bulk plants. However, for
estimating net annualized costs, the throughput of the model plant was
used to calculate gasoline recovery credits for a balance system on
outgoing loads only (or on storage tank draining losses). A gasoline
recovery credit was not calculated for a balance system on incoming
loads (or on storage tank filling losses). This credit is realized by
the bulk terminal which receives the vapors during subsequent transport
truck fillings.
In order to calculate the nationwide cost impact of the "no
exemption" control option, it was necessary to determine the number of
uncontrolled bulk plants. The number of uncontrolled bulk plants was
estimated to be 8,040 (refer to Table 4-10 or Table 7-19). It was
assumed that 50 percent of the facilities incurred capital costs in the
7-11
-------
TABLE 7-6. AVERAGE CONTROL COSTS FOR BULK PLANTS (NO EXEMPTIONS)
(4th Quarter 1982 Dollars)
Model Plant Mo. 1234
Throughput (liters/day) 11,400 24,600 47,300 64,400
Weighted Average Top & Bottom Loading Costs
Balance Incoming &
Outgoing Loads on"
Uncontrolled Plants*
Capital Costb>c
Annual 0 5 H (3%)
Capital Charge (13.IS)
Taxes, Ins. (4%)
Recovery Credit^
Net Annual1zed
Emission Reduction (Hg/yr)
Cost Effectiveness (5/Mg)
Balance Outgoing Loads on
Plants with Incoming
Load Balanced"
Capital CostM 21,240 21,240
Annual 0 & M (31) 637 637
Capital Charge (13.15) 2,793 2,793
Taxes Ins. (45) 850 850
Recovery Cred1td 613 1,322
Het Annuallzed 3,666 2,957
Emission Reduction (Mg/yr) • 4.4 9.4
Cost Effectiveness ($/Mg) 839 314
28,540
856
3,752
1,142
613
5,137
8.1
634
28,540
856
3,752
1,142
1,322
4,428
17.5
253
28,540
856
3,752
1,142
2,543
3,207
33.6
95
28,540
356
3,752
1,142
3,462
2,288
45.3
50
21,240
637
2,793
350
2,543
1,737
18.1
96
21,240
637
2,793
350
3,462
817
24.7
33
Includes the cost of retrofitting two account trucks for use in vapor
balance service.
b
Top Load Cost - $19,490 (915), Bottom Load Cost - 538,970 (95).
References 5, 6, 7.
d
Recovery credits are based on a control efficiency of 90 percent on outgoing
loads from a balance system (or storage tank draining losses).
7-12
-------
base year 1987 and 50 percent the following year. This 2-year phase-in
of the capital .costs was repeated every 15 years, since 15 years was
estimated as the useful life of the control equipment. Net annualized
operating costs were incurred by 25 percent of facilities in 1987,
75 percent in 1988, and 100 percent each year thereafter. As with
terminals, revised recovery credits, and hence revised annualized
costs, were calculated for the control facilities for each year to
account for the decrease in gasoline consumption.
Total cumulative net annualized costs were determined to be approx-
imately $1,260 million. Net annualized costs were discounted at a rate
of 10 percent to obtain an NPV of $337 million. Capital costs were
also determined on both a cumulative and a net present value basis and
approximate $653 million and $252 million, respectively, during the
period 1986-2020.
The calculated cost effectiveness values for this control, option
are shown in Table 7-6 for each of the four model plant sizes. Cost
effectiveness is presented in units of dollars per megagram of emission
reduction and was determined by dividing the total annualized cost by
the emission reduction achieved by the control technique (i.e., balance
incoming and outgoing loads).
7.2.3.2 Size Exemption Option. The second control option would
require bulk plants loading less than 15,100 liters (4,000 gallons) of
gasoline per day to install vapor balance on incoming loads only
for storage tanks and transport trucks. An uncontrolled model plant
number 1 (
-------
Table 7-7. ESTIMATED CONTROL COSTS FOR BULK PLANTS (EXEMPT < 4,000 gal/day)
(4th Quarter 1982 Dollars)
Model Plant No. 1 2 3 4
Weighted Average Top and Bottom Loading Costs
Throughput (liters/day) 11,400 24,600 47,300 64,400
Balance Incoming Loads
and Install Outgoing Submerged
Fill on Uncontrolled Plants
with < 4,000 gal/day*
Capital Costb«c
Annual 0 & M (32)
Capital Charge (13.1%)
Taxes Ins. (4%)
Recovery Credit
Net Annual 1 zed
Emission Reduction (Hg/yr)
Cost Effectiveness ($/Mg)
To Install Outgoing
Submerged Fill on Plants
with Incoming Load
Balanced < 4,000 gal/daya
Capital Cos1^»c
Annual 0 5 M (3D
Capital Charge (13.1%)
Taxes Ins. (4%)
Recovery Credit
Net Annual 1 zed
Emission Reduction (Hg/yr)
Cost Effectiveness ($/Mg)
8,500
255
1,118
340
1,243d
469
6.6
71
1,200
36
158
48
1,243d
-1,002
2.9
—
28,540
356
3,752
1,142
1 ,322®
4,428
17.5
253
21,240
637
2,793
350
1,322®
2,957
9.4
314
28,540
356
3,752
1,142
2,5436
3,207 '
33.6
95
21,240
637
2,793
850
2,543*
1,737
18.1
96
28,540
856
3,752
1,142
3,462e
2,288
45.8
50
21,240
637
2,793
850
3,4626
817
24.7
33
a Includes the cost of retrofitting two account trucks for use in vapor balance service.
b Top Load Costs (91%) - $19,490. Bottom Load Costs (9%) - $38,970.
c References 5, 6, 7.
d Recovery credit is based on control efficiency of 58 percent for conversion
from top splash loading to submerged fill.
e Recovery credits are based on a control efficiency of 90 percent on
outgoing loads from a balance system (or storage tank draining losses).
7-14
-------
throughput less than 15,100 liters of gasoline per day. From a previous
study, it was determined that 42 percent of bulk plants have a throughput
less than 15,100 liters per day.10'11 Therefore, out of the 8,040
uncontrolled plants, 42 percent, or 3,400 of these bulk plants load
less than 15,100 liters of gasoline per day. Again, it was assumed
that 50 percent of these facilities incurred capital costs in the base
year 1987 and 50 percent the following year. This 2-year phase in of
the capital costs was repeated every 15 years, as 15 years was assumed
to be the useful life of the control equipment. Net annualized operating
costs were incurred by 25 percent of facilities in 1987, 75 percent in
1988, and 100 percent each year thereafter. Again, annualized costs
were adjusted each year to reflect the change in gasoline consumption.
Total cumulative net annualized costs were determined to be approxi-
mately $780 million. Net annualized costs were discounted at a rate of
10 percent to obtain an NPV of $208 million. Capital costs were also
determined on both a cumulative and a net present value basis and
approximate $462 million and $177 million, respectively, during the
period 1986-2020.
The calculated cost effectiveness values for this control option
are also shown in Table 7-7 for each of the four model plant sizes.
7.2.4 For-Hire Tank Trucks
Independent owners or operators of gasoline tank trucks transporting
gasoline from bulk terminals and bulk plants will also incur costs as a
result of the control options. The costs to these companies will
include the capital investment required to convert tank trucks to
bottom loading (where necessary) and to install vapor recovery equipment.
Annualized costs include maintenance of the vapor recovery equipment
and yearly vapor-tight tests on the trucks. A gasoline recovery credit
is not given to the for-hire tank trucks for installing vapor recovery
equipment. The credit is realized by the bulk terminal or plant that
receives the vapors during subsequent tank truck fillings. The costs
to the oil companies and bulk plant owners which operate tank trucks at
their own terminals and/or plants have already been included in the
cost analysis for bulk terminals and bulk plants.
7.2.4.1 For-Hire Tank Trucks at Terminals. The average cost of
converting tank trucks to bottom loading and adding vapor recovery
7-15
-------
equipment would be the same for a for-hire tank truck company as for a
tank truck owned by a bulk terminal. Bottom loading conversions average
about $3,604 per tank truck and the addition of vapor recovery provisions
requires an average expenditure of about $3,180 per tank truck as shown
in Table 7-8.12
In order to calculate the cost impact on the for-hire tank truck
industry, it was determined in Chapter 4.0, Section 4.1.4 that 5,800
for-hire tank trucks would be affected. Since it was estimated that
10 percent of the uncontrolled terminals use splash filling, 10 percent,
or 580 of the 5,800 affected for-hire tank trucks using splash-fill would
require both bottom loading and vapor recovery retrofitting. The capital
investment for these tank trucks would be $3,934,720 as shown in the upper
half of Table 7-8. The remaining 5,220 vehicles would already use sub-
merged fill and would thus require vapor recovery provisions only. The
total capital cost for these tank trucks would be $16,599,600 .as shown in
the lower half of Table 7-8. The total capital cost accruing to the
for-hire tank truck industry would be $20.5 million.
The annualized cost due to retrofitted tank trucks includes the
cost of maintaining the vapor recovery equipment and of performing an
annual vapor-tight test. Capital charges on the initial investment on
the equipment are also included. With maintenance costs at $1,000 per
year and tank truck testing at $450 per year,12 the total annualized
cost for 5,800 trucks would be $11.1 million.
In summary, the initial capital investment required by for-hire tank
trucks at terminals will be $20.5 million. The annualized cost for
these tank trucks will be $11.1 million per year. Fifty percent of the
capital costs were assumed to be spent in 1987 and 50 percent in the
following year. This 2-year phase-in, consistent with terminal controls,
was repeated every 15 years (the useful life of the equipment). After
the phase-in period, the annualized cost stayed constant throughout
the analysis since no recovery credits are associated with tank truck
controls.
7.2.4.2 For-Hire Tank Trucks at Bulk Plants. The average cost of
adding vapor balance equipment to tank trucks would be the same for a
for-hire tank truck company as for a tank truck owned by a bulk plant.
The addition of vapor recovery equipment for top loading systems at bulk
7-16
-------
Table 7-8. COST FOR THE FOR-HIRE TANK TRUCKS AT TERMINALS*
(4th Quarter 1982 Dollars)
No. of Affected Trucks
Bottom Loading &
Vapor Recovery
580
5,220
Capital Investment per Truck
Total Capital Costs
Capital Charges (13.1%)
Annual Maintenance/Testing
(@ $1,450 per truck)
Product Recovery Credit
Net Annualized Cost
6,784
3,934,720
517,312
841,000
NA
1,358,312
Vapor Recovery Only
Capital Investment per Truck
Total Capital Costs
Capital Charges (13.1%)
Annual Maintenance/Testing
(® $1,450 per truck)
Product Recovery Credit
Net Annualized Cost
3,180
16,599,600
2,182,412
7,569,000
NA
9,751,412
TOTAL CAPITAL COSTS FOR 5,800 FOR-HIRE TANK TRUCKS AT TERMINALS =
$20.5 million.
TOTAL ANNUALIZED COSTS FOR 5,800 FOR-HIRE TANK TRUCKS AT TERMINALS =
$11.1 million.
Reference 10 shows that bottom loading conversions average about
$3,604 per tank truck and that addition of vapor recovery is about
$3,180 per tank truck. Also, maintenance costs average $1,000 per year
and tank truck testing averages $450 per year.
7-17
-------
plants averages about $3,180 per tank truck. Bottom loading conversion
costs for bulk plant trucks were assumed to be the same as for terminal
trucks ($3,604 for bottom loading conversion and $3,180 for vapor
recovery equipment). It was assumed that the percentage of trucks requiring
bottom loading conversion was the same percentage as the bulk plants
converting to bottom loading (9 percent). Therefore a weighted cost per
truck was generated ($3,180 x .91 + $6,784 x 0.09 = $3,500). In Chapter
4.0, Section 4.1.4, it was determined that of the total 58,700 tank
trucks at bulk plants, 26,400 have vapor balance, and 16,000 are owned by
bulk plants and have been included in the bulk plant cost estimates. The
remaining 16,300 affected for-hire trucks will require vapor balance
provisions under the "no exemption" control option for bulk plants.
Therefore, the total capital cost accruing to the for-hire tank trucks
under this control option would be $57.1 million as shown in the upper
half of Table 7-9.
The annualized cost due to retrofitted tank trucks includes the
cost of maintaining the vapor recovery equipment and of performing an
annual vapor-tight test. Capital charges on the initial investment on
the equipment are also included. Thus the total annualized cost in the
base year for 16,300 tank trucks under this control option would be
$31.1 million.
Of the 32,300 tank trucks without vapor balance, it is assumed
that 58 percent, or 18,700 tank trucks, load at bulk plants with a
throughput greater than 15,100 liters per day, since 58 percent of the
bulk plants have a throughput greater than 15,100 liters per day (see
Table 4-2). Of the total 18,700 tank trucks loading at bulk plants-
with a throughput greater than 15,100 liters per day, 9,300 vehicles
are owned by bulk plants (58 percent of the 16,000 uncontrolled tank
trucks owned by bulk plants). The remaining 9,400 affected for-hire
trucks will require vapor balance provisions under the control option
which exempts bulk plants with a throughput less than 15,100 liters per
day.
The initial capital cost accruing to the for-hire tank trucks under
this control option would be $32.9 million as shown in the lower half
of Table 7-9. The annual!zed cost due to retrofitted tank trucks
includes the same maintenance and annual testing costs as outlined in
7-18
-------
Table 7-9. COST FOR THE FOR-HIRE TANK TRUCKS AT BULK PLANTS
(4th Quarter 1982 Dollars)
NO EXEMPTIONS
No. of Affected Trucks
16,300
Weighted Average Costs - Top or Bottom Loading and Vapor Recovery
Capital Investment per Truck3
Total Capital Cost
Capital Charges (13.1%)
Annual Maintenance/Testing
(@ $1,450 per truck)
Total Annualized Cost
EXEMPT BULK PLANTS < 4,000 gal/day
No. of Affected Trucks
3,500
57,050,000 or $57.1 million
7,500,579
23,635,000
31,135,579 or $31.1 million
9,400
Weighted Average Costs - Top or Bottom Loading and Vapor Recovery
Capital Investment per Truck3
Total Capital Cost
Capital Charges (13.1%)
Annual Maintenance/Testing
(@ $1,450 per truck)
Total Annual ized Cost
3,500
32,900,000 or $32.9 million
4,325,487
13,630,000
17,955,487 or $18.0 million
Top Load Costs (91%) - $3,180, Bottom Load Costs (9%) - $6,784.
7-19
-------
Section 7.2.4.1. The total annualized cost for 9,400 tank trucks would
be $18.0 million as shown in the lower half of Table 7-9. As with
previous analyses, a 2-year phase-in of controls was assumed. After
phase-in, capital costs were repeated every 15 years based on the
useful life of the equipment. The annualized cost remained constant
since no recovery credits are involved.
7.2.5 Service Stations
7.2.5.1 Service Station Stage I. Capital costs and installation
costs of Stage I controls were obtained from two sources.13'14 Average
capital and net annualized costs are presented in Table 7-10 as $1,698
and $341, respectively (4th quarter 1982 dollars). Since the number of
underground storage tanks and the amount of piping at service stations
does not vary considerably with throughput (storage capacity would vary
more), costs to comply with Stage I at facilities were assumed to be
independent of facility size. A gasoline recovery credit is not given
to the service station for Stage I control. The credit is realized by
the bulk plant or terminal receiving the vapors upon subsequent tank
truck fillings.
Data on the number of service stations requiring Stage I control
installations in each year were necessary since nationwide cost infor-
mation was generated from per facility costs. Four control scenarios
were developed for installation of Stage I systems at service stations
- two nationwide options (with and without size exemptions) and two
nonattainment area options (with and without size exemptions). Under
both the nationwide and nonattainment options, service stations with an
average throughput of 37,900 liters per month were exempted from Stage
I requirements. For the nationwide options, it was assumed (see Section
5.1.1) that 50 percent of the facilities incurred capital costs in the
base year 1987 and 50 percent the following year. This 2-year phase-in
of capital costs was repeated every 15 years (2002-2003, 2017-2018),
since 15 years was estimated as the useful life of the control equipment.
Net annualized operating costs were incurred by 25 percent of facilities
in 1987, 75 percent in 1988, and 100 percent each year thereafter.
This was determined in the same manner as the emission reduction
percentages for a Stage I service station nationwide phase-in over 2
years. The nonattainment options assumed a 1-year phase-in for capital
7-20
-------
Table 7-10.
SERVICE STATION STAGE I CAPITAL AND
MET ANNUALIZED COST ESTIMATES3
(4th Quarter 1982 Dollars)
Capital Cost and
Installation
1,698
Annualized Costs
Maintenance (3%)
Taxes, Insurance
and G & A (4%)
Capital Chargesb
(0.131)
50.9
67.9
223
Annualized Cost
Recovery Credit
Net Annual ized Cost
342
MA
342
$/Mg, Cost Effectiveness0
MP1 (18,950 liters/mo.)
MP2 (75,800 liters/mo.)
MP3. (132,650 liters/mo.)
MP4 (246,350 liters/mo.)
MP5 (701,150 liters/mo.)
1,380
345
197
106
37.2
References 11, 12.
D
Capital charges are based on a 10 percent interest rate and on
equipment life of 15 years.
Since the number of underground storage tanks at service stations
do not vary considerably with throughput (storage capacity would
vary more), costs to comply with Stage I at affected facilities
were assumed to be independent of facility size.
Sample emission reduction calculation:
(1130-40) ing
TTter
246,350 liters
mo.
12 mo. Mg
yr. 109 mg = 3.22 Mg/Yr
7-21
-------
costs beginning in 1986. Net annualized operating costs were incurred
by 50 percent of facilities in 1986 and 100 percent each year thereafter.
Capital costs and net annualized costs were calculated for the
four Stage I control options described. Costs were determined both
as a cumulative total and as a net present value (NPV) of total costs
incurred from 1986 to 2020, discounted at a rate of 10 percent. Stage I
control costs are presented in Table 7-11.
TABLE 7-11. STAGE I CONTROL COSTS
(Millions of 4th Quarter 1982 Dollars)
Regulatory
Strategy
Nationwide (NO EX.)
Nationwide (EX.)a
NAb areas (NO EX.)
NAb areas (EX.)a
Capital
Cumulative
1,008
423
35
15
Costs
NVP thru
2020
378
159
14
6.4
Net Annual
Cumulative
2,234
938
81
34
ized Costs
NVP thru
2020
590
248
24
10
a
Size exemptions to service stations with a throughput <37,900 liters
per month.
NA s Nonattainment.
7.2.5.2 Service Station Stage II. Cost data were obtained from
three references15'16*1? for service stations ranging in size from two
nozzles (19,000 liters/month throughput) to 16 nozzles (680,000 liters/month
throughput). These costs are representative of each of the three Stage II
control systems currently in use (i.e., vapor balance, hybrid, and vacuum
assist) and were updated to fourth quarter 1982 dollars using cost
indices.16 The data were grouped according to the service station model
plant characteristics detailed in Table 4-4 (i.e., number of nozzles and
monthly throughput range). Estimates of service station control costs
by type of Stage II system are based on the average of these data and
presented in Table 7-12. Capital and net annualized cost ranges from
which these averages were obtained are also presented in Table 7-12.
7-22
-------
1
(S
1— *—
CO "*
oo ce
f*
UJ "o
D. O
co CM
h-CO
CO CTi
Oi-«
o
t.
H* «
»
UJ
t r** CTt
CO « •
to o
10
CM 10
O^CM
s
_
>t
•—' —
• o
sz
e ^*"
-X.
M C
(/» i. O
^!«J«
• N -P- O
O Mr- 3
z o — -o
4J Z 4-> OJ
C W- 3
« O a. c
r** .O O v>
•S i-fei
0 Z t— UJ
o"
2 -~-
- . tn r-» in co O *o 9)
co eo ^-i f-t ^-H -H o
CM O
*-< cn
CO P^
en ^ ^ ^«j ^
en i co r*- CM r*- ~+ i J •* "^^, ^
in CM ~* en *-*
4* VO
ol^O^r^S
O 1 in (O CM ••* CM^^-t CM^
CM f-
«• en
•^*"
2 ««
S^-o c
w u S>H:5"S -^
o— «OJ= «ACJ*—*-— a>
goa)oc«e» s 2» «*-
4Jr-C: r- ^ U C C UJ
10 "o.*"* C "S. X S -M *^ O
c •
10 v^
CD **
o
CM — •*
\ocM"WCMPom-m«-<
CO CO ^-t CM d O Cft
— « cn^ CM m
U)
s
CM 10
o
en ~-< en «tr o co 10 CM m
H^O • " *^D
••H CM O
o* *—
v4
_
o
*^ ^^
10 If)
• en
m o m
^~— c
v> S>S5'S —
o«— «e 5 viol^— >=S> t
5 "ro «i— "<8 - cu = « *"
V) 4^ CC (B 4^ VI > ^ C£ 4>*
>«^*^>' 3 <*v V O ^" (A
t/VO. CQ.X04-> O
10 c ic « o a) o
7! o>
1 S
S -
CM ^O
* U)
•I'vigisl1? s
—4 o *-« PO CM m PO
CM O^ P^
CO* —
CM in
»o i ^ in PO *o vo i r*.
«eo * *«r *
in to^ CM PO CM *-«
PO *-*
J_.
U [ *•»
in^ 10
vo CM «-< *o m to ^- ^4
to CM * *o *
•-4 • CM
^
3
s. s
to co
o ieM«fentocn«r. in
CM^CMCOOPnr** t CM
-2 ~ "S
' .8 ~
to o
• ^*
co — 4 r» en co 10 o» m »o
i*r ^eo ""j
CM «• CM MIO M
»-* en o — *
o —
^
r^
2
O VI
S.<~ a)
e »ft 4-> C
4^ (A O**— *O OJ >
(S* o •— «e j= w o ^ *— u
** ° ^*«= ° "^ b 1 I1 S
J2'«fl3f-.'« *aie(o uj
v»^ ^— s v a> o •— **
^ A c O. X U •** IA
0 c *^-
,
^
0)
c
CO -p-
X J3
_g o
e a>
*/>
O «A
CJ 0)
+•» F—
c «
« >
Z at
en
Ul 0
u *-
a> >
J= 10
„ |
1 1
**• in
e> «-<
CM 2
2 S
u
L. C
4J fc.
w at
1 £
.c e
t.
*£ £
i-
2 a.
^ L.
S V,
-§ U
0. 0
• o
- c"
'
" 5
u c
U i-
«j" *
•o .
i
5
VI
>)
o
u
«3
p
1^
<0
a>
>>
k
0)
B.
01
^2
*" S
a>
i2 o
£ v>
uj a
o
" 5
°!
3 41
•C O
1
Vi
€
4-*
>»
•o
"C
*
* «2
01 4J
-P
* £
•S£
A e
0) """
je^--
fel
ll
in PO
«s
•^ 0)
u in
| J
*o w
*> V?
e
0 4->
U VI
e *v)
O VI
IO
"S E
u =
A U
o) 2
10 V
+>
^£
^
iS
s a
•0"°
7-23
-------
The cost of installing vapor recovery systems at service stations
can vary greatly, depending on the station throughput, station layout
and the number of nozzles and pumps (see Table 7-12). The average
capital cost for equipment and installation of balance systems ranges
from $5,088 to $13,709 depending on the size of the service station.
The average cost of equipment for vacuum assist also varies with station
size. The cost of these systems is usually the highest due to the
secondary processing unit required and ranges from $12,231 to $21,082.
The hybrid system is basically a vapor balance system with liquid
aspirator or jet pump equipment added. The additional cost for
the equipment puts this system in the middle between vapor balance
systems and vacuum-assisted systems.
In calculating net annualized costs, a credit of $0.29 per liter
of gasoline recovered during refueling was assumed. The procedure
followed for determining Stage II recovery credits is detailed in Table
7-13. Capital charges were based on an interest rate of 10 percent and
an equipment life of 15 years for vapor balance and hybrid systems and
8 years for vacuum assist systems. Annual operating and maintenance
(O&M) costs reflect a nozzle replacement charge of $100 each, once per
year for balance systems and once every two years for vacuum assist and
hybrid systems. Installation costs for the balance system include the
installation of liquid blockage sensors with an average cost of $100 to
$150 per dispenser. Costs do not include installation of over-head
hose retractors.
The weighted average Stage II control costs, presented in
Table 7-14, were determined by using an average cost weighted by the
population of Stage II systems currently installed - 80 percent vapor
balance systems, 15 percent hybrid systems and 5 percent vacuum assist
systems.^ Emission reductions (Mg/yr) and cost effectiveness figures
($/Mg) per model plant size are given both for Stage II costs per
system and for weighted average system Stage II costs in Tables 7-12
and 7-14, respectively.
Additional cost data were obtained from the California Air Resources
Board (ARE)20 and the American Petroleum Institute (API).21 These cost
estimates reflect assumptions which differ slightly from those used in
7-24
-------
Table 7-13. STAGE II RECOVERY CREDIT CALCULATIONS
Balance and Hybrid Stage II Systems
Emission factors: Displacement losses = 1080 mg/liter
Breathing losses = 120 nig/liter
Assume 95% recovery of displacment loss and 50% recovery of
breathing loss
Recovery factor = (1080 mg/liter)(.95) + (120 mg/liter)(.50) = 1086 mg/liter
Example:
1,086 mg x 37,900 liter x kg__ x liter x 12 mo. x $0.29
liter mo. 10° mg .67 kg .yr liter
= $214 per year recovery credit.
Equation reduces to:
[Throughput (liter/mo)] x 0.0056 $ - mo. = $ per year
liter-yr
Vacuum Assist Stage II System
Assume 50% recovery of displacement loss and 50% recovery
of breathing loss
Recovery factor = (1080 mg/liter)(.50) + (120 mg/1iter)(.50) =600 mg/liter
Example:
600 mg x 37.900 liter x kg x 1iter x 12 mo. x $0.29
liter mo. iuo mg TBTTg yr liter
= $118 per year recovery credit
Equation reduces to:
[Throughput (liter/mo)] x 0.0031 $ - mo. = $ per year
liter-yr
7-25
-------
Table 7-14. WEIGHTED AVERAGE STAGE II COSTS3
(Fourth Quarter 1982 Dollars)
Model Plant No.
Throughput (liter/mo.)
18,925
75,700 132,475 246,025 700,225
Emission Reduction (Mg/yr) 0.25
0.99
1.73
3.21
9.13
Capital Costs
Annual 0 & M
Capital Charges13
Taxes, Ins.
Recovery Credit
Net Annual i zed
$/Mg, Cost Effectiveness
5,654
495
778
226
104
1,394
5,653
6,098
621
837
244
417
1,285
1,302
6,604
826
906
264
731
1,266
734
9,766
1,068
1,328
391
1,357
1,430
446
14,790
1,789
2,004
592
3,861
522
57
Stage II control costs for all three systems were determined by using an average
cost weighted by the population of Stage II systems currently installed (i.e.,
80% balance, 15% hybrid, and 5% vacuum assist).17
Capital charges are based on a 10% interest rate and an equipment life of 15 years
for balance and hybrid systems, and 8 years for vacuum assist.
7-26
-------
this analysis. The API costs, given in 1983 dollars, include use of
high-hang retractors, installation and annual permit fees, certification
fees, and annualization based on a 7.7 percent interest rate. Comparison
of capital costs is difficult because representative throughput ranges
are not given; however, the API estimates appear to be much higher than
those presented in Table 7-12. Net annualized costs were generated by
API on a nationwide basis and a determination of per facility costs
from available data was not possible. For comparison, ARB data are
presented in Table 7-15. ARB costs are based on a 12 percent interest
rate and are given in 1979 dollars. A comparison between these costs
and the average costs presented in Table 7-12 is difficult because the
nozzle configurations and throughput ranges representative of the model
plants chosen by ARB are not entirely consistent, with the model plant
characteristics of this analysis. Given the corresponding throughput
ranges, the ARB costs given in Table 7-15 for the 6 nozzle stations
should be compared with model plant 3 of Table 7-12; the 9 and 12
nozzle cases closely approximate the parameters of model plant 4.
ARB capital costs for all systems range from approximately 5 percent
lower to 20 percent higher than the costs presented in Table 7-12. The
ranges of net annualized costs reported by ARB closely approximate the
ranges presented in Table 7-12.
For ease of calculation, average costs were developed so that
future costs could be generated based on loss of throughput due
to decreasing gasoline consumption. The average weighted Stage II
system costs per model plant size (Table 7-14) were adjusted by
percentages representative of each of the five model plant categories
given in Table 4-4 (i.e., 58 percent, 17 percent, 15 percent, 8 percent,
and 2 percent for model plants 1 through 5, respectively), thus obtain-
ing an average weighted net annualized cost, capital cost, and gasoline
recovery credit per facility. Table 7-16 presents these average weighted
costs and credits on a per facility basis for calculation of proposed
control options with and without size exemptions. Service stations
exempted from Stage II control requirements were independent dealers
averaging less than or equal to 189,000 liters per month throughput and
all service stations with an average throughput of 37,900 liters per
month or less.
7-27
-------
TABLE 7-15. CALIFORNIA AIR RESOURCES BOARD (ARB) STAGE II
CONTROL COSTS
(1979 Dollars)16
Model Plant No.
No. of nozzles
Throughput range
(1i ters/mo.)
3
6
4
9
4
12
(113,700-379,000) (189,500-379,000) (265,300-379,000)
Capital Costs
Balance System
Hybrid System
Vacuum-Assist
Net Annual ized Costs3
Balance System
Hybrid System
Vacuum Assist
7,372
9,618
15,502
11-1,324
327-1,641
2,826-3,397
9,133
11,962
17,813
653-1,591
1,043-1,981
3,585-3,992
10,844
14,332
20,074
-
1,287-1,850
1,762-2,325
4,340-4,584
aNet annualized costs are given as a range to reflect the throughput range
(113,700 - 379,000 liters/mo) of the costs reported.
7-28
-------
TABLE 7-16. SERVICE STATION STAGE II WEIGHTED AVERAGE COSTS
(Fourth Quarter 1982 Dollars)
Cost Factor
1
Throughput 18,925
(liters/mo.)
NO EXEMPTIONS
% each MP category 58
Net Annual i zed Cost 1,394
Recovery Credit 104
Capital Costs*
1) 5,654
'2) 12,231
3) 5,308
EXEMPTIONS (< 37,900 liters/mo.'
% each MP category
NIb
I
Net Annual 1 zed Cost
MI
I
Recovery Credit
NI
I
Capital Costs - NI
1)
2)
3)
Capital Costs - I
1)
2)
3)
Model Plant Number
234
75,700
17
1,285
417
6,098
12,720
5,749
132,475 246,025
15
1,266
731
6,604
13,661
6,233
all stations and
40 36
1285
417
6098
12,720
5749
-
1266
731
6604
13,661
6233
-
8
1,430
1,357
9,766
15,768
9,450
Weighted Av.
5 Value
700,225
2
522
3,861
14,790
21,082
14,459
189,500 liters/mo.
19 5
80 20
1430
1430
1357
1357
9766
15,768
9450
9766
15,768
9450
522
522
3861
3861
14,790
21,082
14,459
14,790
21,082
14,459
NA
NA
1,342
427
6,384
12,989
6,036
for independents (I))
NA
NA
1270
1249
872
1858
7391
14,035
7042
10,771
16,831
10,452
Indicates installation of 1) balance, hybrid, and vacuum assist systems,
2) vacuum assist system alone, and 3) balance and hybrid systems.
bNI denotes non-independents, I denotes independents.
7-29
-------
Gasoline consumption is expected to decrease with time and this
will have a corresponding inverse effect on net annualized cost figures.
As recovery credits decrease, net annualized costs will increase. The
average recovery credit (with and without size exemptions) was decreased
annually by a percentage equal to the decrease in gasoline consumption
and the amount of decrease in each year was added to the average weighted
net annualized cost. The projected consumption of gasoline is assumed to
be constant from the year 2000 to the year 2020, therefore recovery cre-
dits and net annualized costs will remain constant from the year 2000 on.
In this manner gasoline recovery credits and net annualized costs
were determined on a "theoretical" basis, that is, assuming that the
Stage II equipment is capable of achieving control efficiencies
approaching theoretical rates. In order to determine credits and costs
at actual "in-use" efficiencies, it was necessary to take the recovery
credits previously adjusted for the decrease in gasoline consumption
and apply the following equation:
In-use efficiency
In-use credits = Theoretical credits x Theoretical efficiency
The in-use efficiencies for Stage II systems under State and Federal
enforcement scenarios are detailed in Chapter 3.0. The theoretical
control efficiency of Stage II systems is 95 percent.
Stage II recovery credits and net annualized costs were also
adjusted for control options combining both Stage II and onboard
technologies. When these technologies are used in combination, the
onboard system dominates and is credited with vapor recovery (note that
no monetary value is assigned to the onboard recovery credit as it is
assumed that any recovery credit is more than offset by the additional
burden of carbon canister weight on the vehicle). In cases where
Stage II systems are used in combination with onboard, the recovery
credits previously adjusted for decrease in gasoline consumption were
further adjusted by the percentage of onboard controlled consumption.
In other words, if onboard technology were estimated to control
10 percent of gasoline consumption in a given year, then the Stage II
recovery credit would be 90 percent of the value determined from
application of Stage II technology alone.
7-30
-------
In summary, recovery credits and corresponding net annualized costs
were adjusted on a per year basis, (see Appendix G, Table G-2) in
consideration of the following factors:
1) Gasoline consumption experienced a gradual decline until the
year 2000 when a constant consumption was assumed,
2) Control scenarios in which Stage II and onboard control
technologies are combined reflect the percentage of onboard-
controlled consumption in each year and result in lower Stage II
recovery credits and higher net annualized costs, and
3) Actual "in-use" Stage II system control efficiencies are less
than theoretical levels, resulting in lower recovery credits and
higher net annualized costs than theoretical cases.
Since the equipment lives of Stage II systems vary (15 years for
balance and hybrid systems and 8 years for vacuum assist), it was
necessary to determine capital cost weighted averages from three phase-in
scenarios: (1) initial capital cost estimates would include installation
of all three types of Stage II systems, (2) after each 8-year cycle
only vacuum assist systems (5 percent of total facilities) would require
replacement and (3) each 15-year period from the date of initial instal-
lation, both balance and hybrid systems (95 percent of total facilities)
would be in need of replacement. All weighted average costs, with and
without size exemptions, are presented in Table 7-16.
The number of affected facilities under each of the control options,
as well as nationwide costs, is detailed in Section 7.3. The facility
phase-in schedule, used to determine capital costs and net annualized
costs under both nationwide and nonattainment control scenarios, is
similar to the schedule presented for service station Stage I control
installations.
7-2.5.3 Onboard Controls. This section summarizes the appropriate
cost sections on onboard control costs presented in Appendix C of this
report. As shown in Table 7-17, an onboard vapor recovery system is
expected to carry a consumer cost of $13.32 for light-duty vehicles
(LDVs) and $18.19 for light-duty trucks (LDTs). Those LDTs using dual-
fuel tanks (approximately 20 percent) may require two separate onboard
control systems for a total cost of $36.38. This is a conservative
7-31
-------
TABLE 7-17. ONBOARD VAPOR CONTROL HARDWARE COSTS
(1983 dollars)
Incremental Costs
Component or Assembly
Charcoal Canister LDV/(LDT)
Purge Control Valve
Liquid Vapor Separator
Fill pipe Seal
Pressure Relief Valve
Hoses/Tubing
Miscellaneous Hardware
Vehicle Assembly
Systems Engi neering/Certification
LDV Totals: Vendor
LOT Totals: Vendor
Vendor
$3.99/(7.83)
0.74
0.71
1.12
0.44
1.90
0.40
—
—
$9.30
$13.42
Retail Price
$5.077(9.94)
0.94
0.91
1.42
0.56
2.41
0.51
1.00
0.50
Retail $13.32
Retail $18.19
7-32
-------
assumption since costs would likely be reduced by using one large
charcoal canister rather than two separate canisters.
A fleetwide estimate for all LDVs and LDTs can be determined by
sales weighting the costs given above. Using the projected sales for
1988 from Appendix C, and assuming 20 percent of LDTs have dual-fuel
tanks, the fleetwide average cost is calculated to be $15.08 as shown
below. For future calculations, this cost will be rounded to $15 per
vehicle.
(10.45 M) ($13.32) + (.8) (2.778 M) ($18.19) = (.2) (2.768 M) ($36.38)
= $15.08
The $15 per vehicle is believed by EPA to be the average
incremental cost above the cost of the present evaporative control
systems. However, there are reasons why the cost of onboard control
systems could be somewhat greater (approximately $25), see Appendix
C, page C-23.
Note that these total cost estimates per vehicle include fill pipe
seal and pressure relief valve modifications. The net annualized
onboard control cost was obtained by multiplying the capital cost by
a capital recovery factor. This capital recovery factor represents a
10 percent interest rate and an onboard control system equipment life
of 10 years for light-duty vehicles and 11 years for light-duty
trucks (the lifetime of the vehicle). A gasoline recovery credit is
not given. Available data indicate that fuel economy is not enhanced,
as any potential fuel credit will be offset by the additional weight,
of the onboard control equipment.
The control strategy proposed assumes that onboard controls be
implemented beginning in 1988. Data on the number of light-duty
vehicles (LDVs) and light-duty trucks (LDTs) requiring onboard control
in each year were necessary since nationwide cost information was
generated from per vehicle costs. The number of new vehicles per
year, as well as vehicle scrappage rates, were determined in a manner
similar to onboard emission reductions (see Section 4.2, Gasoline-
7-33
-------
Consumption and Model Plant Projections). The nationwide costs of
onboard controls are presented in Section 7.3.
7.2.5.4 In-Use Costs. As a further comparison between Stage II
and onboard control options, an analysis was performed to evaluate the
costs incurred when considering in-use efficiencies rather than -
theoretical efficiencies. Section 3.7.3 discusses in-use efficiencies
in detail. In-use efficiencies affect only the recovery credit component
of net annualized cost estimates (i.e., recovery credits decrease and
net annualized costs increase); capital costs would remain unchanged
from the theoretical case.
Table 7-18 summarizes the cost and cost-effectiveness for nationwide
Stage II control options based upon in-use efficiencies associated with
annual inspections (86 percent) and minimal enforcement (56 percent).
These costs and cost-effectivenesses can be compared to the nationwide
estimates obtained from onboard controls, considering in-use efficiencies
(92 percent). Table 7-18 also evaluates the effect on cost and cost
effectiveness if the additional emission reduction associated with
control of excess evaporative emissions with onboard controls are
considered (see discussions in Section 3.7.3). Cost-effectiveness is
shown in Table 7-18 for both cumulative values (cumulative dollars/cumulative
emissions) and discounted values (NPV dollars/NPV emissions).
7.3 NATIONWIDE COSTS OF CONTROL OPTIONS
Costs were derived for each of the industry sectors under both size
exemption and no exemptions options. These control options were then
used to develop the costs for the regulatory strategies discussed in
Chapter 4.0. The nationwide costs of the control options were
determined* by using the individual facility costs discussed earlier and
the total number of facilities requiring control in each of the years
evaluated (1986-2020). As discussed in Section 4.2.2, the number of
facilities was considered constant throughout the analysis; however,
there were different phase in schedules for different control options
(see Section 5.1.1 for a more detailed discuss of facility phase in).
Table 7-19 illustrates the number of facilities requiring controls for
each of the industry segments.
7-34
-------
CO
1
o
i—
o
o
Q
2
o
CD
O
o
Z CO
M *«/>
« 0 S-
^•4 SS ^O
UJ P-
C3 O O
•f. t-i Q
CO Ll_ CM-—
UJ CO O
z en CM
LU LU ^H O
i— i a> \o
ui z f»co
CO t— i fc. O>
(C «-4
Z CD =3— •
o z o>
CO H-4
O£ LU 4*
01
^
U
0>
UJ
1
+>
o
CJ
s
s
C/l
«
O
•o
s
«T*
^
a
e
+j
jg
^
a.
Z£
Id
CO
en
I-M
CJ
TT
IO
§
3
•5
o cfi mr~ «a-p~
-H-H-^CMCMCM -^ -^
o oo o o o oo oo
CM «— 1 O *O Cn 10 O^ CMO
«-CM ^rr». meo
•H ^^ — < CM CM f> , ^H — <
eo r— 10 en-«rto oo oo
cMeno icto icie
ro m m r— o to «oi0 «oio
10 10 IA co en in en en en en
^•1 »"4 «^
inioiomieio enei CM CM
eneoineneou) enen enen
44 4^
M c «i e
CO) C CD O) CA 9 Cn
op o E cc ce
4^CJ **O ^"'^" ^^
O) UU UU O) CJ H) S)
•O O) O CO O 33 S3
^- 0.1k 0.1k «.lk M- Ik
% cne ME; cuoi« oioi*
C r«- C LU «> ^ CUJ r*L.S*Q. WkO>
OIBM C 10 •— <0 « «5
_ u r— O tj I— UOICU> OIOI>
4-> in— r— IB —~- i— •> — >— >— CJ 1— I— CJ
IO C 4J «3 E *•> *-" «J E -I-1 O O O U
Z OO1O1 3^ O-CUOI 3-^- O)-r- >,i- .C O> -^- >,-^ f
~- t- tn c c EWM c c v. .c r— f 4J w ^ .— .c *J
i 4JO=> e: — cj o =3 c — o cj c cj — =5 cj c o> -.-
O.OII *C. X X CJ 1 «C £ CJ > O >. S t>O>X
i-^ E.CCUJJSC: .c c
•— i 35 H- i-* 1 1 H-*— * 1 I^H-I 1 k4i I
X 0 i.
OIUJOO SEOO IO O O
en 0
re i i .0
-u c
co o
_
c
Ol
0)
0.
o
C IB
O
"- -0
4 •
in Q)
•»- 3
V IB
CJ
CJ
e in
O 01
L.
-o o.
in *>
IB 0)
ea z
IB ^3
7-35
-------
Table 7-19. NUMBER OF FACILITIES REQUIRING' CONTROLS
FOR GASOLINE MARKETING OPTIONS
FACILITY TYPE
TOTAL FACILITIES
MP 1
MP 2
MP 3
STORAGE TANKS AT TERMINALS 630
F0R-HIRE TANK TRUCKS
- Teralnal Trucks 5,800
- Sulk Plant Trucks
(Ho Exemptions) 16,300
- 3ulk Plant Trucks
(< 4000 Gal/Day 8P Exempt) 9,400
MP 4
MP 5
TERMINALS
8UU PLANTS
- No Exemption
- Exempt Bulk Plants
< 4000 Gal /Day
SERVICE STATIONS
;(ESHAP Alternatives
STAGE I
- Ho Exemption Option
- Exempt Stations
< 10,000 Gal/Month
STAGE II
- Ho Exemption Option -
- Exempt Stations < 10,000
Sal /Mo and Independent
Stations < 50,000 Gal /Ho
-Independents Only
CTG Alternatives
STAGE I
- NO Exemption Option
- Exempt Stations < 10,000
Gal /Ho
STAGE II
- Ho Exemption Option
- Exempt Stations < 10,000
Gal /Ho and Independent
Stations < 50,000 Gal/Ho
- Independents Only
500
3,040
3,040
197,900
33,100
383,200
120,600
6,895
2,395
114,900
36,200
240 135 105 20
3,400 4,000 560 80
3,400 4,000 560 80
114,800 33,600 29,700 15,800
— 33,600 29,700 15,800
222,100 65,400 57,500 30,700
47,500 34,900 30,700
10,500
4,000 1,170 1,035 550
1,170 1,035 550
66,600 19,600 17,300 9,200
14,300 10,500 9,200
3,150
—
--
—
4,000
4,000
7,500
7,500
2,700 .
140
140
2,200
2,200
800
7-36
-------
Table 7-20 presents the nationwide costs ,for each of the control
options. Cumulative costs represent the simple summation of the total
capital and annualized costs over the period 1986-2020. The net present
value (NPV) costs represent the 1986 present value of all costs over
the analysis period, discounted at a rate of 10 percent. The year 1986
was used as the basis for comparison of NPV costs between options
because this was the first year in which any of the regulatory strate-
gies took effect. NPV costs were developed to compare the strategies
and take into account the opportunity costs associated with spending
money now or in the future. In this analysis, nationwide options (with
the exception of onboard) were assumed to begin taking effect in 1987,
all non-attainment area options were assumed to begin in 1986, and the
onboard option was assumed to begin in 1988.
These nationwide costs for the options were combined with the
cumulative and discounted emission reductions found in Table 5-7 to
calculate the cost effectiveness of each option. Cost effectiveness is
presented based on both the cumulative emission reduction and cost
values and the discounted (NPV) emission reduction and cost values.
Cost effectiveness is expressed as dollars spent per megagram of emissions
reduced ($/Mg). Table 7-21 presents the cost effectiveness values for
each option and for each pollutant (gasoline vapors, benzene, EDB,
EDO. Cost effectiveness values could not be calculated for the tank
truck options because, although costs for the tank truck controls were
calculated, the emission reductions due to these controls were included
in the bulk terminal and bulk plant emission reduction calculations.
Since the EDB emission reductions were the lowest, the cost effectiveness
associated with these emission reductions are the highest. The most
cost effective of the nationwide control options appears to be the
option requiring controls on bulk plants (while exempting bulk plants
with a throughput of less than 15,100 liters/day). The least cost
effective option appears to be the requirement for Stage II equipment
nationwide on all service stations, regardless of size (no exemptions).
7.4 NATIONWIDE COSTS OF REGULATORY STRATEGIES
The nationwide costs associated with the regulatory strategies
discussed in Section 4.3 are a combination of the control options costs
7-37
-------
TABLE 7-20. NATIONWIDE COSTS OF GASOLINE MARKETING CONTROL OPTIONS
Costs (Millions of 1982 Dollars) (1986-2020)
Facility Type
Terminals3
- Loading Racks
- Storage Tanks
Bulk Plants3
- Ho Exenptions
- Size Exemptions
Capital
Cumulative
598
33
653
462
1986 MPV
221
16
252
177
Net
.Cumulative
789
(247)
1,255
781
Annual i zed
1986 NPV
214
(65)
337
208
Tank Trucks
- For-H1re Terminal Trucks 57
- For-HIre Bulk Plant Trucks
o Ho exemptions 161
o Size Exemptions 93
Service Stations (Stage I)
- Nationwide
o Ho exemptions 1,008
o Size Exemptions 423
- All HA Areas
o No exemptions 35
o Size Exemptions 15
Vehicle Refueling
- Stage II
o Nationwide
23
63
36
378
159
14
6
361
1,014
585
2,234
938
81
34
96
270
156
590
248
24
10
- No exemptions
- Size Exemptions
o All NA Areas
- No exemptions
- Size Exemptions
o Selected HA Areas
- No exemptions
- Size Exemptions
o Onboard
7,823
2,977
2,350
895
893
340
6,836
2,847
1,086
977
373
371
142
1,787
18,710
6,323
5,841
1,973
2,220
750
9,666
4,869
1,628
1,672
558
636
212
1,921
Includes the costs associated with trucks owned by the facilities.
7-33
-------
CO
o
£
o
h-
o
o
ts
p
UJ
**
to
I
fiu
vo
en
••*
w
V
•i
^
o
0)
a.
£
U
10
u.
o
g
8
r-.
f-.
88 88 gg
o o «a: o o o o
|g ,
p-Tin"
o o o o
o o o o
o ou «a- vo
rt t-T i»Tfn"
~*
§00 o o o o
O O O 0 0 O
o oo << oo oo
o"
s
IT}
—1
i
1
CO
1
o"
8
in
g
CO
CO
vo
0
s
o
o
I
in
in e
c cZ
1 £
oi a
1— CO
00
O 0 1
in o
— i rt
8g
°-°. *
§"o" i
m°
cn w
f*«- in
00
0 0
o"o"
88. '
§s
gg
cocn^ i «s
S5
00
8S , <
Kg
CM cn i
38 '•".
<
Q
C
i
O)
=? «
O IS
t! i «
O) ^ IB ^/
X 0. 1— U
UJ E 3
01 01 U
OX 01 1—
Z UJ IO
II 0 *C
i
f> ^ VO P^
•^
o o o o
00 00
H it
— < CM ,*en
mia vot~
CM -H i-4
§o o o
o oS
ceTin" ^ToT
cn vo com
§0 00
o o o
w 10 fn
SCO VO CM
^ IO COW
SI? gS
mm ^- CM
o o . o o
vo 5? SS
*•*
a,
cn
t/> *rt M
— C C
o o cn
« •»• M VI ^ (A C
c ** c: to *» c -t-
o v a. o o a. o •—
I* "° 1^ u E"^ **
. C <1> E z a; a:
01 4J ° K _ ° X
u * m z ^ »2
•^ z oo< oo u
5. ' ' i
trt ^
§0 00 00
o o o o o
oo" oo" o"o"
o o o o ooo
. vo^ r— r- Km
m f*. vo CM co m
PO ~* CM —4 CM -*
00 00 00.
c? o o o o o
o o o o o o
o o o o ooo
m vo ^- p«» *H co
i-» m ' in CM vo CM
o o o o o o
00 00 00
o^o^ o o o o
§o o o o o
o o o o o
m co CM cn m to
vo P-. P-. c\i cn m-
f» •"* CM t^ CM ^H
go o o o o
o o o o o
o o o o o o
80 o o o* o* u
o .00 ooo
SCM W O CO W
F- CO CO CM Cn
o o oo ooo
o o o o ooo
o^o^ o o^ ooo
«r co cTco" m"in" vo"
CM to cnm — i«r o
m— CM— m— CM
o o oo ooo
o o oo ooo
o^o^ °°, ooo
in o io"o cTco" m"
m— m lf> *^ in —4
00 00 OO 0
*2X S!S j-io 10
^H cp cn Ol O Cfi M
« » * * »
gg gg Sg g
CM^*-^ °.Ch CM O ^
CM -4 CM CM ^ -H*
M
03
a*
u
c vi e c
^o e o «c o
•o X»c: £1 *»= z ^c
*-a.o < 0.0 -^ cxo
X E=<>~ E^~ VE*^
i— » C +•» «C O>4^ 4J C>4-*
"-"^o xo.z x a. u xo.1^
g^^ Of— § f— " § «
*^ c
W? O II O 1 ! 6 1 1 O
t 1
0)
c
t"-
o
tA
s.
•o
01
!l
i— O>
"« *«!
h- 3
VI*
at
a> «n
ix> O
(J
c a)
•^ >
v»,g
s*
u *»
4-1 ai
ai <*-
C ib>
•9
C O
*»- c
2E
(A
2^
O ^
^- o
Q. in
O 0)
e *o -P-
a> u .j=
o ** ai
u. c >
2S =
4-> S 0
« CJ
O) "C3
•O JB Ol
e x 13
3 0 *»
o JZ in
in •*•
•«> in
•o in O)
Ol &-
>resent value,
>st-effectlvei
ird controls i
u o
zzo
v .a u
7-39
-------
that were just discussed in Section 7.3. In many cases, the nationwide
costs for the strategies could be determined by simply adding the
nationwide costs of the appropriate options. This is true for all
cases except those which combine Stage II and onboard controls.
As was discussed in Section 5.1.3.2, when Stage II controls -and
onboard controls coincide at a vehicle refueling operation, the onboard
controls will dominate. This means that no vapors are returned to the
service station which results in no recovery credits and higher annualized
costs associated with the service station vapor control equipment
(Stage II equipment). The strategies also assumed that Stage II equipment
would be installed for only one useful life of 15 years (two useful
lives of 8 years each for vacuum assist systems). At that time, Stage
II would not be replaced since onboard equipment, alone would result in
greater emission reduction. This reduces the total capital cost burden
of the regulatory strategies. Therefore, the analyses of all -Stage II
system options, when combined with onboard in a regulatory strategy,
were calculated separately. The costs for Stage II options, when
combined with onboard, are shown in Table 7-22. Both the cumulative
and NPV of the annual and capital costs for these options are much
lower than the costs shown in Table 7-20 because the analysis incorporates
only one useful life of the equipment (two useful lives for vacuum
assist.)
Table 7-23 illustrates the estimated cumulative and NPV nationwide
costs of the regulatory strategies. Generally, as the regulatory
strategies get more complex, the costs increase. The lowest cost
regulatory strategies are those which would require Stage II in non-
attainment areas only and the highest cost strategies are those which
require the combination of Stage I, Stage II, and onboard controls
nationwide.
Table 7-24 presents the cost effectiveness of the gasoline market-
ing regulatory strategies. The cost-effectivess values were generated
from the nationwide costs in Table 7-23 and the nationwide emission
reductions presented in Table 5-9. In terms of VOC, the most cost-effective
strategy was Strategy IV - Stage I controls. The least cost effective
strategy was the no exemptions option of Strategy XII - Stage II and
Onboard controls nationwide.
7-40
-------
TABLE 7-22. COSTS FOR STAGE II OPTIONS WHEN COMBINED WITH ONBOARD
Costs (Minions of 1982 Dollars) (1986-2020)
Stage II Options
Capi tal
Net Annualized
Cumulative
1986 NPV Cumulative
1986 NPV
Nationwide
- Size Exemptions
- No exemptions
All NA Areas
- Size Exemptions
- No exemptions
Selected NA Areas
- Size Exemptions
y*K :.
- No exemptions
1,022
2,695
307
808
hiW
-•»' '*'"•"
307
.- •- ;• ' ••
842 vf; . 3, 589 --;, ;' = ";;" 1,575
2,207 9,521 : ^ ;-r:4,261
'"'" 289 ''"';/'" 1,040 T;: r"" -> 518
-;::^': 757 :;,: '. , • 2;803;7i'l:;!;vi:u; -J;l,429
f •::;,:, i.-;-i...5 yi*(«?hii ,r-y
;,;.-; 110 ;:-^c i395;«^ ^;7^ 197
ix'J.S ' 4i*i,'' - ' 9fWrff««iiKS: on . -fi
288 1,Q65.; , i: lgs?£ ,-543
7-41
-------
TABLE 7-23. NATIONWIDE COSTS OF GASOLINE MARKETING REGULATORY STRATEGIES
Costs (Millions of 1982 Dollars) (1986-2020)
Regulatory
Strategy
I. Baseline
II. Stage II - M.A.*
A. size exemptions
B. no exemptions
HI. Stage II - M.A.
A. size exemptions
B. no exemptions
IV. Stage I
A. size exemptions
B. no exemptions
V. Stage II
A. size exemptions
B. no exemptions
VI. Stage I & Stage II -
A. size exemptions
B. no exemptions
VII. Onboard
VIII. Stage II - H.A.* 4
Onboard
A. size exemptions
B. no exemptions
IX. Stage II - N.A. &
Onboard
A. size exemptions
B. no exemptions
X. Stage I 4 Onboard
A. size exemptions
B. no exemptions
XI. Stage II - N.A. &
Stage I & Onboard
A. size exemptions
B. no exemptions
XII. Stage II & Onboard
A. size exemptions
B. no exemptions
XIII. Stage I & Stage II &
Onboard
A. with exemptions
B. no exemptions
XIV. Gas Bz Reduction
A. 62A% Bz reduction
B. 81.3% Bz reduction
Capital
Cumulative
340
890
910
2,390
1,670
2,510
2,980
7,820
4,640
10,300
6,840
6,953
7,140
7,160
7,680
8,500
9,350
8,820
10,200
7,860
9,530
9,530
12,000
3,200
8,500
Net Annual i zed
1986 MPV
140
370
380
990
630
950
1,090
2,850
1,720
3,800
1,790
1,900
2,080
2,080
2,560
2,420
2,740
2,720
3,510
2,630
3,990
3,260
4,950
1,800
4,900
Cumulative
750
2,220
2,010
5,900
3,210
5,410
6,320
18,700
9,530
24,100
9,670
10,010
10,700
10,700 •
12,600
12,900
15,100
13,900
18,000
13,300
19,200
16,500
24,600
30,100-33,300
75,000-87,400
1986 NPV
210
640
s
570
1,700
860
1,440
1,630
4,870 .
2,, 490
6,310
1,920
2,120
2,460
2,450
3,370
2,780
3%360
3,310
4,820
3,500
. 6,180
4,350
7,630
7,390- 8,050
18,600-21,700
7-42
-------
a.
SC
1
^•4
CB S
t—t
l__ _
u3 |
OS
S
UJ 1 1
_J *
O ==
OO 10
S 2
OS
£ 3
oo
UJ 3=
<— 1 g
CD
UJ
^™
OS
t—
CL.
0 | I
< 01 01
i «rt C
__; eu 01
§||
UJ - BO
-QS S
a
U- t z
OL*J 5
U
//* M
to S
UJ
UJ
S 1
Co* CO
UJ 2
u_
U.
LU
1— g
00
o
o
3
•ef
CVJ
I
UJ
CO
£>>
IO 4->
o»**
cc
• §|
o"o
ss
en"eo"
OH CM
§O
0
§o
coS
co"*-*1
CM tO
o o
88
ii
to fo
ro cn
«•* CM
§O
8
||
•o- eo
en CM
CM to
o o
8.8.
(A in
^f ••«
ii
co"o"
in *r
~* p*>
Is
CM
O O
V CO
0 C^
~* CM
IA
C
*.-2S
«c ** o
• Q.^-
z e 4-*
Q> Q.
1 Si
*^ O
0) OKA e
«n to
«c *» • »
» to«teo
-H M
ii
88
~4 CO
CM m
•^ CM
II
o o
o o
m m
to in
CM in
o o
8.8.
oo
80
0
•, *
^f PO
CM tO
•-* CM
ii
||
eo T
to to
CM in
oo
33
Ss"
II*
^•It)
*f O
§8
CO C0_
^
o o
CM
(A
C
0 V)
s.S-5
QJ Q.
1 « i
1-4 X
»•* Cl CU
a> *^ o
cn « c
CO *t CO
t_^
O O O O
88 88
11 l§
r-«. O r*- to
-H ~H PO
|| 8|
§"o o"o"
o o o
v \n in co
LO -H (O a. d> ex
x E x E
OJ QJ CJ tt»
X ^H X
i—iwo; •»• o> 4)
N N
O> •*- O 4) ••• O
cn in c cn «A e
IO (0
CO «C CO Crt -S CO
" ^
o o •
0 O
§o
o o
CM f
-H CM
o o
gg
II °
in r^.
CM ^
o o
33
8*8
°*°» °
m to
*^ CM
00
o o
°.p.
o"o"
38 o
* •>
in eo
CM V
o o o
8.8. 3.
cn en to
o o o
—•CM CM
o o o
8.8. 8.
in"w< f*
*H(M CM
33 g
to ro^ co^
*-T -7
CM CO O
f^« PO ^
•^ ^
1
i— i C
o «
cn*^ o
« a.-*-
sl§-
x
i-» CU O* T3
cn «n c o
4j . . e:
CO «CCO 0
1-4 «-H
88
P.P,
88
<— i ^.t
CO in
O O
0 0
p. p.
o o
§o
.°
oeo
in en
we-
oo
8.8.
II
o o
CO U)
-Z-*
38
00
gg"
o o
§§
to in
i|
CM CM
O O
8.8.
to" en"
^4 CM
CM CM
go
en in
— <"— T
00
CM O
OH «-<
•0 VI
S
*. ££
•c *> o
=E E *»
HI CL
1 g g
1-1 X
1-1 T3 O) O>
L* K
eu «•••: o
cn o in c
-4-> C • •
oo o «C aa
••4
*
0 O
o o
oo
0 0
r— CM
cn co
to to
o o
0 O
o o
O CD
P.O.
coo"
CM cn
CM OH
o o
§0
.p.
oo
33
Oto
SlO
oo
38
II
o o
CMW
CM OH
2°
8.8.
CM CM
oo
8.8.
co'eo"
CM CM
o o
cn eo
enr-
OH'OH
5S
oHoH
1A
•e e
O 1A
X E 5
tu eii
— X
O C . .
) O < CO
X
o o
83
88
^ VJ>
CM CM
§O
O
-°«
oo
§0
C7*
-Hen"
to tn
§0
0
p. p.
§0
0
en en
in in
CM CM
o o
§0
.p.
o"o
83
• •
en CM
tow
o o
8.8.
CM en
o o
§0
„=>„
CM O
oHoH
SO
in
0 0
SS
«A
-o e
L. O (A
0 4J O
C E 4J
•0 Q) CU
X
Qt **"• O
cn «A c
#e
4J * •
CO < CO
X
0 O
8.8.
oo
ss.
CM UD
CM CM
88
P. P.
oo
o o
r— in
•a- to"
tn in
33
P.P,
o"o*
83
Ch in
CM VO
CM CM
00
O 0
00.
o"o
CD CD
00
• *
cn I*-
in r-*
in in
§o
o
55
o o
8.8.
P*TO
•*• f^
§0
o
^
o o
eSS
^^
(A
*«3 -0 =
t- 0 (A
I X E
•e eu o>
i— i X
»•* »-^ cu
N
0} O> •*- O
cn cn iA c
co to -a; ca
*-*
x
||
o o
eo o*
in r-~
o o
0 O
0 O
88
CMO"
in in
o4 oH
00
33
s"§
p. p.
-4 en
OCM
tor~
O O
8.8.
oo
o o
P.P,
oo"
»r en
m in
OH OH
O O
eo en
in e*
CM en
o o
go
.0.
en in
in w
CM en
§0
CM
r-*> to
OH CM
o o
to to
tO CM
OH CM
b. C
a o vi
O -o- c
OJ 0.
•e x E
U V
NH X
N
W-o- o
en IA e
«
V> < CO
^
X
o o
o o
0.0.
80
0
in or
CM en
oo
og_.
oo
00
in «r"
in to
§0
0
o o
g"g"
°.°.
eh" en
in oj>
CM en
oo
00
o"o"
88
M •
to to
in to
II
r-Teo
ss
II
CO CM
tO CM
^CM
gg
O in^
•-T^T
o o
to cn
— t — *
( t
g, V,
a c
+* O IA
CO -o- C
0
^ S *J
OH l- a> a.
•a x E
o a> «
cn'c — 0
"-H < CO
X
3=5=
=!-!
«<:
3E SC
«
11
°»s.
o o
CO CO
§o
o
in'o* •
r*. PO M
cu
c
CU
^
U
OJ
u-
0)
2 z **
o
u
o
|Q
=
15
0
e
4J
gg ^
1
0)
cn
a
e c t.
0 0
*A ^- «— --
« ** 4-» (O
Cfl U U ^™
= || |
CO M N O
CO CO
u »4 »« c
=j ^-n •»-
•o • • o
o> CM *-• ^
Q£ 10 CD ^
> «C CO
1-4 «
x
7-43
-------
7.5 COST PER INCIDENCE REDUCTION
An analysis was performed to determine the residual costs expended
per cancer incidence avoided for selected nationwide and nonattainment
area regulatory strategies. The residual costs were determined by
obtaining the annualized costs of the controls associated with the
regulatory strategy and subtracting several assumed benefit values
per megagram of VOC emissions reduced. The annual VOC benefits are
those in addition to cancer prevention, such as non-cancer health effects
and agricultural damage due to ozone. The residual cost per incidence
is then simply the residual costs divided by the appropriate amount of
cancer incidences avoided by implementation of the regulatory strategy.
In Table 7-25, several of the regulatory strategies are presented
with their corresponding emission reductions. The emission reductions
are presented as the net present value of all the annual emission
reductions over the study period and as a reannualized value representing
equal emission reduction for each year of the study period. Calculations
were then conducted to determine the residual costs after assuming
several different dollar values for the benefit of reducing each megagram
of VOC emissions. For example, in Table 7-25 the reannualized emission ,
reduction associated with Stage I is 0.218 million Mg. Multiplying
this emission reduction by each of the assumed VOC benefit values
yields the annual ized VOC benefit in dollars.
Table 7-26 presents annualized cost (including control equipment
costs and enforcement costs) and annualized incidence reduction due to
benzene exposure associated with several of the regulatory strategies.
The cost per cancer incidence avoided, assuming no additional benefits,
is calculated by dividing the annualized costs by the annualized incidence
reduction. Table 7-27 takes this one step further by incorporating the
annualized VOC benefits into the analysis. The values presented represent
the residual cost, assuming varying benefits for reducing VOC emissions,
of reducing cancer incidences due to benzene exposure.
Table 7-28 contains a similar analysis compared to that used in
Table 7-27, except that Table 7-28 was developed using the sum of
the incidences due to benzene and gasoline vapors. It is assumed that
7-44
-------
OO
I—
1— 1
u.
UJ
•sy
UJ
03
O
0
O
^»
^
00
(—
0
o
of
0
H-
«j
S-
o
4-J
fl3
^—
3
co
cu
C£
• •
M-
O
CU
3
1o
^
•f-3
•r"
M-
CU
C
CD
CQ
O
0
c
0
•(••
CO
• VJ
•f—
UJ
M-»
cu
3
fw>
rt3
>•
C
O
•r-
4^
O
•a
s
to
c
o
•r"
to
to
•r-
3
JS
UJ
cn
2T
•^"^^
o
o
0
CM
CD
^r
^*^
O
o
uo
f"H
CD
^
O
o
o
t-H
CD
s:
0
o
uo
CO
3?"
^%^
o
uo
CM
^^
CD
s:
c
o
2-
^
S
_
en
^g*
O
•p.
!•«•
l_
•p—
S
>*
U
CU
o
<4«
t^
00
in
!**«,
to
LO
r*.
oo
cn
cn
^^
CO
o
o
10
co
O
•— ••
s
LO
^_^
<
•z.
1
1 — 1
I — 1
cu
co
(O
^_)
oo
LO
cn
CM
r-H
CM
CM
p^.
^»
t-H
j*
f^
CO
p^,.
^^
t-H
O
CM
sr
»-H
^
CO
c
o
•r-
to
ry?
1
t— i
i— i
cu
CO
(C
] *
oo
, i
cn
t-H
OO
•=!•
t-H
LO
cn
CO
^>
CM
LO
cn
o
0
cX
0
to
LO
•~—
£.
o
•r-
•»->
its
^»
1
>— 1
l_4
cu
cn
to
4_>
OO
^— .
&Q
CM
CT)
^ -
^3
S.
(O
o
c
o
^" CT^
1 — .— <
CM LO
LO cn
O CO
CM 00
p^- cn
OO LO
T-H CM
CO O
to co
t-H
«* uo
co to
p>- cn
oo uo
t-H CM
o o
CM 0
OO LO
»-r CM
cu
> cu
'•*-> -f—
(O -r*9
^ (13
0 S_
0. O
fO CL
> 03
CU >
CU
0
3 3
7-45
-------
TABLE 7-26. BENZENE REGULATORY COSTS AND INCIDENCE REDUCED
Regulatory Strategy
(with size cutoffs)
(In-use-efficiency)
Stage I
Stage II-NA (87%)
Stage II-NA (56%)
Stage II-Nation (86%)
Stage II-Nation (56%)
Onboard (92%)
w/o evaporative
w/ evaporative
Annuali zed
Costs
($ Millions)3
91
62
52
183
146
199
199
Annual ized
Benzene
Incidence
Reduction'3
0.06
0.83
0.41
1.92
1.13
1.44
1.66
Costs
($ Mil lions
per Benzene Cancer
Incidence Avoided)
1,564
75
126
95
128
138
120
a Includes control equipment and annual enforcement costs.
blncidence reduction after controls. Before-control annualized incidence:
Stage I = 0.18, Vehicle Refueling =4.09.
7-46
-------
LU
O
LU
a
1—4
O
^z
H- *
r**
LU
O
*^ **"^
O 00
(—
LU U_
Cu LU
•z.
OO LU
1— CQ
oo
O O
0 0
^»
^
o 2:
l^" l_^
t ^"**
^C -^
_l 3
:D oo
CD OO
LU <
Qi —
LU Q
Z LU
LU O
M l-H
z o
LU >
CO eC
r^
CM
1
f^
LU
l
CQ
^£
1—
-a
CU
•o
o
>
^
CD
O ••
£Z ti—
to "O O
3 T- 3
s: o -a
c c a>
«c i— i a:
^•^
cn
O)
•p
n3
i-
+-*
oo
>ij
^
o _
fO
r^
3
cn
0)
C£.
cn
•^_
o
0
o
VJ-
cn
2:
O
0
LO
r-t
e ri
^r*
^v^
O
O
O
<— t
CD
^,
O
o
LO
4&"
cn
s:
*^**»
0
LO
C^I
•feO-
^— H
*J
O
£Z
O O
O <^- LT> CT» ^ LD O
CO t-4 ^ ^t-_
O O^ T-H P**» ^D CO OO
co co in co cr» ^t-
c^ «^^ *^^ v,o ^^ c^ co
CM LO O r— CD CM CO
^ CO t-l CM CO •* tO
O CO -tf- O1 ,-H <* to
• • « • • • •
O O O I— 1 i— 1 I— I r-)
s? &s
to to
OO LO cu
&5 &9 T- >
r-. to c c +J -r-
OO LO O O (O •»->
~ ^ ^ ^ (pv >p— ^— ^ t__ fQ
+-> 4J &S O S-
^ «^ (B (O CM CX O
•Z. -Z. Z. Z CTi n3 Q.
1 1 1 1 — >
1— ( 1—4 1 — 1 1 — 1 1 — t "O QJ
S- O
CLJ CU 4^ £3
OO OO OO OO OO O
in
CM
1
II 1 —
"O rt3
S-
03 q—
CU *~*
O CO
t— ' 1 s
CU CO
•a o
•r- 0
c -o
•i- cu
^ •r^
i *
c~
cn
• (-I*
i
-M
3
0
CO
>r_
<4_
01
O)
-Q
03
a
•a
c
CO
o
cu
M
7-47
-------
cet
O
UJ
a
O
«t
UJ
o
367
UJ
S
O
5e
UJ
0
o
C£
UJ
Qu
CO
f—
a
CO
os
ex.
ic£
5*
UJ
se
H-4
—1
O
CO
o
§
^
3E
UJ
J3E
UJ
CO
.
*!*
r^*
UJ
_J
«
*^c
h-
*-».
CO
C£.
CZ3
§E
>•
CO
S
Of
£
a:
0
&
— «
_J
a;
L^
a.
o.
UJ
ca
>™4
CO
S3
W2C
_J
a.
^w{
~""J
u
•o
o
o
.
C43
IO •
Q) CT
O.C
g§
•v* l/>
r- in
z
^^
in
4J
O
U
^^
Q) et
M U
1?
5-S
•g«
N U
||
11
>>
W
IQ
t.
s>
^*
o
"S
en
6>
O)
o"
o
Si
*^
§
g
in
•-i
en
^•4
O
^
1
CD
£
g
Si
I-" ^S
55
=1 N
•^ co
-O
*»3
0 f
"§S
^^
&
QJ
•r-
O
«^—
*
0)
in
C
H-l
M
a>
ai —
i o-o
1 **~ 3
mv>
in
Ql
« Tj
i- S
vl
in
a —
— 3
in
43
^l
g 43
V)
in
«n
in
O)
•E 3
in
in
w5
I- 3
tn
in
15
V)
in
at
i > "Q
IO 3
t- 43
in
o o o o o oo
o o o o o o o
o o o o o o o
o o o o o oo
O O XT <*> «3 » O
O O CO CM •» * O
o tn o co CM cMin
u
O co vo tn co co co
-H f*. ro o m men
fO *^ *^ *"*^ **^
en tn co i«. en O vo
CO CO r» VO CM CM O
— < XT CM C71 CO COCO
^4 VO CO CO CO O ^
O P«- CM f»- X*1 CM"*
en o »— < r** co ch ro
-H o ID -H evj vo co
^4 C\J -H •-< CM
»« »7
VO 'C*
co in 4) a>
r- S c: c 43 43
co vn o o " £ £
«C «c >
_i i— , _( i— i ai o>
1- O
0) O) 0) O> O) 1Q -^-^
o en en O) CTI o ^ f
arsiaaajs
CO CO > fcO I/I O
7-48
in
c
CO
u
o
^
01
^J
f—
IO
i
c
1
to
•
CJ
i
a i—
•e s
0) O
"o —
G in
0) >•
o o
S °
•o -o
•^ o>
C3 N
C •?
c
c %.
a
u
t.
Ol
in
O
«°
w
u
3
Ol
ct
O)
u
01
•o
u
c
^
|g
^
•
C?
•t-
N
CQ
Q
-
is
1
01
43
^
en
0)
3
g
if
£•
Ol
Ol
43
^J
10
5
M
01
U
^K
c
"^
o
i.
VV
u
-------
the incidences due to benzene exposure and the incidences due to gasoline
vapor exposure are additive since the respective exposure results in
different types of cancer incidences (leukemia in the case of benzene
exposure and liver or kidney tumors in the case of gasoline vapor
exposure).
7-49
-------
7.6 REFERENCES
1. Bulk Gasoline Terminals - Background Information for Promulgated
Standards. U.S. Environmental Protection Agency. Office of Air
Quality Planning and Standards. Research Triangle Park, NC. Pub-
lication No. EPA-450/3-80-038b. August 1983. p. .B-3 through B-5.
2. Reference 1, p. 2-42.
3. VOC Emissions From Volatile Organic Liquid Storage Tanks - Background
Information for Proposed Standards (Preliminary Draft). U.S.
Environmental Protection Agency. Research Triangle Park, N.C.
EPA-450/3-81-003a. June 1983. p. 8-2 through 8-7.
4. Control of Volatile Organic Compound Emissions from Volatile
Organic Liquid Storage in Floating and Fixed Roof Tanks - Guideline
Series. U.S. Environmental Protection Agency. Research Triangle
Park, N.C. (Draft) August 1983. p. 5-1 through 5-6.
5. Arthur D. Little, Incorporated. The Economic Impact of Vapor
Control Regulations on the Bulk Storage Industry. Report to U.S.
Environmental Protection Agency. Research Triangle Park,- N.C.
Publication No. EPA-450/5-80-001. June 1979. p. II-6.
6. Pacific Environmental Services, Inc. Study of Gasoline Vapor
Emission Controls at Small Bulk Plants. Report to U.S. Environ-
mental Protection Agency, Region VIII. EPA Contract No.
68-01-3156, Task No. 5. October 1976. p. 6-1 through 6-10.
7. Control of Volatile Organic Emissions from Bulk Gasoline Plants -
Guideline Series. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. December 1977. p. 4-1 through 4-11.
8. Pacific Environmental Services, Inc. Evaluation of Top Loading
Vapor Balance Systems for Small Bulk Plants. U.S. Environmental
Protection Agency, Washington, D.C. EPA Contract No. 68-01-4140,
Task Order No. 9. June 1977. p. V-2, Table V-l.
9. Pacific Environmental Services, Inc. Stage I Vapor Recovery and
Small Bulk Plants in Washington, D.C., Baltimore, Maryland, and
Houston/Galveston, Texas. U.S. Environmental Protection Agency,
Washington, D.C. April 1977. p. II-2, Table II-l.
10. Reference 5, p. 111-19.
11. Reference 6, p. 3-4.
12. Reference 1, Appendix B.
13. Arthur D. Little, Inc. The Economic Impact of Vapor Recovery
Regulations on the Service Station Industry. Report to U.S.
Environmental Protection Agency. Research Triangle Park, N.C.
Publication No. EPA-450/3-78-029. July 1978. p. H-8.
7-50
-------
14. Pacific Environmental Services, Inc. Hydrocarbon Control
Strategies for Gasoline Marketing Operations. Report to U.S.
Environmental Protection Agency. Research Triangle Park, N.C.
Contract No. 68-02-2606, Task No. 13. April 1978. p. 6-4.
15. Luken, Ralph A. Cost and Cost-Effective Study of Onboard and
Stage II Vapor Recovery Systems. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. August 1978. 39 p.
16. Reference 12, p. 6-9.
17. Reference 11, p. H-6.
18. Economic Indicators: CE plant cost index. Chemical Engineering.
Vol. 90, No. 6. March 21, 1983. p. 7.
19. Telecon. Norton, Robert, Pacific Environmental Services, Inc.,
with Simeroth, Dean, California Air Resources Board.
August 23, 1983.
20. Memorandum from Norton, R.L., Pacific Environmental Services, Inc.
to Shedd, S.A., Environmental Protection Agency. Decembe-r 20, 1983.
Trip report to California Air Resources Board.
21. American Petroleum Institute. Cost Comparison for Stage II
and Onboard Control of Refueling Emissions. API Publication
No. 4306. Washington, D.C. January 1984. Appendix V.
7-51
-------
-------
8.0 ECONOMIC IMPACT OF THE REGULATORY STRATEGIES
This chapter examines some of the economic impacts associated
with regulatory strategies that apply nationwide and, in addition,
Regulatory Strategies VIII and IX. (See Section 4.3 for a description
of the regulatory strategies.) This chapter presents a sensitivity
analysis of the total cost of each of these regulatory strategies,
where total cost is defined as the 1986 net present value (NPV) of
both control and enforcement costs. Price increases are then estimated
for each of the nationwide strategies. The price increase per liter
of gasoline is based on average annualized control cost and average
annual gasoline consumption. Price increases for light duty vehicles
(LDV's) and light duty trucks (LDT's) are assumed to equal estimated
unit cost increases. Given these price increases, resulting declines
in quantity of gasoline, LDV's, and LDT's consumed are estimated. The
chapter concludes with an examination of some of the impacts of the
regulatory strategies on facilities of different sizes and regions. A
report to EPA contained in the docket provides a more detailed discus-
sion of the methodology, analyses, and results discussed in this
chapter.1
8.1 METHOD AND SCOPE
The total cost of a regulatory strategy is the principal measure
of economic impact presented in this chapter. Total cost is the 1986
NPV of the control cost incurred by firms and the enforcement cost
incurred by government agencies. Total cost is a "net" cost in the
sense that "recovery credits" from reduced vaporization of valuable
gasoline stocks are subtracted. Other benefits of the regulatory
8-1
-------
strategies, such as improved health and reduced crop damage, are not
considered here. Consequently, there is no attempt in this chapter to
identify a regulatory strategy that balances marginal total cost with
marginal air quality benefits in monetary terms. The total cost of
each of the 10 regulatory strategies considered is computed using the
control cost assumptions of Chapter 7 and the annual inspection
assumptions of Chapter 9. This provides base case total cost
estimates, which are then broken out by industry sector.
The sensitivity of total cost projections to the underlying
assumptions employed in Chapters 7 and 9 is illustrated by changing
key variables and computing revised total cost for comparison. These
variations cover the following: constant gasoline consumption at the
projected 1986 level, declining gasoline consumption incorporated
through declining numbers of facilities with constant throughput as
opposed to constant numbers of facilities with declining throughput
(base case), 5 percent and 15 percent real rates of discount, and a
unit cost of onboard control of $25 per vehicle tank (20 percent of
LDT's have dual tanks). In addition, the effect of varying levels of
enforcement effort and compliance on the total cost of nationwide
Stage II strategies is examined.
In a perfectly competitive market, the control costs required of
firms will be passed on to their customers in the form of higher
prices. In this analysis, it is assumed that the gasoline marketing
industry is competitive and that control cost will be passed along •
through the stages of marketing to the purchasers of gasoline. The
gasoline price increases are estimated by computing a "unit cost
increase." The unit cost increase of gasoline is the control cost for
the entire gasoline marketing portion of a nationwide regulatory
strategy, annualized over the period of analysis (1986-2020) and
divided by projected average annual gasoline consumption for the
period. Gasoline price impacts are not assessed for regulatory
strategies that affect only nonattainment areas because such
assessments would require consideration of local market conditions.
8-2
-------
Average unit cost increases for LDV's and LDT's are inputs into the
analysis. The unit cost increases are used to estimate reductions in
annual consumption likely to occur in response to price increases
equal to the unit cost increases. These reductions are termed "quantity
impacts" and are based on a range of pure elasticity values published
in the professional literature.
The last section of this chapter identifies some of the distribu-
tive economic impacts of Stage I, Stage II, and benzene removal control
options. In particular, the differential control costs per unit of
output for firms of different size and for different regions of the
country are considered.
8.2 SUMMARY OF COST COMPARISONS . • ..
8.2.1 Total Cost by Regulatory Strategy
The NPV of control cost, enforcement cost, and total cost for
each nationwide strategy are given in Table 8-1. The base case assump-
tions used in developing the economic impacts in this chapter are the
same as those used in Chapter 7, including the assumption that decreases
in gasoline consumption over time would result in declining throughput
at a constant number of facilities. Declining facility throughput
results in declining recovery credits over time, which materially
affects both the magnitude and distribution of aggregate and sectoral
costs. Accordingly, an alternative method of incorporating projected
declines in gasoline consumption is considered in Section 8.2.3.
The numbers in Table 8-1 may be interpreted as the amount of
national income that must be committed to control equipment and enforce-
ment to support a 35-year program of reduction of benzene emissions
from gasoline. The total cost values in Table 8-1 range from $0.88
billion to $21.7 billion. Selecting regulatory strategies with exemp-
tions decreases total cost--in most cases by one-half to two-thirds.
As previously stated, the total cost of a regulatory strategy
must include enforcement cost. This cost represents real resources
used even though the cost is not borne by the firms or industry sectors
impacted. Based on the Chapter 9 assumptions for annual inspections,
enforcement cost comprises 0.5 to 4.3 percent of total cost for
8-3
-------
TABLE 8-1. 1986 NPV OF THE COSTS OF THE REGULATORY STRATEGIES
(109 1982 dollars)
Regulatory strategies Option
IV.
V.
VI.
VII.
VIII.
IX.
X.
XII.
XIII.
XIV.
Stage I — nationwide
Stage II — nationwide
Stage I and Stage II —
nationwide
Onboard — nati onwi de
Stage II — selected nonattain-
tnent areas and onboard —
nationwide
Stage II — all nonattain-
inent areas and onboard —
nationwide
Stage I and onboard —
nationwide
Stage II and onboard —
nationwide
Stage I, Stage II, and
onboard — nationwide
Benzene reduction in
gasoline
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Control cost
0.86
1.44
1.63
4.87
2.49
6.31
1.92
2.12
2.46
2.45
3.37
2.78
3.36
3.50
6.18
4.35
7.63
7.39-8.05';!
18. 6-21. 7°
Enforcement
cost
based on
annual
inspections
0.02
0.04
0.07
0.22
0.09
0.27
0.001
0.01
0.02
0.02
0.06
0.02
0.04
0.06
0.18
0.08
0.22
—
Total
cost
0.88
1.48
1.70
5.09
2.58
6.58
1.92
2.13
2.48
2.47
3.43
2.80
3.40
3.56
6.36
4.43
7.85
7.39-8.05'j
18.6-21.7
aEsti(«ates assume declining gasoline consumption, declining recovery credits, and a constant
number of facilities.
For Strategies IV through XIII, Option A exempts certain facilities from the regulatory
strategies and Option B allows no exemptions. For Strategy XIV, Option A requires the removal
of 94.5 percent benzene from the reformate fraction for a total reduction of 62.4 percent;
Option B requires the removal of 94.5 percent of benzene from reformate and fluid catalytic
cracker fractions for a total reduction of 81.3 percent.
The total cost is control cost plus enforcement cost.
These control costs include expenditures required to boost gasoline octane and volume after
benzene reduction.
8-4
-------
regulatory strategies involving Stage I and Stage II controls. Enforce-
ment cost is only 0.07 percent of total cost for onboard controls. No
estimate of enforcement cost is made for benzene reduction in gasoline.
8.2.2 Total Cost by Sector
The following paragraphs briefly describe base case cost impacts
by sector. The estimates presented assume projected declines in
gasoline consumption, declining recovery credits, and a 10-percent
real discount rate. The sectoral impacts of alternative assumptions
are discussed in Section 8.2.3. (For a more detailed discussion of
sectoral impacts, see Reference 1.)
Regulatory Strategies XIV.A and XIV.B would impact petroleum
refiners most directly. Strategy XIV.A would remove 94.5 percent of
benzene from the reformate fraction for a total reduction of 62.4
percent, and Strategy XIV.B would remove 94.5 percent of benzene from
both the reformate and the fluid catalytic cracker (FCC) fractions for
a total reduction of 81.3 percent (see Table 8-2). Table 8-3 gives
the 1986 NPV of the control costs of benzene reduction required of
gasoline producers.
Also included in Table 8-3 are estimates of two additional costs
of benzene removal. The cost of octane loss reflects the cost of
gasoline additives needed to maintain gasoline octane levels at roughly
pre-benzene-reduction levels. The cost of volume loss is the cost of
additional gasoline production needed to make up for the volume of
gasoline lost due to benzene removal.
Table 8-4 presents the 1986 NPV of Stage I control and enforcement
costs under base case conditions for bulk terminals, bulk plants, and
for-hire truck operators. Both bulk terminal and bulk plant NPV's
already include the cost to equip trucks owned at these facilities;
therefore, for-hire truck cost is treated separately. Bulk terminal
cost also includes cost to control fixed-roof storage tanks.
For the bulk terminal sector, the NPV of Stage I cost, including
enforcement, is $246 million. Enforcement cost represents approximately
0.4 percent of total cost. Storage tank control cost is negative
while for-hire truck control cost is positive and substantial.
8-5
-------
TABLE 8-2. ESTIMATED REDUCTION IN BENZENE CONTENT OF GASOLINE
RESULTING FROM REGULATORY STRATEGY XIV
Option
A
B
Estimated
volume of benzene
before reduction
(percent)
1.37
1.37
Estimated
volume of benzene
after reduction
(percent)
0.52
0.26
Reduction
in benzene
content of gasoline
(percent)
62
81
a
Source: Reference 2.
TABLE 8-3. 1986 NPV OF THE COSTS OF BENZENE REDUCTION
(109 1982 dollars)
Costs
Control costs required
of gasoline producers
Investment requirements over
analysis period
Operating costs
Cost of octane loss
Cost of volume loss
Total cost
Option A
5.77
1.80a
3.97
0.28-0.55
1.34-1.73
7.39-8.05
Option B
14.27
4.77a
9.50
2.57-5.
1.75-2.
18.59-21
15 .
26
.68
Investment costs include onsite and offsite construction, equipment
and materials cost, and working capital for a 3-year construction phase .
and a 20-year plant life.
Total costs for benzene reduction do not include an estimate of enforcement
costs but are the sum of control costs to gasoline producers and the costs
of octane loss and volume loss.
8-6
-------
TABLE 8-4. 1986 NPV OF STAGE I CONTROL AND ENFORCEMENT COSTS FOR
BULK TERMINALS, BULK PLANTS, AND FOR-HIRE TRUCKS
(106 1982 dollars)
Facility, option
Bulk terminals3
Bulk terminal costs
Bulk terminal storage tanks
Independent bulk terminal trucks
Total
Bulk Plants— Option A (exemptions)
Bulk plant costs3 -
For-hire bulk plant trucks
Total
Bulk Plants—Option B (no exemptions)
Bulk plant costs3
For-hire bulk plant trucks
Total
Control cost
213.0
-65.0
96.4
244.4
206.9
155.8
362.7
337.9
270.2
608.1
Enforcement
cost
0.8
0.3
—
1.1
6.3
—
6.3
6.3
—
6.3
Total
cost
213.8
-64.7
96.4
245.5
213:2
155.8
369.0
344. 2
270.2
614.4
Includes cost to equip trucks owned at the facility.
8-7
-------
For the bulk plant sector, costs vary by option. Under Option A
(exemptions), the total cost is $369 million. Enforcement cost contrib-
utes nearly 2 percent of this total while for-hire truck controls
contribute 42 percent. Under Option B (no exemptions), the total cost
is $614 million. Enforcement cost and cost to control for-hire trucks
represent 1 and 44 percent, respectively, of the bulk plant sector
total cost.
Table 8-5 presents the 1986 NPV of Stage I and Stage II control
and enforcement costs for service stations. Costs for the Stage II
program with onboard controls are also presented.
The nationwide Stage I total cost for service stations is $262.2
minion under Option A and $624.4 million under Option B. Enforcement
cost constitutes 5 percent of total cost for each option. When Stage I
is instituted in all nonattainment areas only, the total cost is $10.5
million for Option A (enforcement cost is 6 percent) and $25.1 million
under Option B (enforcement cost is 5 percent). (These costs are
relatively low because almost all nonattainment areas are assumed to
have Stage I controls at baseline.) The nationwide Stage II total
cost under Option A js $1.69 billion. Enforcement cost constitutes
4 percent of this total. - Under Option B, the total cost is $5.08 bil-
lion, of which 4 percent is enforcement cost.
Costs are calculated separately for options that require both
Stage II and onboard controls because in these cases Stage II equipment
is installed only for one lifetime for balance systems (^15 years) and
two lifetimes for vacuum assist (a total of 16 years) as onboard
equipment is phased in. For Stage II with onboard, decreasing recovery
credits are assigned to the stations as the onboard population increases
and the vapors are therefore no longer returned to the stations.
Thus, while the annual capital cost for Stage II is lowered when
Stage II is combined with onboard, operating cost increases over time
due to the loss of the recovery credits.
As a result, the total cost for Stage II with onboard differs
from that for Stage II alone. The 1986 NPV of control and enforcement
costs for nationwide Stage II with onboard presented in Table 8-5 is
$1.6 billion under Option A and $4.4 billion under Option B. Enforce-
8-8
-------
TABLE 8-5. 1986 NPV OF STAGE I AND STAGE II CONTROL AND
ENFORCEMENT COSTS FOR SERVICE STATIONS3
(106 1982 dollars)
Facility, option
Stage I — nationwide
Option A (exemptions) . -
Option B (no exemptions)
Stage I--all nonattainment areas
Option A (exemptions)
Option B (no exemptions)
Stage II — nationwide
Option A (exemptions)
Option B (no exemptions)
Stage II — nationwide with onboard
Option A (exemptions)
Option B (no exemptions)
Stage II — selected nonattainment areas
with onboard
Option A (exemptions)
Option B (no exemptions)
Stage II — all nonattainment areas
with onboard
Option A (exemptions)
Option B (no exemptions)
Costs pertain only to service stations
Control cost
248.0
590.6
9.9
23.8
1,622.2
4,859.9
1,565.9
4,259.8
195.8
543.3
515.0
1,429.1
and do not incl
Enforcement
cost
14.2
33.8
0.6
1.3
170.7
224.0
70.7
178.3
7.3
23.2
19.4
61.2
ude costs of
Total
cost
262.2
624.4
10,5
25.1
1,692.9
5,083.9
1,636.6
4,438.1
203.1
566.5
534. 4
1,490.3
onboard
controls.
8-9
-------
merit cost is 4 percent of each total. Note that Table 8-5 shows only
costs to service stations and excludes costs of onboard controls.
In addition to the nationwide regulatory strategies, two additional
strategies covering nonattainment areas only were considered. The
total cost for Stage II in selected nonattainment areas with onboard
is $203 million under Option A, with 4 percent of this total attributed
to enforcement cost. Under Option B, total cost is $567 million, of
which 4 percent is enforcement cost. Total cost for Stage II in all
nonattainment areas with onboard is $534 million under Option A and
$1.5 billion under Option B. Enforcement cost is 4 percent of total
cost for both options.
Although Stage II costs change when combined with onboard control,
onboard cost is the same whether or not it is combined with Stage II.
The analysis uses unit cost estimates of $13.32 per LDV and $21.83 per
LOT (based on a weighted average of control costs for single- and
dual-tank trucks.) Given these inputs, the 1986 NPV of onboard control
costs is $1,922 million. This figure includes an enforcement cost
estimate of $150 thousand per year, which has a 1986 NPV of $1.3 million
or 0.07 percent of total cost.
8.2.3 Sources of Variation in Cost
For several reasons, the total cost, unit cost, and quantity
impact of these regulatory strategies might deviate from the estimates
presented above and in Section 8.2.4. This section investigates the
sensitivity of total cost estimates to certain variations in base case
assumptions. The meaning of each variation is discussed below and the
impact of each variation on sectoral and total cost is presented.
Table 8-6 summarizes the results of this portion of the sensitivity
analysis. Base case results are included for purposes of comparison.
8.2.3.1 Constant Gasoline Consumption. The base case cost
estimates presented in Table 8-1 incorporate a projected decline in
annual gasoline consumption of 23.3 percent over the period of analysis.
(Gasoline projections are discussed in Section 4.2.1.) However, there
is always considerable uncertainty attached to projecting values over
a 35-year period. Cost estimates are sensitive to these projections.
Accordingly, Table 8-6 provides estimates of the total cost associated
8-10
-------
to
o
1— t
a.
yr
I/)
to
UJ
t—
•z.
oc
UJ
_J
C£.
UJ
a
z
rj
CO
UJ
s
UJ
1-
cc co
,_(
o: c
O i-
§ .2
UJ r-
ce o
•o
UJ
X CM
1— CO
cn
U. rH
0
01
r— O
CO rH
O *•>
_l
r-
g
UJ
X
u_
o
a.
z
CD
CO
cn
CD
1
CO
UJ
CO
1—
p
IA
o
u
•o
to
o
o
0)
c
E
12
u
ftl
a
c
10
p
(A
C
o
o
01
"u
j=
flirt
> '-/
X. C
in to
CM P
«•
t-
O
01
p
P ro
c c.
01
0 P
i. C
01 3
Q- O
I O
in in
•5
j.,
01
p
P ro
c s-
01
0 P
i. s:
ft) 3
a. o
1 U
in in
•5
u
14- in
o oi
s- -P
ft) •!—
o r— i
3 U
C ro
It-
c
0
ft) .,-
C P
•r- O.
0 3
tA IA
ro c
Ol 0
U
.a
ai oi
in in
ro ro
m u
ro
c
o
p
a
o
IA
ftl
Ol
01
ro
p
IA
i.
o
p
10
's
Ol
ftl
CfL
CO
CO
o
o
IX
O
in
CM
rH
CM
r~
o
px
f^
o
co
CO
o
^
01
•a
I
o
ro
c
1
1
ftl
Ol
to
p
CO
I-H
CO
rH
r^
H
rH
CD
CM
0
CO
r-i
co
ro
r-i
co
<3-
rH
CO
cn
CO
,_H
CM
ro
rH
in
CM
CM
ro
pH
o
l-i
o
rH
<
01
-o
o
p
to
c
I
1
HH
rH
01
O)
10
p
CO
CO
o
in
CO
co
ro
CO
p^
CM
^.
f^.
ID
^
cn
o
in
CO
i*-
in
CM
CM
o
CM
CO
r~
ro
in
o
CM
CD
rH
CM
CO
tn
CM
<
t
1
rH
rH
ai
Ol
to
P
CO
•a.
c
10
rH
01
Ol
10
P
CO
HH-
f^
CO
CD
in
0
in
ro
CO
cn
ro
IX
in
in
0
ID
8
CO
m
01
c
o
p
ro
c
rH
ro
ro
•o
ro
iH
T3
O
rH
ro
CM
cn
rH
CM
cn
rH
CM
cn
rH
Ol
•a
I
o
10
c
1
1
•o
to
o
JD
c
o
rH
CM
in
CO
CM
in
rH
ro
ro
rH
rH
CM
p_4
H
CM
ro
rH
CM
<
1
C
O
c
Ol
u
01
ftl
rH
rH
U
Ol
to
p
CO
»
CO
co
ro
rH
IX
CO
f»^
1-
CM
r^
•3-
CM
?
CM
CO
•o
c
ro ft)
TJ
(A •!-
ro 3
01 C
S- 0
ro -r-
P ro
c c
01 1
C S-
•r- 10
10 O
P JQ
-P C
ro o
VD
CO
ro
3
rH
in
IX
co
ro
CM
CO
*
CM
£
CM
<
1
C
o
c
r—
ro
I
rH
I— 1
01
Ol
ro
P
CO
r- '
ro
CO
^
cn
ID
CM
cn
CO
4-°
cn
ro
ro
cn
ro
ro
CO
ro
CO
ftl
c
o
in p
ro ro
Ol c
S- I
(0 I
TJ
P S-
C 10
01 O
§J3
C
•r- O
to
p -a
p c
ro ro
cn o
rH CO
*^ <3*
•a T3
•3" rH
o in
CM CM
•o -a
in co
ro CM
^ in
•3- CM
CO CM
CSJ CO
cn o
co ro
CM ro
O 0
CO «»•
CM CO
C
O
•o
c
ro Oi
•a
i-i 3
C
01 O
Ol-r-
ro P
P ro
CO c
rH
rH
X
rH -3-
CO CM
in cn
cn CM
ro CM
ro co.
•o -a
rH in
«*• co
ID O
rH
rx cn
rH CO
•31 IX
rH 0
CM CO
^> f*.
CO IT)
«t CO
< CO
•o
c
ro
ftl
-•o
rH -r-
C
01 O
01 .r-
ro P
P ro
CO C
I
rH TD
S-
01 ro
Oi O
ro JD
p C.
CO O
X
I
'
CM
ro
in
1
CM
cn
•3-
cn
o
c^
rH
1
m
co
CM
rH
1
1
1
in
o
CO
1
cn
CO
rx
<
c
c
o
u
•o
01
s-c
01
c
ftl
N
c
01
m
rH
X
1
1
rH
ro
"3-
rH
CM
^
CM
H
in
cn
rs^
ro
CO
rH
CM
ro
l
i
l
^
rH
CM
CO
CO
rH
CO
D
01
C
o
IA
10
01
01
ro u
P ro
o o s-
c p u
3 .;_
o s- P
to ro
c p
co o ro
•r- U
C P
O O "O
•f- ro -i-
P t- 3
a. i- •—
O H-
01
•0 P T3
C ro C
ro E to
IA O 0>
11 It- p
•r- ftl ro
Ol t. S
01 £-
p 01 O
to jr. i+-
S- p ftl
P S-
IA E
O E
>,£ 0
O 1-
p ftl
ro c ai
r- ftl C
3 N Ol
Ol C N
01 ftl C
s- a 01
a
ai P
J= C 14-
P ftl O
sS-
O ftl C
I- D. ftl
<4- U
in s-
IA . 01
Q) ^* o
•r- cn
p in
r- O *3-
— cn
O r-
it- > O
•i- oi ro
ro s- >
t a. 1
ftl JZ ftl
O P t
in in ot •
ts- P c
•r- 0)
oi 3 in u
x. cr oi s.
ftl ft) c. ,01
t- -i- O.
< 3
•4 O" CO
C ftl .
0 C S- rH
•r- 0 CO
p .r- CO
o a. c o
o o
rH - P O
r-l > 0.-1-
rH rH O P
XX 0
•- 3
jr >, p -o
Ol Ol C 01
3 ftl 01 S-
O P 0
S- to S- i—
c i- Ol ro
P P D. P
CO O
> «3- P
rH £-
O CM 10
(A U. CD
Ol i-
•i- It- O
O) • O »t-
fti in
p c c in
to o o c
S- -r- -r. 0
p p p •(-
CO D, U P
E 3 U
i- 01 T3 ro
O X
•D
S
u
>J
01
o
fj
ftl
s.
Ol
c
c
13
01
•o
o
p
D.
E
3
in
c
o
u
Ol
c
'o
IA
ro
Ol
Ol
C
E
u
ftl
01
E
3
in
tA
to
tA
01
P
10
E
P
(A
UJ
a
in
0)
c
o
ftl
c
o
p
ro
"D ,
a.
o
a
E
O
E
3
^3
C
ro
C
0
p
3
IA
O
u
ftl
c
o
(A
to
Ol
Ol
c
c
"u
ftl
IA
ftl
p
ro
o
a.
i.
o
u
c
p
ro
g)
O
*o
•a
o
P
QJ
QJ
O U)
l4_ Q)
(0 0)
QJ
•a P
QJ JO
•P U
c
QJ -P
w c
QJ CO
S- r-
a a.
W r—
0) QJ
3 O
CD E
C
QJ O
-C E
U
QJ
U)
JZ
>»
•a
QJ
Q.
E
QJ
•P
ro
s_
•p
c
. 3
O
u
in
"a
(0
QJ
S-
QJ
C
in
0)
u
c
QJ
S_
*P
•D
O
"D
QJ
3
(TJ
O
C
in
in
O
•P
C
o
u
S-
fO
o
c
o
o
^_)
o
u
p
c
3
0}
.c
D
O)
c
U) C
QJ 3
S- O
3 U
•P UJ
•5 -5
C
QJ QJ
a. ^
X -P
QJ
c •
to
p *•
•r— S- '
a. QJ
fO >
u CD
t- o
o .c
c -
o ~o
P P
(t) O
N QJ
i — 4—
fO QJ
3 S-
C
C QJ
ft) i-
to
c
to in
QJ
C P
o to
£-
0) -P
in c
m 3
A O
u
t/) in
"D QJ
P -P
ro -P aj
JZ C E
+J QJ -r-
S- -P
O) C QJ
JC 3 J=
P O -P
u
w in -P
o T- co
P in
C (U P
QJ £ in
C -P O
Re U
E "- -a
0 £-
U QJ CO
D) O
>, C jQ
C (0 C
< .c o
CO fO
in > c
CD .D C
in co
i — QJ QJ
CO 4-> -P
c u to
tO QJ Ol
**- QJ
P CO D)
> QJ CO
•P QJ
••- TJ JT
U5 r— +J
C 3
QJ O «*-
in .5 o
C
o
•P
3
"D
QJ
S_
QJ
C
QJ
- N
C
QJ
.2
QJ
+j
TJ
QJ
3
o
"O
. c
to
QJ
C
CJ
O
QJ
C
O
to
O)
•p
in
o
o
.Q
O
•D
QJ
3
a*
QJ
t.
in
Q)
S-
3
-P
C
QJ
Q.
X
QJ
QJ
T3
3
u
c
in
in
O
u
,_
o
_j_>
c
o
u
QJ
W)
QJ
.C
H-
QJ
8-11
-------
with the regulatory strategies when annual gasoline consumption is
assumed to remain constant at the projected 1986 level of 326 billion
liters (86 billion gallons) per year.
When constant gasoline consumption is assumed, the total cost
estimate for bulk terminals decreases 55 percent if storage tank costs
are included and 38 percent if storage tank costs are excluded. For
bulk plants, constant gasoline consumption decreases estimated total
cost by 14 percent under Option A and 7 percent under Option B. For
service stations, nationwide Stage II total cost decreases 17 percent
when exemptions are allowed and 8 percent without exemptions. (Stage I
costs for service stations are unaffected because service stations
receive no Stage I recovery credits.) The total cost of nationwide
Stage II with onboard, which still incorporates some decline in recovery
credits reflecting the phase-in of onboard controls, decreases only
about 6 percent under Option A. Total cost for Stage II with onboard
strategies in all nonattainment areas and in selected nonattainment
areas decreases 7 percent under Option A. Decreases are only 3 percent
for Option B under Stage II with onboard both nationwide and in non-
attainment areas.
As shown in Table 8-6, constant gasoline consumption decreases
the total cost associated with the nationwide regulatory strategies by
2 to 17 percent depending upon the individual strategy and option
examined. For strategies that affect only nonattainment areas, total
cost decreases 0.4 to 2 percent. Constant gasoline consumption also
affects the NPV ranking of regulatory strategies. Under Option A,
Regulatory Strategies VI and IX exchange rankings, while under Option B,
Regulatory Strategies VI and XII exchange rankings.
8.2.3.2 Declining Gasoline Consumption with Declining Numbers
of Facilities. Impact estimates are sensitive to the manner in which
the gasoline marketing industry is assumed to accommodate projected
declines in gasoline consumption. Two alternative methodologies—one
varying recovery credits and the other varying the number of facilities
and only indirectly varying total recovery credits—can be used to
estimate the cost impact under projected declines in gasoline consump-
tion. The declining recovery credit methodology that is incorporated
8-12
-------
in the base case assumes that projected declines in gasoline consumption
are shared by all existing facilities, with no facilities closing.
The result is a constant number of model plants with throughput
declining over time in each model facility. The declining number of
facilities method is based on the assumption that declines in gasoline
consumption would result in closure of marketing facilities, which
decreases the number of model plants over time with each facility
maintaining a constant throughput. Two variations of this assumption
are considered: uniform decline and skewed decline. Under the former
variation, the decline in number of facilities is assumed to be uniform
across model plant categories. Under the latter variation, smaller
model plant categories experience greater declines. This skewing
reflects recent industry trends toward larger, more efficient facilities.
For bulk terminals, when the percentage decline in number of
facilities is assumed to be uniform across model plant categories, the
total cost decreases 71 percent from the base case if storage tank
costs are included and 49 percent if storage tank costs are excluded.
When the decline in facilities is restricted to the three smallest
model plants (skewed decline), these total cost estimates decrease 60
and 42 percent, respectively. For bulk plants, if reductions in
facilities are assumed to be uniform across facility size categories,
Stage I costs decrease 18 percent under Option A (exemptions) and
11 percent under Option B (no exemptions). If the same decline in
throughput is restricted to the two smallest model plant size cate-
gories, Stage I total cost decreases 18 percent under Option A and
13 percent under Option B.
For service stations, the total cost for Stage I programs (nation-
wide and in nonattainment areas) decreases 6 percent when the station
population is decreased in uniform proportions as consumption declines.
Under the uniform decline, Stage II with onboard total cost decreases
6 percent under Option A and 3 percent under Option B for Stage II
control applied nationwide. For Stage II with onboard controls in all
nonattainment areas or in selected nonattainment areas, total cost
decreases 7 percent under Option A and 3 percent under Option B. The
8-13
-------
total cost for Stage II programs without onboard control decreases 22
percent under Option A and 13 percent under Option B. When it is
assumed that population decreases would be greater among smaller model
plant groups rather than larger ones (skewed decline), the total cost
for Stage I decreases 7 percent under Option A. All other changes in
total cost to service stations are the same under the skewed population
decline as they are under the uniform decline.
Table 8-6 shows the total cost figures associated with the
regulatory strategies assuming that the projected decline in gasoline
consumption results in comparable declines in numbers of marketing
facilities. The figures presented are for reductions in facilities
that are uniform across facility size categories rather than skewed
toward smaller model plants. For regulatory strategies that include
either Stage I or Stage II control (all except Regulatory Strategies
VII and XIV),.the total cost decreases 0.4 to 22 percent when the
facility population decreases over time. The smallest declines are
associated with regulatory strategies that include onboard control.
Regulatory strategies that include only Stage I and/or Stage II control
exhibit 13- to 22-percent declines in total cost when a decline in the
number of facilities is posited.
Clearly, the total cost impact of declining gasoline consumption
depends upon the manner in which the industry accommodates this decline.
If the decline is to be absorbed through decreases in the number of
facilities rather than decreased throughput at all facilities, cost
estimates decline. Moreover, the cost ranking of regulatory strategies
depends on the methodology chosen to incorporate declining consumption.
The relative positions of Regulatory Strategies VI.A and IX.A are
affected as are the rankings of VLB and XII.B. In recent history,
much of the gasoline marketing industry has been characterized by
declining numbers of facilities and increasing average facility size.
This pattern supports the assumption that declining consumption would
result in declining numbers of facilities rather than decreases in
average throughput.
8.2.3.3 Discount Rates. Varying the discount rate applied to
streams of costs is tantamount to changing the economic weight given
8-14
-------
to the future. A lower discount rate gives more weight to future
costs, while a higher rate gives less weight to these streams. However,
because streams of costs may be negative due to recovery credits, the
net effect of changing the discount rate is not easily predictable.
For bulk terminals, when storage tank control cost is included,
reducing the discount rate to 5 percent increases the total cost
13 percent and increasing the rate to 15 percent reduces this cost
7 percent. When storage tank costs are excluded, decreasing the
discount rate increases the total cost 36 percent and increasing the
discount rate reduces it 17 percent. The direction of these changes
is dictated by recovery credit streams for bulk terminals and bulk
terminal storage tanks, for bulk plants, reducing the discount rate
from 10 to 5 percent increases the total cost 35 percent under
Option A and 39 percent under Option B. Increasing the discount rate
from 10 to 15 percent reduces the total cost 17 percent under Option A
and 19 percent under Option B.
With the 5-percent rate, the total cost at service stations for
Stage I controls nationwide is raised 46 percent above the base case
under both Options A and B. Total cost for strategies involving
Stage I in nonattainment areas only increases 37 percent. Nationwide
Stage II without onboard total cost increases 48 percent under Option A
and 51 percent under Option B. Increasing the discount rate from 10
to 15 percent decreases the total cost for nationwide Stage I 22 percent
under both Options A and B. Total cost for Stage I in nonattainment
areas is decreased 16 percent under both options. Under Option A,
total cost for nationwide Stage II without onboard controls decreases
22 percent; under Option B it decreases 24 percent.
For both the 5- and 15-percent rates, the costs for nationwide
Stage II with onboard vary less dramatically than for Stage I and
Stage II without onboard. This is because most installations of
Stage II in combination with onboard involve only one equipment
lifetime, and no further Stage II costs are incurred after this point.
At a 5-percent discount rate, the total cost for nationwide Stage II
with onboard goes up 27 percent under Option A and 26 percent under
8-15
-------
Option B. Total cost for Stage II with onboard in all or selected
nonattainment areas increases 18 percent under both options. At a
15-percent rate, the total cost for nationwide Stage II with onboard
decreases 17 percent under Option A and 16 percent under Option B.
The decrease is 11 percent under both options for Stage II with onboard
in all or selected nonattainment areas.
In the case of onboard controls, reducing the discount rate from
10 to 5 percent increases the total cost over 60 percent. Raising the
discount rate to 15 percent reduces that estimate approximately 30 per-
cent. Note that in the case of onboard controls, the per-vehicle cost
is not adjusted for differences in the real discount rate. (Any
component of the unit cost estimate that is based on an annualization
of capital expenditures would be affected by a change in the discount
rate.) The different discount rates are reflected only in the discount-
Ing of the aggregate annual onboard costs at the given unit cost
estimate.
Based on the estimates provided in Table 8-6, lowering the discount
rate from 10 to 5 percent increases the total cost associated with the
regulatory strategies by a low of 37 percent (Regulatory Strategy
XII.B) to a high of about 74 percent (Regulatory Strategy XIV--benzene
reduction). Most regulatory strategies exhibit an increase of 40 to
50 percent. Reducing the discount rate changes the ranking of Regula-
tory Strategies IX and X on a cost basis. Raising the discount rate
from 10 to 15 percent reduces the total cost associated with the
regulatory strategies by a low of 20 percent (Regulatory Strategy
IV.A) to a high of approximately 34 percent (Regulatory Strategy
XIV—benzene reduction). Under Option B, raising the discount rate
affects the cost ranking of Regulatory Strategy VI. Under the base
case, Regulatory Strategy VLB is more expensive than Regulatory
Strategy XII.B; under the 15-percent discount rate, these strategies
are equally expensive.
8.2.3.4 Onboard Control Costs. While EPA's best estimates of
onboard control costs are used in the cost analysis, it is instructive
to consider what effect substantially higher onboard costs would have
8-16
-------
on the rankings of regulatory strategies. Accordingly, a higher
per-unit cost is also considered. The total cost of strategies
including onboard control rises by $1.4 billion (72 percent) when the
cost is $25 rather than $13 per vehicle tank. The effect of this
variation on the total cost rankings of the regulatory strategies
depends upon the option selected. Under Option A, Regulatory Strategy
VI becomes.cheaper than Regulatory Strategies VII, VIII, and IX when
the higher onboard unit cost is used. Under Option B, the higher unit
cost of onboard control reverses the total cost rankings of Regulatory
Strategies VI and XII.
8-2.3.5 Enforcement Cost. Total costs are sensitive to assump-
tions concerning the level of enforcement required to maintain various
levels of in-use efficiency. The enforcement costs in this chapter
are based on annual inspections; however, scenarios assuming more or
less frequent inspections would result in different costs. Two alterna-
tive enforcement scenarios are considered in this section. In keeping
with the in-use efficiency analysis in Chapter 7, these variations
involve only Stage II controls. The first sensitivity analysis assumes
quarterly inspections of Stage II facilities. In the second sensitivity
analysis, "minimal enforcement" assumptions are incorporated. As
discussed in Chapter 3, a minimal enforcement situation is one in
which no State or Federal resources are allocated to Stage II program
enforcement.
When quarterly inspections of Stage II facilities are assumed,
nationwide Stage II total cost is increased 7 percent under both
Options A and B. Total cost for Stage II with onboard is increased
about 6 percent under both options. It should be noted that compliance
levels under quarterly enforcement are assumed to be the same as under
annual enforcement in this analysis.
When the minimal enforcement scenario is applied, enforcement
cost is eliminated, but several other changes also occur. Therefore,
the net effects of such a scenario are difficult to assess. Since
fewer facilities comply, lowered capital cost can also be assumed. As
in the "in-use efficiency" analyses of Chapter 7 (Section 7.2.5.4),
8-17
-------
it was assumed that, due to noncompliance, minimal enforcement control
cost would be 20 percent lower than the cost under full enforcement.
Also, since fewer facilities comply, credits for vapor recovery are
correspondingly reduced. Total cost for nationwide Stage II is reduced
15 percent under Option A and 20 percent under Option B. Stage II
with onboard control total cost decreases 19 percent under Option A
and 21 percent under Option B.
8.2.4 Unit Cost and Quantity Impacts
Substantial cost variations translate into widely varying unit
cost and quantity impacts for the regulatory strategies. These values
are shown in Table 8-7. (As discussed in Section 8.1, unit cost and
quantity impacts are estimated only for nationwide strategies. Conse-
quently, Regulatory Strategies VIII and IX have been omitted from
subsequent tables.) The estimated increase in unit cost of gasoline
ranges from 0.034$/liter to 1.17t/1iter (0.ISC/gal Ion to 4.43
-------
to
UJ
(3
LLJ
t
g
H-
>
1
— J
S
UJ
Q
rH
O
rH
1-
O 1-H
U. l~
CL
O CO
UJ CO
U- <
u.
UJ UJ
£5
^— UJ
=E to
Z> CO
o*
ce
O UJ
z o
< z
i —
CO
0
o
1—
z
LLJ
CD
ce
LU
i
LU
.J
CO
•o
c
O ^^
o E n
-J 3 O
I/I rH
C C ^
O O
•P
o
3
01 C
S- O
r- 4J i-
5 > i- >>
C Q E \
O -J 3 CT
•P C rH
10 o ^-*
C U
a
CT
(0
> U S-
** £ -?>
o \.
Ol •!- S-
C 4-> Ol
O S r—
in in
10 C (D
o o o
U iH
^•^
J3
If- S-
4J 0 01
in oi 4->
O 01 C -r-
O l/l -r- r-
10 1— \
•r- S- U)
C U 10 CM
n c oico
i-H
'*•'
C
O
&
>1
D}
P
(0
S-
•p
>>
o
p
10
Oj
01
ct:
it ii
ii ii
ii ii
00 rH rH CM
H CM CM CO
* co in *3-
ro in ic cri
O O O rH
O O O O
< m < co
O)
Ol T3
•O •!-
'r- 2
2 C
C 0
O i-
P «
ro c
C 1
i i
rH HH
O> Oi
O) Ol
ro to
CO CO
>-« >
II ro to ro
in in in
ii r>.° i-: i>;
•3- «3- 1 ro rH
rH Lf) 1 ^- «*
*f O rH CM
rH
01
*O TD ro T3 T3
C -r- C C •!-
ro 3 i ro 3
C 1 C
rH O -O rH O
01 4-1 ro oi P
01 ro o 01 ro
ro c JD ro c
4-> C 4->
CO O CO
«
rH rH
> > X
ro ro
in in
r-: f>:
0) Ol
ro to
«3- CM
CO O
i-H
Ol 01
CM in
00 «3-
O CM
o o
< CO
t
"O
s-
10
o
c
o
•o c
c -a
"•5
l-l C
rH O
01 P
01 ro
10 C
CO
^
rH
X
ro ro
in in
01 01
ic co
00 CO
*fr CM
rH
01 01
CD CM
rH 0
rH ro
0 0
<. CD
rH
I
a> i
O1T3
ro s- 01
4J ro -o
CO O •!-
ri s
» c c
rH O O
01 c ro
ro 10 c
p
CO
«
X
1 1
1 1
1 1
1 1
o o
en ic
I-H *(•
1 1
O 0
CD O
CD CM
rH ti-
ro r~
* rH
O rH
1 1
O rH
11^ O
o' r-!
< CO
01
c
o <—
4-> U!
01 O 10
C 3 Ol
N 01 C
C S- -r-
co
..;
X
4J
C
10
4->
in
c
o
u
10
•a
c
10
„
I/I
4J
?
S-
o
&
01
•>
o
u
01
£-
Ol
c
c
u
01
•o
«
c
o
4J
in
c
o
u
Ol
c
, —
o
ro
Ol
Ol
c
c
"o
Ol
•o
_2
u
c
ul .
C I/I
O 01
4J P
D.-I-
E r-
3 -r-
in u
in ro
ro !r-
01 n-
01 O
10
u s-
01
01 JD
CO C
ro
CO
ro
>,
>,
"a
4J
3
E
O
ro
O)
\
S
L.
01
>
c
o
u
r-
HI
4-1
I/I
ai
i. .
S 01
in u
01 £
E a
*o c
10
E 01
•o ro
ai
14- S-
o u
c
>>•"-
•r- P
U C
•r- 01
P U
l/l t-
10 01
r- O.
0) 1
rH
01
U 10
S- O
a. P
01 01
1— "D
01
in.5
in p
O P
1- C
0 3
Ol • t-
C 0)
•r- O.
'o -o
in oi
ro "o
01 C
o a>
&1
•r- 4-1
U V-
P C
in ro
r- O-
01
c
u
'£ o
ft5
ro o
Ol "O
c a.-
•i- S-
m
3 o>
•o ro
01 P
P C
3 01
Ife
U 1
^
o
i-
p
c
o
u
•o
i.
ro
&
c
o •
rH
!- rH
01 •
•O rH
j;
3 in
10
1/1 1
01
in ~o
10 01
0) in
S- 3
o
c in
•^ 01
01 U
O 'r-
i- 01
a. >
Ol 4J
0 01
.C r—
> »-
O
££
(U U
^ U)
r ,2
4J OJ
p
re a)
u
o ••-
S- S-
*0 CL
O £
••- 1—
4_J
t .
3 in
in ai
C in
o ro
U 01
r- O
0 C
•o oi
c u
ro *^
> Q.
a
c
c •.-
o
C (0
O Ol
P O
O P
3 .
3
O
1
01
ro
a.
, t
01
Ol
ra
P
CO
o
t(_
01
in
4-)
u
01
14—
01
p
p
c
ro
•o
ro
P
in
o
u
P
a>
8-19
-------
I *>
I 4J C
* C 30
I I Cl O
I' U) U U <
I wi t- U>
I Cj 1-
' C_ Tl
s
*> e
o o CM o
O r-i O CM
O O O O
O O CM O
•r-H O CO O
O 00 C
01 0)
I •!- t-
(J 4-> t. S-
!—•(-(/} 0) -P
f— r- C -P fO U) •
m o o co
O) v) u
-------
these estimates are discussed in terms of deviations from the base
case, which assumes declining gasoline consumption, declining recovery
credits, and a 10-percent real discount rate.
When constant gasoline consumption at the 1986 level is assumed,
unit cost impacts decline 29 to 35 percent under Option A and 26 to 27
percent under Option B. These declines are attributable to different
recovery credit streams and a different level of production over which
costs are spread. When declining gasoline consumption accompanied by
a declining number of facilities of constant throughput is assumed,
unit cost impacts decline 18 to 23 percent under Option A and 13 to 14
percent under Option B. These declines are attributable principally
to different recovery credit streams. .... >
The unit cost impacts under various discount rates are also
presented in Table 8-8. If the discount rate used to evaluate future
streams of expenditures is 5 rather than 10 percent, unit gasoline
cost impacts generally decline 9 to 16 percent. (The phasing of
Stage II and onboard controls under Regulatory Strategies XII and XIII
produces some differences in impact patterns.) The largest decline is
experienced under Regulatory Strategies IV and X, where the unit cost
impact decreases 15 percent under Option A and 16 percent under Option B.
Assuming a 15-percent discount rate, the unit cost impact increases 0
to 18 percent depending upon regulatory strategy and option selected.
The smallest increases occur under Regulatory Strategies XII and XIII.
All other strategies exhibit increases of 8 percent or more.
Certainly, the magnitude of average unit cost impacts depends
upon the costing assumption chosen. Moreover, the ranking of regulatory
strategies by average unit cost impact sometimes changes when assump-
tions are varied. (Regardless of option, Regulatory Strategies VI and
XII switch rankings when a 5-perceht discount rate is substituted for
the 10-percent discount rate.) It should be noted, however, that such
switches occur only when the impacts associated with the regulatory
strategy are close in magnitude.
Quantity impacts are estimated by applying price elasticities of
demand for gasoline and automobiles to estimated average unit cost
8-21
-------
increases. Table 8-9 presents the sensitivity of gasoline quantity
impact estimates to differing costing assumptions. Quantity adjustments
are not used in this preliminary study to refine cost estimates of the
regulatory strategies. Such refinement is difficult to make and is
not likely to have significant impact on the costs. Assuming constant
gasoline consumption, quantity impacts decline 13 to 18 percent under
Option A and 7 to 8 percent under Option B. Under the assumption of a
declining number of facilities, quantity impacts decline 18 to 23
percent under the exemption options and 12 to 13 percent under the
nonexemption options.
Table 8-9 also presents estimates of quantity reductions under
different discount rates. When a 5-percent discount rate is used, the
effect on quantity impacts ranges from a 5-percent increase for Stage II
with onboard under Option B to a decrease of 16 percent for Stage I
alone under Option A. The wide variation is primarily due to the
different recovery credit streams implied by the strategies. Using a
15-percent discount rate, the quantity impacts either remain constant
(Regulatory Strategy XII.B) or increase from 2 to 17 percent.
For the purposes of the analysis above, a long-run gasoline
demand elasticity of 0.55 was used, based on Department of Energy
estimates.3 It should be noted, however, that no consensus exists as
to a specific, national, long-run elasticity of demand for gasoline,
and that 0.55 is a mid-range estimate. For sensitivity, low- and
high-range estimates that bracket the published estimates were substi-
tuted.3 Table 8-10 demonstrates the sensitivity of gasoline quantity
impact estimates to variations in price elasticities of demand for
gasoline. Using a higher estimate for gasoline price elasticity
nearly triples the quantity impact estimates calculated using the base
elasticity. Using the lower price elasticity for gasoline more than
halves initial quantity impact estimates.
Table 8-11 shows the sensitivity of automobile quantity impacts
to changes in per-unit control cost assumptions. Increasing the unit
cost for onboard controls to $25 per vehicle tank (about a 90-percent
increase in unit cost) increases LDV quantity reduction 88 percent and
LOT reduction 38 percent.
8-22
-------
00
LU
to
Ul
u
TZ
r— I
I—
00
o
1-
z
3
LU
111
>
O Z
r- O
O r-
Ul a.
^ ~^
CQ V) (•—*
l-l 00 (-
o: < IB
ti-Sl
^
3 «0
O LU ^->
— ••
3.
rH Hi
Si
g =
z
hH
W
§
a
LU
ce
en
i
CO
LLI
1—
+>
4J C
C 3 01
1 01 O *>
in u u ID
rH S- Ul S-
a. TJ
*> 'c
C 3 01
01 O P
1 O O 10
in c- in s.
01 •*-
Q. TJ
.Q
Ul
Ol«t- 01
C 0 i-
C i. -r-
•r— 01 r—
u e u
01 3 10
a c «-
«..£
C C -P
ID *r- D.
•P r- E
Ul O 3
C in in
one
o 01 o
u
ai oi
in in
IB IB
CO U
c
o
p
o.
o
01
10
s-
m
o
10
"3
01
01
oc
rH CM
CM O
rH CM
rH rH
rH CM
•J- «!•
CM CM
H CM
CO rH
rH CM
< CO
01
TJ
1
-natib
i
o>
01
10
•p
00
t-l
eo co
0 0
ro en
ID CM
CO CM
CM r**
O «3-
i-l 0
CM r-.
CM PO
CM *
CM r»
rHCM
CM oo
< CD
.1
o
.p
03
C
i
i
o
D)
(0
UO
>
m m
r*. co
co en
rH in
«*• o
rH
10 TJ
"'I
01 O
O1T-
10 <->
•P ID
00 C
>
1
1
1
1
1
1
,
1
01
-natio
i
•o
£_
ID
O
a
c
0
.
>
rH CM
o r-
CM O
rH CM
f- rH
rH rH
rH CM
- *f *3-
CM CM
rH CM
PO rH
rH CM
< m
i
i
•o
s-
o
c
o
•a
c a>
10 TJ
01 O
Ol-r-
ID -P
•P ID
OO C
>C
CTI o ID o in r*» PO f**
CO O ~~" in rH CO CM
rH ""rn""'
en o IT>^ O CM ID CM
CM O It rH rH CM
rH
COOVOO ID 00 CO rH
CO O ^f rH CM CM
rH rH
< m < CD
1 'O
•a IDO
ID -TJ
O HH •!-
.Q rH 3
o ai o
O) 'r.
"O 10 +J
CO 4J 10
ID 0) 00 C
"O 1
KH *r- . |
l-l S 1— T5
C t,
ai o at ID
Ol'r- O! O
P ID +J C
OO C OO O
H^
rH HH
X X
1 1
OOOO
co en in CM
rH rH «• in
OOOO
o vo co in
rH rH PO *
Illl
Illl
i i i r
il il
OOOO
ID Cn O ID
ID r- CM CO
rH rH «•
>
5
in
m
o
4-
O
0)
^
o
in
ID
Ol
0
2"
•r-
U
•f-
.p
tn
01
01
u
i-
O-
ID
Ol
C
in
3
^j
3
&
O
p
C
2
a
1
O!
C
o
E
10
in
0)
c
u
a>
TJ
g
o
<*-
c
3
"
ID
O
Q.
O
U
c
^
ID
£
<->
oi
r^
O
o
£
•p
i
.c
*>
i.
£
g
resent*
a.
u
2
•?
'+-
0)
3
u>
C
O
. u
3
•a
2
>>
p
c
C7
4->
C
01
in
s-
a.
0)
IA
a>
in
.c
P
c
m
O.
C
Ul
1
Dl
Lu
•£,
o
1
01
in
Q
o.
rH
rH
Ol
Dl
ID
p
00
2 .
O P
<*- 3
01 O
.a i
in ui
C ID
O S.
5 a
U t-t
3 t-t
•D
2 S,
ID
^_
P W
c ai
ID P
3 Ir-
o- w
U
8-23
-------
TABLE 8-10. GASOLINE QUANTITY IMPACTS: AVERAGE NATIONAL REDUCTIONS IN
CONSUMPTION UNDER CONSTANT GASOLINE CONSUMPTION
WITH VARIOUS ELASTICITY ASSUMPTIONS3
(106 liters/year)
IV.
V.
VI.
VII.
X.
XII.
XIII.
XIV.
Regulatory strategy
Stage I — nationwide
Stage II — nationwide
Stage I and Stage II —
nationwide
Onboard — nat i onwi de
Stage I and onboard —
nationwide
Stage II and onboard —
nationwide
Option
A
' B
A
B
' A
B
A
B
A
B
Stage I, Stage II, and A
onboard — nationwide
Benzene reduction in
gasol ine
B
A
B
Base
elasticity
,(n = 0.55)
143
241
271
812
414
1,054
—
143
241
343
(0)
1,026
(0)
486
(143)
1,268
(241)
1,660-
1,790
4,200-
4,860
Higher
elasticity
(n = 1-59)
413
697
784
2,349
1,197
3,046
—
413
697
990
(0)
2,967
(0)
1,404
(413)
3,664
(697)
4,800-
5,160
12,100-
14,000
Lower
elasticity
(H = 0.24)
62
105
118
355
181
460
,
62
105
149
(0)
448
(0)
212
(62)
553
(105)
725-
779
1,830-
2,120
aThe price elasticity of demand (n) measures the percentage reduction in
quantity demanded per unit time due to a 1-percent increase in price.
Quantity reductions before Stage II phase-out. Figures in parentheses
represent quantity reductions after Stage II phase-out.
8-24
-------
TABLE 8-11. REDUCTIONS IN VEHICLE CONSUMPTION ATTRIBUTABLE TO
UNIT COST INCREASES UNDER ALTERNATIVE COST ASSUMPTIONS
LDV
Unit price =
$13/vehicle tank
Unit price =
$25/vehicle tankc
103/yra
17.7
33.3
Percent
0.16
0,31
LOT
103/yra Percent13
5.3 0.18
7.3 0.25
Computed using a price elasticity of light vehicles of 1.11.
Percentage of average LDV or LOT consumption over the period of the analysis
(10.78 x io6 LDV's per year; 2.91 x io6 LDT's per year).
'Represents about a 90-percent increase in unit control cost.
TABLE 8-12. REDUCTIONS IN VEHICLE CONSUMPTION ATTRIBUTABLE TO A
UNIT COST INCREASE OF $13/VEHICLE TANK WITH VARIOUS PRICE ELASTICITIES'
LDV
LOT
103/yr
Percent
103/yr
Percent
Base elasticity
(H = 1.11)
Higher elasticity
(H = 1-70)
Lower elasticity
(p = 0.22)
17.7
27.2
3.5
0.16
0.25
0.03
5.3
8.1
1.1
0.18
0.28
0.04
The price elasticity of demand (r)) measures the percentage reduction in
quantity demanded per unit time due to a 1-percent increase in price.
Percentage of average LDV or LOT consumption over the period of the analysis
(10.78 x io6 LDV's per year; 2.91 x io6 LDT's per year).
8-25
-------
Table 8-12 demonstrates the effect on the quantity Impact for
vehicles of varying price elasticities. Assuming a higher price
elasticity of demand for light vehicles increases estimated quantity
reductions over 50 percent from the base estimate -for both LDV's and
LDT's. Assuming a lower price elasticity reduces estimated quantity
reductions approximately 80 percent. Clearly, quantity impact estimates
are very sensitive to price elasticity assumptions. Moreover, estimates.
of the price elasticities of demand for light vehicles vary consid-
erably from study to study.4
8.3 DISTRIBUTIVE IMPACTS
Imposing a standard, whatever its design, will affect the profit-
ability and competitive position of firms and the well-being of con-
sumers. Some industries may benefit from increased demand (e.g.,
carbon canister producers) if onboard controls are selected, and some
firms may be better positioned than others to respond to a standard's
requirements (e.g., a new service station may already have installed
Stage I and Stage II systems in anticipation of a standard requiring
them). There will, -in short, be "winners" and "losers," but these
gains and losses are not necessarily economic costs in the strict
sense of the term. The distribution of these outcomes over classes of
firms and individuals as well as over time is, however, part of the
economic impact.
As consumers and firms adjust to the new set of prices and costs
resulting from a standard, a number of transitory conditions are
likely to occur. Holders of fixed capital may find that their rate of
return is less than what led them to inves.t in, that line of business.
Because of the nature of the fixed capital commitment, however, the
facility will continue to produce until operating costs cannot be
recovered. In the short run, then, the firm is selling its output at
less than a normal rate of return. Similarly, a firm may find that,
because it has production facilities in place, it is in a good position
to take advantage of the new cost and market conditions and that its
profits are higher than expected. Over time, capital depreciation and
reinvestment opportunities permit firms to leave or enter various
8-26
-------
lines of business and new, normal profit levels of activity and industry
structure will be established. Some of these adjustment gains or
losses and the longer term effects of the new cost conditions on
industry structure are discussed below.
The regulatory strategies covering Stage I, Stage II, and benzene
removal undoubtedly would impact the competitive balance and structure
of gasoline production and marketing. The data summarized in Table
8-13 show that under the regulatory strategies the average cost of
smaller distributors would increase more than that of large distrib-
utors. Where measurable, the ratio of additional control cost per
unit of throughput for larger facilities to additional cost per unit
of throughput for smaller facilities ranges from 3 to 76. (For top-
loading bulk terminals this ratio is undefined because the unit cost
increase for the largest model facility is negative, but close to
zero.)
The qualitative analysis presented in Reference 1 shows that such
cost increases would most likely result in fewer, larger facilities.
Because of economies of scale, this reduction in the number of facili-
ties would be more than proportional to the quantity reductions
presented in Table 8-7. This situation would amplify a trend already
well established in the gasoline marketing industries: a 9-percent
decline in bulk terminals between 1972 and 1978, a 30- to 40-percent
decline in bulk plants between 1972 and 1982, and a 36-percent decline
in conventional service stations between 1972 and 1982. Use of exemp-
tions reduces but does not eliminate the cost differential between the
largest and smallest facilities that have to control. Exemptions
would actually improve the competitive position of the exempt firms and
increase competitive pressure on firms just above the cutoff levels.
Differential regional impacts also would be associated with the
regulatory strategies involving gasoline marketing. Rural and attain-
ment areas would experience relatively greater cost increase because
these areas generally have smaller average facility sizes and do not
have baseline Stage I or Stage II controls. For similar reasons,
benzene removal would result in about twice the average cost increase
8-27
-------
TABLE 8-13. DIFFERENCE IN AVERAGE UNIT COST FOR SMALL AND LARGE FACILITIES
UNDER DECLINING GASOLINE CONSUMPTION
Facility
Smallest
model
facility.
(
-------
for refineries in Petroleum Administration for Defense District (PADD)
IV (northern Rocky Mountains) as in PADD III (Gulf Coast).
It is difficult, on this preliminary basis, to estimate whether
regulatory strategies involving onboard control would result in differ-
ential cost impacts or changes in industry structure. If, on an
international basis, automobile producers are competitive and U.S.
producers are the marginal producers, most of a sales reduction would
accrue to U.S. producers.
8.3.1 Petroleum Refineries
The distributive impacts of the benzene reduction regulatory
strategies can be analyzed by both facility size and region of the
United States. According to ADL's 1978 report,5 the control cost of
benzene reduction varies dramatically with the scale of the refinery.
For small refineries (10,000 barrel/stream-day), ADL's scaling factors
result in direct control costs of reduction per liter of gasoline that
range from three to six times the average direct cost of reduction for
U.S. refineries as a whole. For a small refinery with both reforming
and FCC capacity, the cost multiple is closer to three. Differential
unit costs, based on the national averages of Table 8-3, are displayed
in Table 8-14. With an average wholesale gasoline price of about
$0.26/liter ($0.99/gallon) (as stated in Chapter 7), these figures
suggest greater than a 3.6- to 9.0-percent rise in the average unit
cost to small refineries. The U.S. average unit cost increase would
be about 1.2 to 2.9 percent, depending on the regulatory option.
A substantial number of small producers operate in the petroleum
refining industry and, to the extent that each of their refineries is
also small, these figures indicate they would be put at a substantial
competitive disadvantage relative to gasoline producers operating
large refineries. Based on an analysis of differential cost shifts,
these regulations would tend to reduce the number of small refineries,
especially if, as suggested by scale factors and age distribution,
small refineries are the marginal producers of gasoline. The regula-
tion, operating in conjunction with forecast declines in gasoline
demand over the next decade, would likely accelerate this industry
8-29
-------
TABLE 8-14. INCREASED COSTS TO PETROLEUM REFINERIES DUE TO BENZENE
REDUCTION IN GASOLINE FOR A 10,000-BARREL/
STREAM-DAY REFINERY VS. THE U.S. AVERAGE
Costs
Option A
Option B
Increased unit costs for a 10,000-
barrel/stream-day refinery with
reforming and FCC capacity,
-------
"shakeout." Furthermore, cost differentials could increase if the
cost of capital for small refineries were higher than that for large
refineries.
The cost of benzene reduction varies considerably with geographic
region. When price indices are applied to results of ADL's 1978
report,6 the distribution of estimated average unit costs for each
PADD is obtained; these are shown in Table 8-15.
Other markets, especially chemical markets, may be affected by
benzene reduction regulations. The ADL report,7 particularly, cites
the possibility of a benzene glut, which it considers a cost of regula-
tion. Environmental considerations aside, it is economically and
commercially more correct to consider the additional benzene produced
by benzene reduction as a credit or benefit of the regulation. This,
indeed, is what ADL did when it used the difference between benzene's
value in gasoline and the next best alternative use of benzene to
estimate the cost of volume loss. Chemical benzene was wholesaled at
$0.41/1 Her ($1.55/gal) in May 1982,8 a price 50 percent greater than
the prevailing wholesale price of gasoline. If benzene reduction
doubles the benzene supply and if the price of benzene drops as a
result of this increased supply, industrial chemical users and producers
may find it profitable^to switch to benzene in their processes. As an
upper bound, the benzene credit as an industrial chemical may exceed
the direct-volume-related loss of benzene in gasoline. If this is the
case, a "volume credit" rather than a "volume loss" should be incorpor-
ated into the indirect cost analysis.
Markets for other chemicals, especially those used to boost
octane, also would be affected by the two benzene reduction strategies.
If their prices increase as a result of an increase in demand for
octane boosters, the. cost of octane loss due to benzene reduction
would have been underestimated by the ADL analysis. In particular,
the price per octane-gallon would tend to be higher than estimated.
8.3.2 Bulk Terminals and Bulk Plants
As noted in Chapter 4 of this report, an estimated 67 percent of
total gasoline throughput is handled currently at controlled terminals,
8-31
-------
so 67 percent of the gasoline terminal population is assumed to be
controlled. The remaining 33 percent of gasoline terminals would be
controlled under several of the regulatory strategies considered here.
In particular, Regulatory Strategies IV, VI, X, and XIII include
Stage I controls nationwide. Gasoline bulk terminals would be impacted
under each of these strategies, regardless of option, and the impacts
would be the same in each case. No differences in coverage are proposed
and no exemptions apply.
Cost variations across model terminals are significant. Top-
loading facilities experience the largest impacts. Within each plant
size category, both the total cost and the annualized cost of control
over the projection period for a top-loading terminal are higher than
the corresponding costs for a bottom-loading terminal. For the three
smallest plants, the ratio of such costs for a top-loading facility to
the corresponding costs for a bottom-loading facility ranges from 2 to
57. For the largest size plant, the bottom-loading facility exhibits
negative 1986 NPV of control cost and annualized control cost while
the top-loading facility incurs positive costs.
Within loading classifications, the two larger model terminals
always generate lower 1986 NPV of control cost and annualized control
cost than do the two smaller terminals. The largest model terminal
always shows the lowest such cost. The smallest model terminal
generates the highest such cost for bottom-loading facilities while
the next-to-smallest terminal realizes the highest cost for top-loading
facilities.
In summary, larger terminals experience lower NPV of control and
annualized control costs, whether viewed on a total or per unit of
throughput basis. Thus, there are economies of scale in Stage I
control for terminals. These economies will reinforce any economies
of scale in production. Accordingly, the process of market rationali-
zation, which would probably continue to exert pressure on smaller
terminals even in the absence of additional control, can be expected
to continue.
8-32
-------
As Regulatory Strategies IV, VI, X, and XIII include Stage I
control nationwide, bulk plants would be impacted under each strategy.
However, each of these strategies includes two options. For each
strategy, Option B requires incoming and outgoing vapor balance systems
for all bulk plants. Option A exempts small bulk plants (throughput
less than 15,000 liters/day) from outgoing balance requirements but
retains the requirement- of submerged filling of account trucks.
Consequently, the number of bulk plants affected would be the same
under each option (8,040), but costs assessed for the smaller plants
(Model Plant 1) would vary by option.
Because cost impacts are not uniform across model bulk plants,
industry organization would experience some impact. Generally, NPV of
control cost and annualized control cost decrease as plant size
increases. Thus, control costs for bulk plants exhibit economies of
scale as do basic production costs. Accordingly, additional regulation
would reinforce the industry trend toward larger, more efficient
plants. Regulatory cost would put additional pressures on small
facilities, many of which are already marginal. However, because of
exemptions considered, potential impacts would vary by regulatory
option.
Under Option A, some of the pressure of control cost on the
smallest facilities would be eased. However, for plants with through-
put greater than 15,000 liters/day, economies of scale would still be
associated with control cost as well as production cost. The industry
trend toward larger, more efficient bulk plants, therefore, is likely
to continue under Option A. Under Option B, Stage I regulation com-
pletely reinforces this trend.
Geographic distribution of impacts is difficult to assess.
Impacts would be restricted to attainment areas, which are not concen-
trated on the East Coast and in the Midwest (PADD's I and II). However,
the bulk plant population is concentrated on the East Coast and in the
Midwest, only portions of which are attainment areas. Consequently,
impacts would be spread across much but not all of the United States.
It is not possible given present data to be more specific about regional
impact variations.
8-33
-------
8.3.3 Service Stations
Costs of potential regulations do not affect all service stations
in the same manner. Small stations in particular are subjected to
disproportionately higher unit costs than are larger firms since
larger firms are able to distribute the slightly higher fixed cost
over a much larger throughput. Stage II unit costs for stations with
high throughput (Model Plant 5) are only 1.0 percent of those of
low-throughput (Model Plant 1) stations. For Stage I, high-throughput
unit costs are 2.7 percent of low-throughput costs. Some low-throughput
stations will likely have difficulty competing under this skewed cost
burden, particularly given the shrinking margins in today's market, and
might close, allowing larger stations to take up the sales volume.
The difference between the costs of nationwide Stage II programs
with and without onboard controls also affects the distribution of
costs across model plants. The 1986 NP.V of control costs per station
for the various model plants under nationwide Stage I and Stage II
with and without onboard are presented in Table 8-16.
Differences in distributional effects of the two programs are
evident in the table. Smaller model plants bear a higher per-station
control cost as a rule when Stage II without onboard is implemented
largely due to higher recovery credits realized by the larger model
plant groups. When onboard is combined with Stage II, however, the
larger stations are assumed to lose the recovery credit advantage over
time, and thus bear a higher per-station control cost.
The assumptions incorporated in the sensitivity analyses can
cause variation in the distribution of per-station costs. Under the
constant gasoline consumption case, the differences between per-station
costs over the model plant sizes for Stage II with and without onboard
are more extreme. Recovery credits remain constant in the constant
gasoline consumption case, causing annualized control cost per station
to be as much as 70 percent lower than in the base case for the large
model plant for Stage II without onboard. The decrease for the smallest
model plant is less substantial: 2 percent less than in the base
case. The small model plant thus has a much larger annualized control
8-34
-------
TABLE 8-16. 1986 NPV OF CONTROL COST PER STATION BY MODEL PLANT'
(1982 dollars)
Stage I
Stage II — without onboard
With exemption
No exemption
Stage II — with onboard
With exemption
No exemption
MP1
2,984
12,399
10 , 134
MP2
2,984
12,278
12,093
10,770
10,629
MP3
2,984
12,923
12 , 647
12,018
11,809
MP4
2,984
15,669
15,669
15 , 685
15,685
MP5
2,984
14,260
14,260
20,449
20,449
Enforcement cost is not included because it is not borne by the service
station sector.
8-35
-------
cost per station than the large model plant with constant recovery
credits when Stage II without onboard is imposed. For Stage II with
onboard, which includes declining recovery credits to reflect Stage II
phase-out, the changes in per-station control cost relative to the
base case are less dramatic, ranging from a negligible decrease for
the smallest model plant to a 16-percent decrease for the largest.
Stage I per-station costs do not vary in the sensitivity analysis
since no recovery credits are assigned. The per-station costs for the
declining facility cases are not directly comparable to the base case,
due to the variation in station populations over time.
The service station industry is currently undergoing structural
change. The number of stations is decreasing as small stations close
and are replaced by larger, more profitable stations. This trend is
due in part to declining gasoline consumption in recent years, which
has lowered price margins and has forced the higher cost stations to
leave the industry, and in part to the "stand alone" economic philosophy
that has induced oil companies to take advantage of large-station
economies of scale, which is discussed in more detail in Reference 1.
The higher unit cost estimated for small service stations would
exacerbate this existing market trend. Although some small firms will
still have a market niche in the industry, others would be forced to
leave the industry more quickly and in greater numbers if the additional
cost of regulation were imposed upon them. As increased costs are
passed along to customers and retail prices rise, quantity demanded
decreases. Station owners will seek means to maintain volume and cut
costs. These means might include moves toward high-volume locations
and might tend to speed the trend toward self-serve stations.
The market implications of regulations exempting small and inde-
pendent stations are twofold. First, the exemption of small and
independent stations effectively subsidizes them. If, as may be the
case, smaller stations are marginal in the industry, such exemptions
would protect the less efficient facilities at the expense of the
mid-size stations. Second, with exemptions, the middle throughput
range stations would bear higher increases in unit cost. They would
8-36
-------
be less competitive with both the largest and smallest exempt segments
of the industry.
8.4 REFERENCES
1. Research Triangle Institute. Preliminary Economic Impact Analysis
of the Regulatory Strategies for the Gasoline Marketing Industry.
Draft report to U.S. Environmental Protection Agency. July, 1984
Docket No. AT84-07. , .
2.
3.
6.
7.
8.
National Petroleum News. Factbook Issue. July 1983. p. 91.
Memorandum from Aitken, Mary, Research Triangle Institute, to
Morris, Glenn, Research Triangle Institute. April 26, 1984.
Gasoline Demand Elasticities for Use in Benzene/Gasoline Marketing
Analysis.
.Memorandum from Giberson, Linda, Research Triangle Institute to
Robson, John, EPA. December 21, 1983. Price Elasticity of
Demand Estimates for Automobiles.
Arthur D. Little, Inc. Cost of Benzene Reduction in Gasoline to
the Petroleum Refinery Industry. U.S. Environmental Protection
Agency. EPA Contract No. 450/2-78-021. April 1978. pp. 5-25
through 5-31.
Reference 5. pp. 5-17 through 5-22.
Reference 5, pp. 6-11 through 6-17.
Chemical Marketing Reporter. May 10, 1982. p. 37.
8-37
-------
-------
9.1
9.0 ENFORCEMENT STRATEGIES AND COST CONSIDERATIONS
ENFORCEMENT STRATEGIES
9.1.1 Stage -II Programs
Generally speaking, enforcement of an air pollution control
program occurs in two phases. The first phase consists of assuring
installation of acceptable control systems and the second, applying to
cases where the control systems are subject to damage or other malfunction,
consists of assuring that the installed systems are properly used and
maintained. [Hereafter, this latter process shall be referred to as
"in-use enforcement"].
9.1.1.1 In-Use Enforcement. In cases where control technology is
subject to malfunction, the preferable method of in-use enforcement,
all things being equal, is to assess system performance by some sort of
in-use emissions test. EPA's Field Operations and Support Division
(FOSD) has endeavored in the past to develop an accurate and expeditious
in-use emissions test. EPA's primary effort has been the development
of the so-called "Short" Test. This test was first proposed in EPA's
November 1, 1976 notice of rulemaking proposing amendments to already-
promulgated federal Stage II regulations. The Short Test measured
vapor recovery system emissions occurring during actual vehicle
refuel ings. Vapors emitted at the nozzle-fill pipe interface were
captured by a flexible sleeve and fed into recording instrumentation.
The Short Test required two people—one to conduct the
vehicle-refueling/vapor-recollecting process and one to handle the
instrumentation. As conceived, the test required that emissions from
100 cars be measured in order to establish a violation. Experience
with the short test indicated that about 75 cars could be measured at
high throughput stations in an average eight hour workday. Thus,
roughly 2.7 person-days would be required to perform a "Short" Test.
EPA estimates that, on average, state inspection effort yields about
195 person-days (260 days per year x 75% available field time) per
year. The number of stations at which short tests could be performed
per inspector man-year would be about 72 (195 person-days /2.7 person-
days per test). Methodology presented later in this chapter estimates
9-1
-------
that Stage II programs will achieve about 56 percent efficiency in a
minimal enforcement scenario but that this figure may be raised to about
86 percent if regulated outlets are inspected annually. EPA estimates
that Strategy V.A (Stage II nationwide, with size exemptions) would
cover about 120,000 outlets, and that strategy III.A (Stage II in all ozone
nonattainment areas, with size exemptions) would cover about 36,000 outlets.
Accordingly, the person-years of effort which would be required to inspect
each outlet annually in the Strategy V.A and Strategy III.A scenarios,
using the Short Test, would be roughly 1,660 and 500 respectively.
A second approach to in-use emissions testing which has been
investigated by EPA/FOSD is the so-called REST (Refueling Emission
Simulator Tank) procedure. This procedure would, not measure emissions
during actual vehicle refuel ings. Rather, emissions would be determined
by measuring the vapors escaping when gasoline was dispensed into
portable fuel tanks equipped with a number of interchangeable fill necks
representative of fillnecks in the on-the-road vehicle population.
As in the case of the Short Test, two persons are required to
perform the REST procedure. By eliminating the need for actual vehicle
refuel ings, however, REST achieves a time advantage over the Short
Test. It is estimated that about three hours would be required to
perform the test at each station. In addition, a daily calibration of
the equipment requiring about 45 minutes would be necessary. Adding in
travel time to and from stations, the most reasonable estimate presently
is that, using REST, two service stations could be tested per day by
each team of two inspectors. This means that one team could inspect
about 390 stations per year (195 field inspection-days x 2 tests per
day). Accordingly, about 195 tests can be performed per person-year.
Thus, the resources necessary to inspect each outlet once annually in
the Strategy V.A and Strategy III.A scenario's, using the REST
procedure, would be about 600 person-years, and 200 person-years,
respectively.
9-2
-------
It appears that there is an enforcement strategy capable of
achieving given in-use inspection frequencies with substantially fewer
inspection resources than would be required with either the Short Test
or REST procedure. Examination of Tables D-2 and D-3 (Appendix D)
setting forth the frequency of occurrence of each type of Stage II
system defect in a minimal-enforcement scenario, along with the average
effect on efficiency of each type of defect, suggests a strategy. From
the tables it can be seen that most Stage II system defects are visually
observable—e.g., nozzle and hose defects for all systems, tampering
for all systems. Indeed, the only system defects which are not directly
visually detectable are misinstallations (all systems) and miscalibrated
aspirators, or jet pumps, in the case of Hybrid systems.
The visually observable defects constitute by far the dominant
portion of all system efficiency losses. For example, programs
requiring balance systems, which have been certified at 95% efficiency,
are estimated later in this chapter to be 54 percent efficient on
average in a minimal-enforcement scenario. The majority of this
efficiency loss is attributable to visually observable defects and a
high rate of noncompliance. The percentages of system efficiency
losses attributable to visually observable defects for other systems
are comparable.
That such a high proportion of system efficiency losses would be
attributable to visually observable defects and a high rate of non-
compliance suggests that periodic visual inspections of in-use control
systems should serve as the cornerstone of any Stage II enforcement
strategy. Operators of gasoline-dispensing outlets determined to be
without controls or to be using one or more defective or
tampered-with system components could be served with "notices of violation"
or some type of citation, depending on the type of enforcement
authority available, as well as on the Stage II enforcement policy
chosen in each jurisdiction. In either case, the visual inspections
would ultimately lead to correction of observed violations.
Success of a Stage II program whose enforcement strategy was
premised on visual inspection of in-use systems presupposes several
9-3
-------
factors. First, there would have to be a procedure for certifying, on
a generic basis, that permissible-to-use systems were capable, if
properly installed, maintained and handled, of achieving the recovery
efficiencies upon which the Stage II program's emission reduction
credits would be premised. Second, some form of quick check for system
deficiencies not visually detectable would need to be developed—at
least to ensure that losses from such deficiencies remained as low as
theoretical analysis appears to indicate. The vapor recovery system
certification process operated by the California Air Resources Board
can serve as a model in these respects: the process requires that
systems achieve high efficiency levels when tested under actual operating
conditions; in addition the procedure requires the sponsor of each
control system to submit an operating manual "identifying critical
operating parameters affecting system operation...[and identifying] the
operating range of these parameters associated with normal,in-compliance
operating of the control system..." Checks of these Stage II operating
parameters have been developed which are simple and expeditious enough
to serve as supplements to visual in-use system checks.
9.1.1.2 Installation Monitoring. Some of the principal system
defects not subject to visual detection are installation defects—parti-
cularly, in the case of Stage II systems, defects in the installation
of the underground piping. These deficiencies can be detected by the
parameter checks just referred to.
The checks can be performed after a system has become operational.
However, there is a substantial advantage to structuring Stage II
enforcement efforts so that performance of these checks occurs prior to
a system's becoming operational. If the test is performed before the
earth-and-concrete covering over the pipes is replaced, a service
station owner will incur substantially less expense in correcting a
misinstallation problem. Thus, during the installation phase of a
sensibly-run Stage II program, enforcement resources will be concentrated
on assuring that control systems are properly installed.
9-4
-------
9.1.2 Onboard Control Programs
The modified fill pipe program scenario could be enforced, at
probably modest incremental cost, by including the onboard vapor
recovery feature among those emission-control features already
monitored within the established EPA certification and in-use mobile
source testing programs. The regulatory standard would be a performance
standard couched in a form capable of being tested in conjunction with
the Federal Test Procedure's evaporative emissions tests.
9.2 RESOURCES REQUIRED TO PURSUE ENFORCEMENT STRATEGIES AT VARIOUS
LEVELS OF EFFORT
In the immediately preceding subsection, the general approach to
enforcement of Stage II was discussed. The present subsection sets
forth the specifics of Stage II enforcement and describes the
methodology for determining the amount of resources needed to enforce
the various gasoline marketing program options.
9.2.1 Installation Monitoring Resources
It was assumed that the resources applied to the task of monitoring
installation of systems during the phase-in of a Stage II program would
be the same as the resources applied to performing in-use inspections.
9.2.2 In-Use Inspections Resources
The resources necessary for performing in-use inspections at the
various frequencies may be determined from the following formula:
R = S x i x (TI + T? x r) x.~B
where
R
S
r
M
B
M
resources,.expressed in man-years of effort
number of outlets covered by the control option or alternative
number of times per year each regulated outlet is inspected
time necessary to perform each in-use inspection (hours)
time necessary to perform follow-up inspection at outlets
found in violation upon first inspection (hours)
percentage of outlets reinspected after initial inspection
field time available in each inspector man-year (hours)
multiplicative factor to account for supervisory overhead
9-5
-------
These individual factors were determined as follows:
S - The number of facilities covered by each control option is
shown in Table 7-18.
i - 0.5 (bi-annual enforcement)
- 1.0 (annual enforcement)
- 4.0 (quarterly enforcement)
Tl> T2 " The tl'me Pen'ods to perform each in-use inspection (T]_) and
each reinspection (T£) were developed based on field inspection
experience for each of the industry sectors except for Stage II.
Table 9-1 indicates the inspection and re-inspection times assumed
for this analysis.
The time to perform each'in-use inspection (Tj_) for Stage II
facilities was computed as the weighted average of the times to
conduct inspections at typical (9-nozzle) stations utilizing each
type of control technology. The inspection times for stations with
the various types of control were computed as follows:
Common Elements
- Conduct explosimeter checks during refueling occurring
while inspector is at station
5 minutes
- Visual checks of nozzles/hoses
15 minutes
- Time for recording observations and discussions
with station personnel
20 minutes
- Travel Time per station
Total, Common Elements
Individual Elements
Balance System Stations
- Check No Seal-No Flow features
9-6
10 minutes
50 minutes
5 minutes
-------
Table 9.1 INSPECTION AND RE-INSPECTION TIME ASSUMED FOR
ENFORCEMENT-COST ANALYSIS
Inspection Time,a
Hours
2.5
2.0
0.5C
Re-Inspection Time
Hours
1.5
1.0
0.5C
0.5
1.26
0.33
0.5
Industry Sector
Bulk Terminals'3
Bulk Plants'3
Storage Tanks
Service Stations'^
- Stage I only
- Stage II only
aBased on EPA estimates.
blncludes inspection of trucks at the loading racks.
Inspection time per tank.
dCombination of Stage I and Stage II inspections could result in a time
savings of approximately 0.17 hours per inspection. The enforcement
impacts analysis assumes a worst case in which Stage I and Stage II
inspections would be performed independently.
e Weighted average of inspection times for balance system (80% @ 1 2
hours , Hybrid systems (15% @ 1.3 hours), and vacuum assist (5% @ 1.1
hours).
9-7
-------
- Check Back Pressure-Shut Off
Features (All Nozzles)
-• Conduct Liquid Blockage Test
(20 minutes per test, 1 dispenser
in 50, at random)
Total, Indiv. Balance Syst. Elements
Aspirator-Assist System Stations
- Check Nozzle Vacuums
(3 nozzles per station at random)
- Check Back-Pressure-Shutoffs
Check Aspirator Calibrations
(Check 5 nozzles at 1 station
in 20, at random, 15 minutes
per nozzle)
Liquid Blockage Test
Came as bal. syst. stns.)
Total, Individual Aspirator
Assist Station Elements
Vacuum-Assisted System Stations
- Gross Check of Nozzle Vacuum
(1 nozzle per station, at random)
Check Gross Function of
Incinerator, Blower, Underground
Storage Tank Pressure Gauge
Total, Individual Vacuum-Assist
Station Elements
15 minutes
.4 minutes
20.4 minutes
10 minutes
15 minutes
3.75 minutes
.4 minutes
29.15 minutes
3 minutes
10 minutes
13 minutes
9-8
-------
Total Inspection Times (Common and Individual Elements)
- Balance 70.4 minutes
- Aspirator 79.15 minutes
- Vacuum-Assist 63.0 minutes
Weighted Average
Total Inspection
(Assumes 80%/15%/5% throughput coverage
distribution evidenced in South Coast
and San Diego, CA, areas)
= (70.4 x .8) + (79.15 x .15) + (63.0 x .05) = 71.5 minutes
T£ for Stage II was determined to be 25 minutes (10 minutes travel
time; 15 minutes to record observations and to discuss matters
with station personnel)
r - The percentage of stations reinspected is expressed as (a x v)
where "a" is the fraction of .initial violators reinspected and "v" is
the percentage of facilities initially determined to be in violation.
The fraction of initial violators reinspected depends on a judgement
respecting the need to ensure that initial violators, who at some point
agree to correct system violations, actually do correct the violations.
Where enforcement is done on a quarterly basis, it is judged that the
succeeding quarter's inspection can generally serve as a follow-up
inspection and, accordingly, "a" is assigned the value zero for that
enforcement scenario. Where enforcement is performed only biannually,
it is judged that a 100% follow-up rate is desirable. For annual
enforcement, it is believed that a 50% follow-up rate would be adequate.
"v", the percentage of facilities which would be determined to be
initially in violation upon an in-use inspection, has been determined
to be 15 percent for Stage I equipment based on experience in over
4,000 Stage I inspections. No Federal experience is available for
Stage II systems, therefore it was assumed that the initial violation
rate was a function of the percentage chance (probability) that an
individual installed control system will have a defect. A curve
9-9
-------
expressing the relationship between the probability that an individual
in-use control unit would.have some defect and the probability that
any given outlet would have one or more defective units--i.e., be
initially in violation—was developed by the following procedure.
First, a curve expressing the relationship between these two
probabilities, using only the mathematical laws of chance, was plotted.
(See Curve A in Figure 9-1). The mathematical probability "P" that
any station would have one or more defective units is expressed by
the following formula:
P = 1 _ (1 _ p)m
where "p" is the probability than an individual control unit will
have a defect and "m" is the number of nozzles-(estimated to be 9)
at an average station.
The mathematically determined curve would be a correct potrayal
only if individual control system defects were randomly distributed.
Presumably, however, at least some dealers would be conscientious
about maintaining control equipment, and 'thus the actual distribution
curve would deviate from the mathematical model. It was decided to
use available data to generate several data points for such an actual
distribution curve, extrapolating the remainder of the curve based on
a comparison with the mathematically-predicted curve.I/
Analysis of empirical data collected in EPA's survey of service
stations in the District of Columbia revealed that the rate of unit
defects and the rate of stations exhibiting one or more defective
units varied considerably from one type of system to another. Among
stations using one type of nozzle, the percentage of unit defects was
roughly 44%. The percentage of stations with one or more defective
units was roughly 90%. By contrast, units at stations using another
type of nozzle had only a 16% unit defect rate and "only" 50% of the
stations had one or more violations. These two points were plotted
Basing the extrapolation on the mathematically-predicted curve
assumes that randomness will be the dominant factor in determining
the distribution of defects among stations, and since it is believed
that most service station operators will not voluntarily maintain
vapor recovery equipment, this assumption seems reasonable.
9-10
-------
-_ O
03
O)
>
CJ
cu
O)
Q
(13
3
•o
s=
t—<
c
4-
O
O
0_
O)
3
CT>
-. o
.. a
+J
C
... O ,_
"^ 13
-a
T3
o
CO
C
(13
O
CM
-Q
(T3
J3
O
a_
-- o
s s ° c
O CO 10 ^
3
r
1 "
o
CM
99
9-11
-------
and from them was extrapolated a curve (Curve B, Figure 9-1) express-
ing the empirically determined probability that any one station
would have one or more defective control units as a function of the
probability that an individual unit would be defective.
The rates (probabilities) of individual unit defects, including
the weighted average of such defects for each level of enforcement
effort, were determined from the information in Tables D-8 through.
D-10 of Appendix D and are set out in Table 9-2.
Little EPA data have been compiled for violation rates at bulk
terminals and bulk plants. Results of a recent survey by EPA
Region IX of terminals in California,! where a relatively active
inspection program exists, indicated 68 percent of the terminals
inspected resulted in violations either for truck leaks or for
loading rack operations. Therefore, the same percentage of facili-
ties in violation for Stage II was assumed for the bulk terminals
and bulk plants. Table 9-3 contains violation rates assumed for
bulk terminals, bulk plants, and Stage II at service stations.
M - Experience with pollution control program enforcement at the Federal
level using contractor assistance for inspections indicates that an
inspector, on average, spends about 75% of a man-year in the field
doing inspection work. Enforcement officials in the South Coast and
San Diego areas of California indicate that their inspectors spend
75%, and almost 100%, of their time in the field, respectively.
For purposes of this study, M was determined to be 1560 hours per
man-year at both the Federal and State "level (260 days x 0".75 x 8 hours
per day).
B - South Coast and San Diego enforcement officials indicate that the
supervisor-to-inspector ratio's planned for in-use enforcement of Stage
II in their districts are 1 to 6 and 1 to 5, respectively. For this
study, the supervisory overhead factor, 'B1, was set as 1.2.
Using the formula set out at page 9-7, along with the above-
derived values of the independent variables, the inspection resources
required to enforce the various control options at the various levels
of efforts were calculated.
9-12
-------
TABLE 9-2. PROBABILITY THAT INDIVIDUAL CONTROL UNIT WILL
HAVE AT LEAST ONE DEFECT, AS A FUNCTION OF
ENFORCEMENT EFFORT
Inspection
Frequency
Quarterly
Annually
Bi -Annually
Bal
Dir.
1%
21%
37%
ance
N.O.V.
10%
30%
50%
Aspi
Dir.
5%
22%
33%
rator
N.O.V.
8%
33%
42%
Vac.
Assist
Dir. I
2%
8%
16%
J.O.V.
5%
15%
27%
Weighted
Average
Dir. N.O.
6.5% 9.
21% 29%
35% 48%
V.
5%
Using Curve B, Figure 9-2, values of "v" for the various weighted average
rates of individual unit defects (i.e., for the various levels of enforcement
effort), were then determined. The results appear in Table 9-3.
9-13
-------
Frequency of
Inspections
• Table 9-3
PERCENTAGES OF FACILITIES (STEADY-STATE
AVERAGE) WHICH WOULD BE IN VIOLATION AS
A FUNCTION OF FREQUENCY OF IN-USE INSPECTIONS
Percent Violators
(Steady-State Average)
Quarterly
Annual
Bi-annual
State
'Enforcement3/
28
65
85
Federal
Enforcement3/
34
72
90
aThe Federal figures are based on 100% N.O.V.-type enforcement-. Some
States have N.O.V.-type and some have direct enforcement mechanisms.
The State figures presuppose a 50%-50% split between direct and N.O.V.
enforcement.
9-14
-------
9.2.3 Test Observation Resources
For the analysis, it was assumed that an initial performance test
of control equipment would be required only at bulk terminals. The
level of resources necessary to observe a performance test was determined
according to the following formula:
R = S(Tj + T2 x r) x B
where
R = Test observation resources, person-years
S = Number of controlled facilities
TI = Time to observe initial test, hours
T2 = Time to observe re-test, hours
r '= Fraction of facilities requiring retest
B = Overhead factor ,
M := Field time available, man-years .
The number of facilities, the overhead factor, and the field time
available was the same as used in the previous analysis. The observation
time for both the initial test and the re-test was assumed to be 16 hours
(8 hours to observe the test and 8 hours to review the report). This
assumes that the testing for bulk terminals would be conducted in the same
manner as that required by the NSPS for bulk terminals (40 CFR 60, Subpart XX
- Standards of Performance for New Bulk Gasoline Terminals). The fraction of
facilities requiring re-test (one-half) was taken from EPA's analysis of
reporting and recordkeeping burdens performed for the NSPS.
9.2.4 Legal-Clerical Resources - - .
The level of legal resources necessary to process detected
violations of regulations imposed under the regulatory strategy was
determined according to the following formula:
R = S x i x v x (Ti + T? x c)
^
where
R = legal-clerical resources, in person-years.
S,i,,v = same as in Section 9.2.1.
9-15
-------
TI = total time, including legal, secretarial, and supervisory, to
prepare either a notice of violation or a complaint, depending on the
enforcement mechanism utilized. Based on the Federal experience
enforcing Stage I vapor recovery, TI is estimated at 3 hours. 'It is
assumed that State enforcement processes would not be substantially
more expeditious than Federal processes.
T2 s total additional time required to process a case where the
violation is not "voluntarily" corrected and the case settled as a
result of the N.O.V. or complaint. Based on the Federal experience,
the typical case is settled upon issuance of an administrative
order, a process consuming 5 hours of legal, clerical and supervisory
time. It is assumed that the "difficult" case would be handled with
the same effort at the State level.
c ~ the percentage of cases which could be expected to go beyond the
complaint or N.O.V. stage ("difficult" case). As most violations
will be relatively uncomplex and, compared to the cost of a legal
proceeding, relatively inexpensive (to correct), it is anticipated
that the rate of "difficult" cases will be small. EPA estimates a
2.% rate for quarterly enforcement, 5% for biannual enforcement, and
4% with annual enforcement.
M * the time available in a legal-clerical person-year, expressed in
hours. There is assumed to be no difference between the federal and
state figures. EPA estimates the average amount of legal-clerical time
per year which would actually be spent on processing cases at 1920
hours per-year (240 days per work year x 8 hours/day).
Using the formula set out above, this page, and those values of
the independent variables just derived, the legal-clerical resources
required to enforce the various control options and regulatory
strategies at the various levels of enforcement efforts were calculated.
9-16
-------
9.2.5 Onboard Control Inspection Resources
The onboard-control, modified-fillpipe option can be enforced
through the existing certification and in-use testing programs. It
appears that the amount of personnel and equipment needed to effectively
enforce this type of onboard control would be much less than that
required for enforcing service station controls. It was estimated that
incorporating a test of the onboard control system would add only about
1/2 man-year to the EPA certification program, and would require only
about $50,000 in equipment costs. Monitoring onboard controls would
probably not require any additional personnel. The increase in the
testing budget necessary to accommodate the monitoring of onboard
controls would be modest, running at most 1Q% of the FY79 budget of
$1.5 million—i.e., $150,000. At a rate of $30,000 per person-year
(see section 9.3), this equates to about 5 person-years.
9.3 ENFORCEMENT COSTS
In preparing budget estimates, EPA estimates the annual cost of
its "average" employee (legal, technical, secretarial-clerical all
considered) at $30,000 per year. A check of enforcement costs with
personnel in the South Coast and San Diego, California areas reveals
that this figure is probably a reasonable estimate on the state level
as well. Accordingly, the annual costs of enforcing the various Stage
II and onboard control program options can be determined by multiplying
$30,000 by the number of person-years involved. The results appear in
Tables 9-4 and 9-5. The costs of equipment, being a minor portion of
the overall cost, have not been estimated.
9.4 ENFORCEMENT COST EFFECTIVENESS ANALYSIS
A final analysis concerning enforcement costs was performed to
determine the effect of enforcement costs on the cost effectiveness of
the regulatory strategies. Table 9-6 contains a comparison of the
different levels of enforcement for the Stage II nationwide control
option and the Stage II in all nonattainment area control option.
Cost effectiveness of strategies based upon theoretical control
efficiencies and cost effectiveness based upon in-use efficiencies,
given different levels of enforcement, are presented both with and
9-17
-------
without Including enforcement costs. As indicated, enforcement costs
do not significantly effect the cost effectiveness of the strategies.
In addition, Table 9-6 indicates that annual inspections are the most cost-
effective approach for any enforcement level under the no exemption
option and that annual or quarterly inspections are roughly equivalent
under the size exemption option.
9-18
-------
+•» S. H-
S«2S
o c t*
O
O
O
00
o
o
1—0
Z CM
LU O
O IO
a: co
o CTI
U_ i-H
oo
LU
<-
u tu
v- a.
o —
I rf —( O O
to co CM CM *-H *^
O O LO O O IT)
t^* c\j ^H r** CM •—(
o o o
o o o
m co r—
§§g
o o o
00 «3- CM
o <
o <
CM <
CM <
CM
> o
I O
i tn
• o o o
o o o
!_
C
o
c
o
+J
01
c
o
•t—
a i—
00 O)
en
T- OO
> •*-*
(U
00
s
oj <0
•a x
C (U
O M
• ^ oo
z o
1
.a
o
+j
' o.
S
0)
X
UJ
o
o
9-19
-------
•si!
4-> Sl^
« U-r-
0 F— c ia e *>
s. 3
c
o
10 4J
ut a.
-------
a.
C c
0 §£
•*•* {_> i—
• U) fc. •*•
TSr O O £
tJ O **—
Qi C **
H- LLJ— ^
32 3
0 o
O
Lu
0
CO
oo
o
o
h-0
IZ CM
LU O
LU^^O S
O IO CU u
OS CO T3 ^_
O CTi 3 o S.
32*— U .2 S
LU c ":£
CO O *i i
LU 2Z 0 | §
1—4 HH V ^
i— i— £ S.
«
e
at >, o
50 o •»-
r— e +J
« l°l
r— aj «
t. c
U- i— •
e
0
i *
Q.
O
UVO» ^O O O O r-4
CM r~- m CM
o o .0 ' o o o u)
CM »H
0 O O O O 0 O
o of o o o r**
US O «3 C\l CM O r-t .
H ft ft f, fl
CM •-< 00 CO CM
S- 3 S- 3
0) r— = 0) i— C
•4-> «9 C +•> o
Q. •!-
•< £ 4-*
^K m i-i
x e
o x
4-> 0) LU ^3
U M -0
CD •"- O S-
• — CO ^. - iQ
03 O
? r t ^3
c
0 O
1
.
"3 !
c e
C 0
O f-
U> 4J
i. Q.
(U O
o.
s-
^_>
Js
2
cj 15
E 4J
0 «
M- e
s- o
0) C
+•> >>
c^
§g
0
0 1-
s- o
•a c
C V
t) S
«0
• IQ o
s- a
0 •!->
Q. CO
.
"a!
c
o
tn
i.
0)
a.
ia
s.
-------
o o
.
. OS
o o o o o o
co co CM in o in
o o o o o o
o o o
to
fc
S LU
LU I—I
O CJ
CC LU
O \- *~*
u. «c o
sz: oi CM
LU I— O
C-0 CM
LU I
>> >- VO
•—« oi co
!5gS
§<
(
SS8
I O
I O
CM *-4 CO CO CM
O t-. r» to 10 «*_
t— CM ••* CM CO LO
I 0
i o
S CD
O UJ
ID
i
cn
9-22
-------
e e
«- cu o
o e i-
- W3 4->
•-• GC. CO =
H- O O
=3 CD
O LU
Cf
LO
I
CT)
O)
<0
s.
CD <9
c£ 01
S in
cu s_
o •* CM
o o o
en CM «a-
LO CM ^H
> O O
> to co
§00 O
r^ r^. o
O O
m co
CM co in
o o
10 en
CM
o o o
CO lf> CM
oo o
«3- «3 IO
r~- CM •-<
O •-*
co o 10
o o o
000
CO CM CO
So o o o o
o o o o o
fl- O 10 -^ ui v^
f> en in •* co r<
CM r-~ CM *-t
§o o
o o
oo^r-O^
CM -H r».
§0
o
en t~
00
CM oo o
cnco CM
O-«£ eo
X
UJ
o
r- (O r— Iw
S. 3 S. 3
0) r— = 0) I— C
*» Id C *J fO C
• V. 3 CL
-------
*•>«-»
c c
U. gj O
O M *^
O
CO
O
O
SSR3
LU CO
s: LU
LU 1—1
O O
o£ LU
u. «c c?^
S: o: CM . >- to 4->
»-« OS CO C
I— O CFt O
«CI— *-< O
LO
O»
(U
o m tr>
to CM •-<
o o o
ir> 10 f
o of> oo o
^> CM «—< ^^ u) en
o o o o o o
01 oo
o o o
o o o
r- oo in
SO O
O O
en CM >->
i— id e
I
u>
e
0.
I
01
N
O.
i I
X
I UU
H-. 01
1-1 T3 M
0) ID OO
O) O
tO -O
4J C •
00 O «S
w>
e
o
O"«S I
tn
c
•« o
X
Ul
I
9-24
-------
U> i. -r-
002:
o <<-
U_
O
00
oo
O
o
LU 00
s: LU
UJ i—i
O CD
0£ UJ
O I— ^—•
Ll_ < O T3
Z C£ CM O)
LU I— O 3
oo CM s=
LU I •!-
> >- yO -M
i—< C£ CO C
I— O CT> O
< I— ^M O
O LU
LO
I
(Ti
(U
o o to
ro 0)
o Q-
c
UJ
5> 5
s-s
8"
> o o
) O O
o o o
o o o
o oo
o o o
co co r-
oo cn LO
CM
«a- m eo
po -^
us f— «
3 t. 3
C 01 i— C
r— (O *—
S- 3 S-
03 i— C 0) i—
3 «£
= I
e •!—
oa
s- = «c
S;
0)
o
. js
c
o
i en
> 4->
I OO
C
o
UJ
a.
§
x
UJ
o
9-25
-------
U.
o
CO
o
o
LU
E:
00
UJ H- I
O C£J
a: ui
o i— *— •—
LL.
tu j— o -a
CO CM =3
U4 1 i—
>. >- U3 O
»-i cai co c
H- o en o
=5 —1
O UJ
o:
01
(U
I—*
X)
«J
*,-*
c c
<4_ S 0
0 §£
in s_ t-
0 0 Z
o "-
e v»
UJ ^"*
o
tn
cj
u
3 Ul
0 t-
tn «
03 CO
o:>-
c o
I aj
o
£
C
UJ
VI
c
o1 •»-
OJ M- O
=3 O «U
CT OL
(U CA
U C
LL. ^<
_
4?
«
|t
e/>
O OO O O O
CO CO tn CM CM f>
CM to, CM i— t
'
§o o o o o
•-H CO O W CO
tO CM •"•* VO trt PO
i-H
C3 CO ^3 CO ^3 ^3
o t-. »r CM oo -H
CM m -«.-4
r— fl r- rtS
J- 3 J- 3
0) r— C 0) i— C
S- 3 «C I- 3 «5
(0 C 1 (SCI
3 C i- 3 C T-
C3* ^C CQ ^3*^C CQ
V>
S o tn
0) 4J 0
OJ O. t^-
IB H 4J
00 X g"
-(. -O X
s- a> uj
i— i fa N
Of- 0
z
e>c
jj • . °
tO + «S CQ
»
*^
X
I/I
01
O)
-------
1
oo
~^
0
!"**
o.
o
_J
o
0
1—)
t-H
LU
CD
1— ,
CD
O
3Z
^D
^2
r— 1
i
r^
O
LU
I .
I_L»
1 t
LU
LU
00
13
1
-2
0
"~"
3
M>
«J Oi
• t— i y
™ ^"""i,
UJ *•%•
i -—
.>
0
LJ *-^«
Cfl O
3 3
CM
4J 1
= to
S — CO
_ sen
V ^r-l
•» ~— '
O IE
k- =)
5 °
«n
S I
0
CU 1
N ^O
n- CO
io^2
3 s:«—
C' ^
O
cu
U)
o
*T ^*
i 3
CU CM
ce . i
C CO
o — en
££E
JL! ^J
8
LU — —
a:
*
•
sO
i
LU
_J
CO
*c
ooooo ooooo ooooo ooooo
tO~4LOCOCM i-ICOCMtOtO tOLO^HtOCO -OCOUOLOLO
CMCMCntOf^ f-ttOLOPOOO OCnLOPOCO O^"CM*-I«-H
CMOOCMCMCM i-4*-^»-I^H»-( CMCMCMCMCM »-4»-<^^»-4^^
OOOOO OOOOO OOOOO OOOOO
to**i — r^ro ~4coco^ro totOLoto^* o co CM »— t «•
CMCMCOLO^ «-HtO«9*fOCM OCn«CTCM»-* O"*CM*^O
CM CO CM CM CM •—! »~* •••* •"* *** CM CM CM CM CM i-< *H r- 1 «— < *— (
1 IOOO 1 IOOO IIOOO 1 IOOO
I iroLOCM I ir^r^o I itoLOLO i ILOCO*-H
1A 00 CM ** CM r*. -H CM to CM
CM
ooooo ,00000 ooooo ooooo
O LO o O LO OOOOO O O LO O LO O LO O LO O
^» to to CM o co to cn r*-» to co cn en cn co c? r^> *— < o o
cOLOcnencn tointovoto IO^LOLOLO CM*^CMCMCM
»H ^H i-l 1-1 — 4
§0000 ooooo ooooo ooooo
oooo ooooo o.oooo ooooo
CS C5 C3 C3 C3 CS C3 C3 C5 C3 C3 C? C3 C3 C3 C3 C3 C? C? C?
ooooo ooooo ooooo ooooo
r-r-encoco r~mr»cor» ^Htoco>-Hro OCOCMLOCO
cMcor^^-co vocoto^Hco coto*ertor-* o»-nr^.cocy>
CO <^to"r~l~ 00 CO ^- LO LO CM •-! CM CM CM CM .-1 ~4 <-H -H
, *c c r— >, z >— >,zc: i— >>
zinr— 10 »— ^ o i— «F- «i— ioi— o>— U (OCi— CU •— O O lOCf—CU i— •!-> O IO C r- CU
Z T- i- EC(O*J Z CLt- E C OJ *J < i- •»- EC «C CUT- EC(O+J
+J +J •?<3l. S +•> -^ ••-> »-<3i. E •>-> t- <: 3 S-
1 CL CUCUCtCrO 1 CU CUCUCICIO 1 Q. CUCUCIC(d 1 CU CUCUCICIO
E u S.^ t- e s x s-ini-^-e:3 E i-in^-^-cs x s-in-^--r- = 3
MXCUI "•• CU 1 "-IXCUI •— CUI
uj.cc: cufc LU jz c Q) j= c
CU 1— i— O O O O CU N 1— i-HOOOO CU t— >-" O O O O - CO M t— « O O O O
O1 O O^ •*" Ol O O} «r"
IQZII raooii (ozii 10 (/) i. i
OO 00 OO 00
trt
c
o
^
u
3
^3
cu
ec
c
o
in ,
in
. I
in U
1 t
•*•> in
0 +»
3 in
•o o
cu o
06 *,
c c
O 0)
•t— E
in u
€ O
LU *4—
e
O LU
O
:» +
in in
in in
O O
o o
•0 XI
CO CO
•^ t—
« 10
3 3
c c:
c c
ie J=
9-27
-------
9.5 REFERENCES
1. Presentation by Gary Uvagnino, EPA Region IX, at EPA Technical
Workshop "Control of Volatile Organic Compounds from Gasoline
Storage and Transfer Facilities," Fresno, California. December
8-9, 1983.
9-28
-------
APPENDIX A
HISTORY/BACKGROUND
A-l
-------
-------
APPENDIX A - HISTORY/BACKGROUND
Date
Late 1973
6/11/73
Early 1974
3/8/74
1974-75
10/9/75
11/75
1976
11/1/76
1976-77
5/13/77
6/8/77
8/77
Activity
EPA approves San Diego and S.F. Bay Areas Stage II SIP
Regulations.
EPA proposes NSPS regulations for Petroleum Liquid Storage
Vessels. Published in Federal Register 38 FR 15406.
EPA promulgates Stage II and other gasoline marketing
regulations for all or part of 13 AQCR's.
NSPS for Petroleum Liquid Storage Vessels promulgated.
Published in Federal Register 39 FR 9308.
San Diego and S.F. Bay Area implements first generation
Stage II technology. • •
EPA proposes amendments to the Stage II regulations.
Published in Federal Register 40 FR 47668.
EPA issues Criteria for Stage I Vapor Control Systems
at Gasoline Service Stations.
California implements Stage II programs in all California
ozone impacted areas.
EPA proposes additional amendments to the Stage II regulations
as a result of comments received on the October 9, proposed
amendments. Published in Federal Register 41 FR 48043.
D.C. implements first generation Stage II technology.
Final compliance deadlines for EPA-promulgated Stage II.
Regulations indefinitely deferred. Published in Federal
Regi ster 42 FR 27674.
Benzene listed as Hazardous Air Pollutant under Section 112.
Published in Federal Register 42 FR 29332.
Clean Air Act Amendments of 1977
Section 202(a)(5) - Requires promulgation of fill pipe
standards for new cars if Stage II required by EPA.
Section 202(a)(6) - Requires EPA study of Stage II versus
Onboard Technology.
Section 324 - Precludes lessee from paying cost of procurement
and installation of Stage II.
A-3
-------
Date
8/77 (cont.)
10/77
12/77
8/77-1/79
6/78
8/78
10/78
12/78
4/4/80
12/80
Activity
Section 325 - Limits applicability of any EPA Stage II
.regulations and provides for phase-in schedule for small
independents.
.Control Techniques Guideline issued for Control of Hydro-
carbons1 from Tank Truck Gasoline Loading Terminals.
EPA Publication No. EPA-450/2-77-026.
Control Techniques Guideline issued for Control of Volatile
Organic Emissions from Storage of Petroleum Liquids in
Fixed-Roof Tanks. EPA Publication No. EPA-450/2-77-036.
Control Techniques Guideline issued for Control of Volatile
Organic Emissions from Bulk Gasoline Plants. EPA Publication
No. EPA-450/2-77-035.
EPA reviewed feasibility/desirability of Onboard versus
Stage II (no action taken or assessment document released).
Standard Support Environmental Impact Statement for Control
of Benzene from the GasoTine Marketing Industry (Draft Report).
Final report never completed.
Comments received during an open NAPTAC meeting, from industry,
environmental groups and other interested parties on a
draft 112 regulation and the 6/78 support document for gasoline
bulk terminals and plants and storage tank loading at service
stations (Stage I). Regulation or document never proposed.
American Petroleum Institute (API) releases study
demonstrating the feasibility of an Onboard (refueling)
control system. API Publication No. 4306.
Control Techniques Guideline issued for Control of Volatile
Organic Compound Leaks from Gasoline Tank Trucks and Vapor
Collection Systems. EPA Publication No. EPA 450/2-78-051.
NSPS for Petroleum Liquid Storage Vessels promulgated.
Published in Federal Register 45 FR 23374.
EPA proposes NSPS regulations for Bulk Terminals. Published
in Federal Register 45 FR 83126. Also publishes Background
Information tor Proposed Standards Bulk Gasoline Terminals.
EPA Publication No. EPA-450/3-80-038a.
EPA proposes NESHAP regulations for Benzene Storage.
Publishes Background Information for Proposed Standards -
Benzene Emissions from Benzene Storage Tanks.
EPA Publication No. EPA-450/3-80-034a.
A-4
-------
Date
4/13/81
1982
12/82
5/83
7/14/83
8/83
1/84
6/84
7/25/84
Activity
Reflecting the Vice-President's announcement of a reduction
o-f EPA's regulatory burden on the Auto Industry, Federal
Register 46 FR 21628 states that Onboard controls~wiTl"
not be required.
Six States committed in
Stage II regulations.
their 1982 03 SIP's to consider
STAPPA and ALAPCO passed resolution to request EPA to
review Stage II and publish CTG.
EOF and NRDC filed notice of intent to file citizens suit
to compel EPA to enact NESHAP to control benzene from
gasoline marketing and other sources.
EPA starts assessment of control alternatives for the
control of volatile organic compound (VOC), benzene,
ethylene dibromide (EDB), and ethylene dichloride (EDO
emissions and risk from the gasoline marketing industry.
EDF and NRDC file suit.
NSPS for Bulk Terminals promulgated. Published in Federal
Register 48 FR 37578. EPA publishes Background Information
for Promulgated Standards - Bulk Gasoline Terminals. EPA
Publication No. EPA-450/3-80-038b.
Control Techniques Guideline Document for Control of Volatile
Organic Compound Emissions from Volatile Organic Liquid
Storage in Floating and Fixed-Roof Tanks (Draft) released
for public comment.
EPA starts preparing support document entitled - Evaluation
of Air Pollution Control Alternatives for the Gasoline
Marketing Industry.
EPA releases draft staff paper on "Estimation of Public
Health Risk From Exposure To Gasoline Vapors Via The
Gasoline Marketing System."
Science Advisory Board reviews June 1984 EPA staff report.
A-5
-------
-------
APPENDIX B
BASELINE EMISSIONS ANALYSIS
B-l
-------
-------
APPENDIX B
BASELINE EMISSIONS ANALYSIS
The purpose of establishing an emission baseline is to be able to
estimate the impacts of reducing emissions from this baseline through
the implementation of additional control measures. The baseline
emissions must take into account the level of control already in place
in the base year to get an accurate assessment of the impacts of the
control alternatives. The base year for the gasoline marketing source
category was selected as 1982. This year was selected because this was
the final implementation date for many State regulations concerning gasoline
marketing sources and because the latest data available on facilities
and gasoline consumption at the time of the development of this
document was representative of 1982.
The general approach for establishing the emission baseline was
basically the same for each sector of the industry. Data were obtained
on the level of control already used by the States and emission factors
were selected to represent this level of control. Uncontrolled areas
were defined and emission factors were selected to represent the type
of loading or type of operations in those areas. Emissions were
calculated by multiplying the emission factors by the corresponding
throughput for the controlled and uncontrolled areas. Nationwide
throughput, on a county by county basis, was obtained from EPA's
National Emissions Data Base (NEDS). Although these data were dated
1980, the total throughput in the NEDS system was the same as our 1982
reported total. Therefore, this data was considered representative of
our 1982 base year.
B.I Control Techniques Guidelines
Control Techniques Guideline (CTG) documents have been prepared by
EPA for every sector of the gasoline marketing industry (with the
exception of automobile refueling). The purpose of these documents is
to outline what EPA defines as reasonably available control technology
(RACT) for existing sources. Table B-l summarizes the CTG recommended
limits for each of the industry sectors. Some of the recommendations
are in the form of emission limits and others are in the form of
recommended control equipment to be installed.
B-3
-------
Table B.I. SUMMARY OF CT6 RECOMMENDATIONS
FOR GASOLINE MARKETING SOURCES
Facility
Bulk Terminals
Fixed-Roof
Storage Tanks
(40,000 Gal. Capacity)
Bulk Plants
Tank Trucks and
Vapor Collection
Systems
Limit truck loading emissions to
80 Mg VOC/liter from vapor processor
Retrofit with internal floating roof,
Cover or seal floating roof vents,
and visual inspection requirements.
Equipment and work practice specifications
for submerged fill, balance system loading,
and storage tank pressure relief settings
Pass an annual
leak-tight test
which requires
having <3" H20
pressure change
when under
18" H20 pressure
or 6" H20 vacuum
No leaks
greater than
100 percent
L.E.L.9 when
monitored at
any time with
a portable com-
bustible gas
analyzer.
Vapor collec-
tion system
Back pressure
not to exceed
18" H20 when
measured at
the truck.
Service Stations
(Underground Tanks)
Design criteria for drop tube specifications,
vapor recovery requirements, tank truck inspections,
vent line restrictions, etc.
a lower exposure limit
B-4
-------
States with nonattainment areas for ozone are required to adopt
regulations consistent with these CT6 recommendations to provide for
attainment of the ambient standards.
To accurately determine how the States implemented these regulations,
a recent summary report1 and contacts with States were used. Table B-2
lists the States which had implemented requirements for bulk terminals
during the base year of 1982. The States listed in the first column
require that all terminals within these boundaries achieve a level of
control consistent with that of the CTG recommendation (80 mg/liter).
The second column includes States which require controls consistent
with the CTG only for areas within the States which do not meet the
ambient standard for ozone (nonattainment areas). The third column
includes States which do not have any emission control regulations
pertaining to gasoline terminals. A similar table was assembled for
bulk plants and service stations.
In determining baseline regulatory coverage for the tank truck CTG
equivalent controls, two cases were considered: trucks in "normal"
service and trucks in "collection" service (i.e. truck tanks equipped
with vapor collection equipment). Normal service pertains to
areas where no controls are required at the terminal or bulk plant
loading racks. "Collection" service pertains to loading when vapor
balance systems are employed. For normal service, there are no collec-
tion systems, therefore there can be no leakage of vapors from the
vapor recovery system or sealed truck tanks. In these cases, the tank
truck baseline emissions are included in the bulk terminal emission
estimates. For areas where collection systems are used, the CTG
recommendations are to have vapor tight tank trucks. Table B-5
indicates areas where collection systems are required at terminals
along with vapor tightness requirements for tank trucks. All areas
which require bulk plant vapor collection systems also require tank
truck controls.
Baseline emission levels were calculated for fixed-roof storage
tanks and external floating-roof storage tanks for both breathing
B-5
-------
TABLE B-2. STATE REGULATORY COVERAGE FOR BULK GASOLINE TERMINALS
Entire State
Consistent with
CTG Controls9
CTG Controls3
Nonattai nment
Areas Only
No Control
Regulations'1
Alabama
California
Connecticut
District of Columbia
Georgia
Illinois
Louisiana
Massachusetts
Michigan
New Hampshire
New Jersey
Pennsylvania
Rhode Island
South Carolina
Tennessee
Wisconsin
Arkansas
Col orado
Del aware
Fl ori da
Indiana
Kansas
Kentucky
Maine
Maryland
Mi ssouri
Nevada5
- New Mexico
New York
North Carolina5
Ohio
Okl ahoma5
Oregon
Texas
Utah
Virginia
Vermont
Washington
West Virginia
Alaska
Arizona
Hawai i
Idaho
Iowa
Mi nnesota
Mississippi
Montana
Nebraska
North Dakota0
South Dakota
Wyomi ng
-
aCTG Controls = 80 mg/liter standard or lower, tank truck vapor-tight
controls.
Portion of State not covered by CTG controls is covered by submerged
fill requirements.
CNorth Dakota has no nonattainment areas for ozone but entire State
covered by submerged fill regulations.
Approximately 90 percent of total throughput is loaded by submerged
fill.
B-6
-------
(storage) losses and working losses. Most States regulate emissions
from storage tanks in their State Implementation Plans (SIPs) with CTG
recommended controls.2 Based upon the historic pattern shown in a
previous storage tank survey,3 it was assumed that the storage tanks
located in nonattainament States or nonattainment areas of States would
be controlled by external floating roofs. This survey also indicated
that the majority of floating roofs had a mechanical type primary seal.
This survey further estimated that 10 percent of the storage tanks
would be new tanks subject to New Source Performance Standards (NSPS)
and would be controlled by external floating roofs with primary and
secondary seals. The storage tanks located in attainment States or
attainment areas of States were assumed to be 10 percent new tanks
controlled by external floating roofs with primary and secondary seals;
63 percent would be external floating-roof tanks with primary metallic
shoe seals and 27 percent would be fixed-roof tanks.4
B.2 CALCULATION OF BASELINE EMISSION LEVEL
Once the extent of regulatory coverage was established, the
methodology used was to determine the base year gasoline throughput for
each of the gasoline marketing facility operations in the regulated and
nonregulated areas. An emission factor corresponding to the regulatory
coverage, loading method, type of storage used, etc., was then selected
and emissions were calculated by multiplying the corresponding throughput
by the corresponding emission factor. Table B-6 summarizes the baseline
emissions calculated for the gasoline marketing industry in base year
1982. The following sections describe the methodology for each of the
industry sectors.
B.2.1 Bulk Terminals
Given the regulatory coverage in Table B-2, entire States or areas
within States were divided into controlled and uncontrolled areas on a
county basis. Gasoline throughput representative of 1982 consumption
was obtained on a nationwide, countyby-county basis, from the EPA area
source computer file (National Emissions Data System).
B-7
-------
Table B-3. STATE REGULATORY COVERAGE FOR BULK PLANTS
Entire State
Consistent with
CTG Controls9
CTG Controls3 in
NonAttai nment
Areas Only
No Control
Regulations0
Alabama
Connecticut
Illinois
Louisiana
Massachusetts
Michigan
New Jersey
Pennsylvania
Rhode Island
South Carolina
Tennessee
Wisconsin
California0
Col orado
Delaware
Indiana
Kentucky
Mary! and"
Missouri
Nevada
New York
North Carolina
Ohio
Oregon
• Texas
Utahb
Virginia
Washington
Alaska
Arizona
Arkansas0
Florida0
Georgia
Hawaii
Idaho
Iowa
Kansas
Maine
Minnesota
Mississippi
Montana
Nebraska
New Hampshire
New Mexico
North Dakota
Oklahoma
South Dakota
Vermont
West Virginia
Wyomi ng
aCTG recommendations include submerged fill and pressure relief setting
for storage tanks, and balance system loading for the loading racks.
bSubmerged fill required at loading racks in some portion of State where
no CTG equivalent controls required.
cTypical facilities were assumed to have 25 percent splash fill and
75 percent submerged fill at loading racks unless otherwise specified.
dCTG equivalent controls in nonattainment areas on storage tanks only.
B-8
-------
Table B-4. STATE REGULATORY COVERAGE FOR SERVICE STATIONS
(UNDERGROUND TANK FILLING)
Entire State
Consistent with
CTG Controls3
•CTG Controls3 in
NonAttai nment
Areas Only
No Control
Regulations0
Alabama
California
Connecticut
District of Columbia
Illinois
Louisiana
Massachusetts
Minnesota
New Jersey
Pennsylvania
Rhode Island
South Carolina
Tennessee
Wisconsin
Colorado
Delaware
Florida
Indiana
Kentucky
Maine
Maryland
Mi ssouri
Nevadab
New Mexico
New York
North Carolina13
Ohio
Oregon
Texas
Utah
Virginia
Washington
Alaska
Arizona
Arkansas
Georgi a
Hawaii
Idaho
Iowa
Kansas
Minnesota
Mississippi
Montana
Nebraska
New Hampshire
North Dakota
Oklahoma13
South Dakota
Vermont
West Virginia
Wyomi ng
aCTG Controls specifications include submerged fill of storage tanks, vapor
balance between truck and tank, and a leak free truck and vapor transfer
system.
bportion of State not covered by CTG controls is covered by submerged
fill requirements. 3
typical facilities were assumed to have 50 percent splash and 50 percent
submerged filling of storage tanks.
B-9
-------
Table B-5. SUMMARY OF REGULATORY COVERAGE
FOR VAPOR-TIGHT TANK TRUCK
REQUIREMENTS AT TERMINALS
USED IN BASELINE ANALYSIS
All Areas Where Vapor
Collection Systems Required
Consistent with CTG
Recommendations
No CTG
Recommended
Controls in
Vapor Collection
Areas
Arkansas
Cal i form' a
Colorado
Del aware
District of Columbia
Florida
Georgi a
Illinois.
indi ana
Loui sisana
Maryland^
Michigan
Mi ssouri
Nevada
New Hampshire
New Jersey
New Mexico
North Carolina,
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
Texas
Utah
Washington
Wi sconsi n
Al abama
Connecticut
Kansas
Maine
Massachusetts
New York-
South Carolina
Tennessee
V,ermoat
Vi rgi ni a
West Virginia
B-1.0
-------
Emission factors were then selected to represent the different
levels of control already installed at terminal loading racks. For
areas where vapor recovery equipment was installed, three emission
factors were used. The CTG recommended limit for vapor recovery
controls was 80 milligrams of hydrocarbons per liter of gasoline loaded
(mg/liter). However, a recent EPA study indicated that approximately
60 percent of the systems installed to meet these requirements were
operating at or below an emission rate of 35 mg/liter.5 Therefore,
unless otherwise specified, an emission factor of 35 mg/liter was used
for 60 percent of the throughput in areas where terminal loading rack
controls were required and 80 mg/liter was used for 40 percent of the
throughput. Some States, however, specified a 90 percent reduction
regulation rather than the 80 mg/liter CTG recommended emission limit.
In these areas, an emission factor of 96 mg/liter was used. This
represented a 90 percent reduction in EPA's AP-42 emission factor for
loading trucks in collection service.6
Emission factors associated with tank truck leakage were also used
to estimate emissions in areas where vapor recovery was installed. EPA
estimates indicated that tank truck leakage averages about 30 percent in
areas where there are no vapor tightness testing and maintenance
requirements and can be reduced to about 10 percent with vapor tight-
ness regulations.8 An emission factor of 288 mg/liter was used to
represent areas without vapor tightness requirements (30 percent of the
960 mg/liter factor for truck loading in "balance" service) and 96
mg/liter to represent areas with vapor tightness regulations (10 per-
cent of 960 mg/liter).
For areas with no vapor recovery regulations, two emission factors
were used: 600 mg/liter for submerged fill and 1440 mg/liter for splash
fill.9 Some States required submerged fill in all areas not controlled
by vapor recovery. In these cases the 600 mg/liter factor was associated
with the corresponding area throughput. If there were no specific
requirements for submerged fill, it was assumed that 90 percent of the
B-ll
-------
Table B-6. SUMMARY OF BASELINE EMISSIONS
FOR GASOLINE MARKETING FACILITIES
FOR BASE YEAR 1982
Emissions, Mg/yr
Facility
Bulk Terminals
Storage Tanks
Bulk Plants
Service Stations
- Underground tanks
- Automobile refueling
Total
VOC or
Gasoline
Vapors
140,000
56,000
208,000
222,000
407,000
1,033,000
Benzene
840
340
1,250
1,330
2,690
6,450
EDBa
3
1
5
5
10
24
EDCa
30
15
50
50
100-
245
a
Applies to leaded gasoline only. Leaded gasoline consisted of approximately
48 percent of total consumption in 1982.7
B-12
-------
loading took place with submerged fill and 10 percent was loaded with
splash fill.10
Table B-7 contains the emission factors and throughputs used to
calculate the baseline emissions for bulk terminals. These calcula-
tions result in the baseline value for gasoline vapor emissions.
Reference in this document to volatile organic compound (VOC) emissions
(photochemically reactive hydrocarbon emissions) are considered
equivalent to the gasoline vapor emission from terminals since the
amount of non-photochemically reactive compounds (methane, ethane) in
the emissions are very small. Factors representing the fraction of
benzene, EDB, and EDC in the gasoline vapor emissions were used. These
factors were calculated based upon liquid temperatures and vapor pres-
sures of the compounds. For determining the emissions of these
compounds, the following factors (calculated at approximately 60°F),
multiplied by the gasoline vapor emissions, were used: benzene -
0.0060, EDB - 0.000046, EDC 0.00047. Since EDB and EDC appear only in
leaded gasoline, these factors were applied only to the emissions
associated with the loading of leaded gasoline.
B.2.2 Storage Tanks
As stated earlier in Section B.I, it was assumed that 10 percent
of all tanks were considered to be new tanks and would have both primary
and secondary seals installed on floating roofs. The remaining 90
percent of the tanks in the attainment areas were assumed to be 70
percent floating-roof tanks with primary seals only (63 percent of all
tanks in attainment areas) and 30 percent fixed-roof tanks (27 percent.
of all tanks in attainment areas). In nonattainment areas, if there
was a State regulation specific to grimary and secondary seals, then it
was assumed that all tanks required primary and secondary seals.
Conversely, if no applicable State regulation could be found, then it
was assumed that only the 10 percent of tanks considered to be new
tanks would have both primary and secondary seals installed on
floating roofs. The remaining 90 percent were assumed to have
primary seals only.
Emission factors were calculated using the latest information from
EPA and the latest American Petroleum Institute equations. The major
B-13
-------
£
re
ro
(U
0
_
^
1—4 m
ro 10 CM CM 10 in ro
ro CM 10 CM 1-4
«a- en 10
i i-i co cn
1 «S- — 1 CM
t •
CM CO 10 10
i in cn co <0
1 (O CM CM
ro
E
CO
CO
CO
L CM
I CO
cn
••» o;
-i «:
UJ LU
i
cn
.a
ra
in
j
10
o,
c
a
to
o
_J
,
c
0
o
cu
to
CO
tn o
-^ ro
CO 10
CM 10
CM
C
V,
•f-
X
i— 4
o
CM C
to — *
t-*
10 r- * o
O^ ("•» O
cn 10 ch
r-T oT i—T
»-i »-4 ro
CM in o,
i^» *^
. — to
~~ •!-> O.
Is 5
J2 "re a
10 10
E «o c:
ro ^^ O
ja w, N
to to •r-
*£.*£.*£.
CO r-4 >H
in ro
IO 10 CTt
ro 10 CM
o ^r CM
CM —<
in en
r-4 ro
i-4
r"-. ro en i — »-4
^< ro co in CM
CM i-l
•"•
r-t en ro co ro «a- CM
*-H o 10 in o »-H CM
CO —4
r— «f OO CMCO ro t-~O 10
oco CMCM co«a- cn cncM r^
r*^co O T— * rocM I-H cor*- I-H
Oin m«s- in--4 «-4 ror^ co
r*-*a- CMCO »^r^ r-4 CMin CM
r*» in men r-.r*- cn ^fio o
ro co CM CM CM «3- 1-1
ro
+J r—
^s ^ - - (Q _/-\
at Q in O E
(O t= 4-> U- 3
r— IOC 103 SOi—
3 t- 10 -O O
-------
0>
en
to
to
J
o
?
C
to
uj_js_| cn-HcocMOf «-r^ io
S £ -J> 2 'J=!lor~0 »a-in CM
f o w " «^co-nin«d- loco ^j-
oo i— E 1 P- co —4 CM in" co" CM" CM"
o
cn
€
CO
CO
CM
ii i in *$•
II 1 LO O^
II 1 CM ^O
•r— 1
O
i
10
CTl
(-- tn i i 10 in i
O l*» 1 1 U) CM 1
Ol O 1 1 1^ CM 1
CO <-H
^
cn
o
to
00
UJ CM
CO
I 00
: <
: ca
-a
a>
O
O
i
ca
a
.a
u
10
a:
o>
c
•5
o
_J
i
E
t^ 0 0
te
||
h- o in
CJ C— 3 PO
a i.
Q- CO IO ^*
in CM 10 CM in cn o 10 co
-HO co co co vo o cn co
cnin P^ co cn «3- cn co CM
-H^J- VO ~H IO CM COCO CO
co cn o •— 4 i— 4 p>* P^ CM in
-a~~- — *j ~- ^ T
i- O "O £ •*-» 5 o -— .
tO4-» 2 c S4-> j^4^>4->
3oj-i-><->c:u> i-oiE
O— to 4J -t-> tO' — • C
i- — to -U r- JJ
CO *-~ to (/> O 4-*
to to •— ~ «^- to (O
•o — .^- o c ~
•7- cn .^. o c: to
o o s to ^ -5 s
i — W tO +•>
cn tjj - _ - ^ flj
3' — i- • — .
O (O
O >> CD
.21
in o
tO 3
C/l 4->
c c
to Ol
_
B-15
-------
1
10
K-S:
0
en
B
CO
CO
CM
u
i
to
en
O co CM co co csj r*. .-4 f** ***
ui «• "i CM i? rC co ^1 S S
-^.co °» oiroocno oS
-H — < OJCMtnCVJ^J- -H"_"
.
8 SIS i i
CM
•r* •-* co i i en r^ i i
**• co --« t i ou o i i
1*- UO LO 1 1 ^-t CO 1 |
"
•o o
0.?
oo
vo
O
m o «r
vo in o
co —-a-
S£
CO
UJ O\
CO
gi
A£
^
ec
T3
01
C
cc
i1
C3 C VO
i— I— o en
« oo
J§
=110
C CM
o ro
o — <
10
Sen.
a =
>— o <
o o <
CQ
(U
IQ
2^
<=Zm
^ o m
o on
*J*wi
=3 t-
QL (U
JS +j
cn*»-
!o
J=O
1— O
C\J
CO
in
in
CM
01
in
r--
in
1—1
CO
r*.
ft
in
r-t
^~
r*.
CO
«T
CT»
in
to
CO
ff
co r*-
CSJ 10
en *f co
<— • o ^r
^r vo en
co r^ «"
r*» ^ cr*
r*. co «-i
^- ^ co
o^
i-H
in
co"
en
in
•-H
vo «s-
m o
•^ *r
O CO
CO TT
p-« r-
CM
•= en
-O Q «-~
•S 3
S
VI U i- U
O -*-» < +3
^ Q) O
•0 C —
JS 3
>> c
r- i.
i "5
O Ol in in 3
re •»—
-------
!s*
et
1 — o
CO
CM
in
in
en"
OO UJ
«s >-
CD
-a
jQ
(O
E
cs c
— >- O 10
IB o o en
C3 C
i— o o
u o co
O .
i- en
i— e> in
o o co
o
vo
OCMCM
O-l
«a-
CM
CM
I~.
CM
3 i-
CL V
f 4-1
O
i. 0
-C O
m o -H
10 in -)
O -^
o en
CMfO f-H
in rt co
^i- ^- in
in
en
inioro
voooen
CM en in
o
no
in
co
ir>
eo
en
«a-
cnia
cSS
O rt ^v
U
4->
IB -,->-> -^ ^c
*- -4-> trt r— c_>
>> O IB «i- o. O OJ
s ^^ ^° = ^^
i- X i.
OJ CO O
^3 S >• JZ
O
g
s
ai
IB —
4->
O
^
O
-C
o
B-17
-------
\-
—I CM
I
to
II
:§
O)
o
CO
cu
:S
i—I i-* If) VO CO <-H LD
P-» O GO
f-4 f- i—t (O CM
— 2 E ~
-------
1
!
!
i.
en
z
0
C/1 •— *
i— CM
E CO
uj en
UJ
_J UJ
CO 21
_J 1
3
i—
O
O
1
ca
0)
o
1—
Ul _J i-
s •?
1— O CT1
(/I f— E
U
1 ^
Q- CM
O>
_J
1 ° ~
- t
~ vo
h~
en
M
(Q O
CO ^H
O
) .a o
TS
cc
7)
3 „£*
1— f— O LO
ITS o o en
1
S
11
ts c
H- oa
O
CM cn ^H co o o
<& *r or-* m co
1^3 CM ,-t —* CM CO
ir> ro CM *f m co
r"» CM to co
U *~^
•— > 3:
CO
*
r-
I-H
0
CO
t— 1
S
CO
in co
ro CM
en ro
o m
in to
CM
•"7
CJ
c 40
40
VI
H ^O
co.ul
CO CM
CO tO
en cn
—
40
c
Wisconsin
Bn
;i
CMJ!
g
1
!
1
1
1
1
t
1
!
O
cn
CM
i
cn
CO
CO
40
4J
40
cn
c:
5
en «r
t-H CO
o
'o
o
o
en
I~
CO
0)
fl
c
OJ
ct
o
c
- - j | |
cn
c
<*-
V) • JZ
•^ 40 c/>
ui 0) <^
aj <*- a.
o a -c
40 3
S 2^3
C
0 0
*— o -~
r— OJ Wl
•^ t_ in
t_ s
at a.
!gTB
V) CJ C
L. en
C CJ —
3 VI
JZ 4O -t—
3 40 40
O 10 3
t_ a.
JZ CJ JZ
40 ui 31
O 3
CU JZ O
C 40 U.
C 4^
O -f-
en c
D 40
^— o c
(O C_ 1)
jo u O
o a. t-
40 0)
o a.
"o CT o
en 31
•T3 O C
C C
a) -a T-
a> ID 5
ei. s t-
U u Q}
t— CJ 1 —
en
c c.
•~ 0
t- a.
3 >
cr
CU *^
cn u u
•»- 3
V) U
CJ rtj 4O
•^ L. j>4
> fl C
t. (T3
0) C 40
JZ CJ
CU 40 U
(J -f- CJ
= 2JZ
2 d) *
(Q U cn
> CU
C U i-
•f- CU «J
J2 40
o
0)
o
i-~ C
0) O>
-r-
t. en
0.40
vi a.
JZ
"a» 3*
40 O
cn (_
•f~ JZ
r— 4O
t. JZ
cn -a
3 .-a
' a.
en 3
O JZ
w en
JZ 3
4O O
3 >*,
0 40
O C
3
OJ O
t- O
a>
§CJ
JZ
40 =
V) 3
•r- VI
>>JZ
C
3 Ul
CJ
a> oi
t- 40
o ja
3 .<-
a.
•i,^
3 Q.
O JZ
t. en
JZ 3
40 O
ai jz
J= 40
4O
cn 40
C 40
ai
a. 40
•3
fj
40
OJ
JZ
o
-3
C
5
CJ
cs
t—
0
CJ
JZ
4O
OJ
O JZ
40 40
40 -a
u c
CJ TO
T-J
J3 40
3 3
cn a.
v» cn
a> 3
•»- o
40 C_
C JZ
3 40
O
•— C
^- ZJ
£-5
4-> =
3 3
a. cn
JZ
en cu
3 JZ
O 40
JZ VI
4O C
0 ••-
4O rt5 •
40 CJ
! = S
V) tO
V) -r-
C 40 JZ
. CU 3 40
• v) a.
u "en^
a. 3
a> o c-
t. t_ ai
= 40 C
- O) ^
CJ 40 S
40 rQ QJ
ai 40 t-
O)
40 'ra 40
C 4O
§o t_
40 O
JJ *«-
= JZ C
o >
40 • en
?O 40
C C 40
•p- S 3.
tn O) JZ
a) t_ 31
-0-^3
3 O
H3^
I t_ 40
a>
B-19
-------
emissions from a fixed-roof tank are breathing and working losses.
Breathing and working losses can be estimated from emission equations
developed for.EPA Publication No. AP-42.H The equations used in
estimating emission rates from fixed-roof tanks are:
1.02 x ID"5 M,
0.68 D1.73 H0.51 T0.5 F
14.7 - P
-w
1.09 x 10~8 MvPVNKnKc
(2-1)
(2-2)
where,
LB = breathing loss per tank (Mg/yr)
Lyj = working loss per tank (Mg/yr)
Mv = molecular weight of product vapor (Ib/lb mole)
P = true vapor pressure of product (psia) •
D = tank diameter (ft)
o
H = average vapor space height (ft); use an assumed value
of one-half the tank height
T = average diurnal temperature change in °F; assume 20°F as
typical value
Fp s paint factor (dimension!ess)
C = tank diameter factor (dimension!ess )
for diameter equal to or > 30 feet, C = 1
Kc s product factor (dimensionless) =1.0 for volatile organic
liquids (VOL)
V = tank capacity (bbl)
N = number of turnovers per year (dimensionless)
Several assumptions were made in order to calculate emission factors
on a per tank basis, for both breathing and working losses, from a typical
fixed-roof tank storing gasoline. The following assumptions were used:
Mv = 66 Ib/lb mole (for gasoline)
P =5.2 psia (for gasoline)
D = 50 feet
H = 48/2 = 24 feet
T = 20°F
Fp = 1.0 (for a white tank)
C = 1.0
B-20
-------
Kc = 1.0 (for VOL)
Therefore, after substituting into equation 2-1,
LR = 1.02 x 10~5 (66) / 5.2
I :
I 14.7 - 5.2
=8.8 Mg/yr
0.68 (50)1-73 (24)°-51 (20)°-5(
Using equation 2-2 and these additional assumptions:
N = 13 turnovers per year
V = 16,750 bbl tank capacity
Lw = 1.09 x lO'8 (66) ( 5.2) (16,750) (13) (1) (1)
- 34.2 Mg/yr
In summary, the VOC emission factors for a typical fixed-roof tank storing
gasoline are 8.8 Mg/yr from breathing losses and 34.2 Mg/yr from working
losses.
Standing-storage losses and withdrawal losses are the major sources
of emissions from external floating-roof storage tanks. From the
equations presented below it is possible to estimate both the withdrawal
loss and the standing-storage or roof seal loss from an external floating-
roof tank. These equations are taken from AP-42.
SE
4.28 x 10'4 QCWL/D
Ks Vn P*DMVKC/2205
(2-3)
(2-4)
where ,
LSE
Q
C
V
N
withdrawal loss (Mg/yr)
standing-storage or seal loss (Mg/yr)
Product average throughput (bbl/yr)
product withdrawal shell clingage factor (bbl/103 ft2)
density of product (Ib/gal)
tank diameter (ft)
seal factor (dimensionless)
average windspeed (mph); 10 mph assumed average windspeed
seal windspeed exponent (dimensionless)
B-21
-------
P* = vapor pressure function (dimensionless)
Mv = molecular weight of product vapor (Ib/lb mole)
Kc = product factor (dimensionless) =1.0 for VOL
For the purposes of calculating the external floating-roof emission
factors, several additional assumptions were made as follows:
Q = the value for product throughput varied from State to State
(bbl/yr)
C = 0.0015 (for light rust)
W|_ - 5.6 Ib/gal (for gasoline)
Kg s 1.2 (for a metallic shoe with primary seal) and 0.8 (for a
metallic shoe with secondary seal) •
V ~ 10 mph
N = 1.5 (for a metallic shoe with primary seal) and 1.2 (for a
metallic shoe with secondary seal)
P* = (P/PA)/C1 + (1 - P/PA)°-5]2
where,
PA = average atmospheric pressure = 14.7 psia
P = true vapor pressure at average actual organic
liquid storage temperature =5.6 psia
Therefore,
P* - (5.6/14.7)/[l + (1 - 5.6/14.7)0-5]2 = 0.10871
Therefore, after substituting into equation 2-3 and 2-4,
-W
= 4.28 x ID'4 Q (0.0015) (5.6J/78
= 36 x 10-7 Q/78 Mg/yr
= (1.2) (10)1-5 (0.10871) (78) (66) (D/2205
=9.6 Mg/yr for a metallic shoe with primary seal
= (0.8) (10)1-2 (0.10871) (78) (66) (D/2205
=3.2 Mg/yr for a metallic shoe with secondary seal
In summary the VOC emission factors for a typical external floating-
roof tank storing gasoline are 36 x 10-7 Q/-/Q Mg/yr from withdrawal
losses, 9.6 Mg/yr from storage or seal losses on a tank with a
metallic shoe primary seal and 3.2 Mg/yr from storage or seal losses
on a tank with a metallic shoe secondary seal.
B-22
-------
Since some of the emission factors for fixed-roof and external
floating-roof seal losses are expressed on a per tank basis, it was
necessary to convert the State gasoline throughput to reflect the number
of tanks per State. The number of tanks per State was calculated using
the following relationship:
State throughput (bbl/year) = State capacity (bbl) x number of
turnovers/year
Therefore, State capacity (bbl) = State Throughput (bbl)
Number of Turnovers/year
Storage tank capacities of 36,000 bbl. and 16,750 bbl. were assumed
for external floating-roof storage tanks and fixed-roof storage tanks,
respectively.
Number of Tanks/State
State Capacity (bbl)
Storage Tank Capacity (bbl)
This method was used to calculate the number of fixed-roof and floating-
roof tanks in each State. These tank numbers were then applied to the
appropriate emission factor to obtain VOC emissions (in megagrams per
year) for fixed-roof working and breathing losses and external floating-
roof primary and secondary seal losses.
The VOC emission factors for external floating-roof withdrawal
losses are expressed in terms of the throughput of gasoline for each
storage tank and were simply applied to the gasoline throughput for
each State. Table B-8 contains the complete results from the calcula-
tions of VOC emissions from fixed-roof and external floating-roof
storage tanks. Emissions for benzene, EDB, and EDC were calculated
in the same manner as for bulk terminals.
B.2.3 Bulk Plants
The approach to calculating baseline emissions from bulk plants
was essentially the same as for terminals. The emission factors for
uncontrolled bulk plants also came from AP-42. Uncontrolled emissions
from the storage tanks were estimated using the following factors:
breathing losses - 600 mg/liter, filling losses - 1150 mg/liter, and
B-23
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
STATE
ATT
THROUGHPUT
(lOOO'a liters)
NA
THROUGHPUT
(1000's liters)
ATT
THROUGHPUT
ClOOO's barrels)
Alabana
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vernont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
NATIONWIDE TOTAL
0
519, S77
2.715,332
0
423,897 .
0
8,191, 527
0
1,188,202
1,698,519
0
7,39O,O8O
5,328,341
0
547,954
1,649,208
0
0
7,415,178
4,365,948
4.778,419
1,598,519
265,654
0
0
2,264,452
7,494,581
10,096,147
1,241,544
1,805,332
4,672,230
2.024,229
O
0
0
1,396,124
O
16,454,602
998,100
565,232
7,210,118
3,525,005
2,570,935
1,317,589
111,712,675
7,217,995
0
545,884
2,771.248
4,911,193
657,720
1,028,116
9,450,263
10,678,632
• 0
0
17,246,996
2,339,667
0
885,274
7,721,294
1.372,066
5,536,032
15,759,255
0
0
4.544,547
O
1,529,614
1,530,400
11,858,791
632,413
12,959,957
569,124
O
16,299,183
1,718,663
2,843,932
17,725,227
1,356,552
5,766,603
0
8,913,653
12,744,871
1,516,506
324,950
2,417,832
3,388,070
615,328
0
197,377,851
O
3,269
14,824
24.9O3
67,597
17,079
• 0
2,666
0
51,523
0
7,474
1O,6S3
0
46,482
33,514
23,568
36,990
0
3,447
10,373
0
0
46,640
27,461
30,055
10,054
13,400
1,671
0
0
14,243
47,140
63,503
7,8O9
11,355
29 , 388
12,732
0
0
0
8,781
0
103,497
6,278
3,555
, 45,350
22,172
16,171
49,609
8,287
933,545
B-24
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(continued)
STATE
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecti cut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
NATIONWIDE TOTAL
NA
THROUGHPUT NO .
ClOOO's barrels)
45 , 4OO
0
16,200
3,434
193,517
17,431
30,891
4,137
6,467
59,441
67,167 "-
0
0
108,481
14,716
O
5,568
3,872
48,566
8,630
34,821
5,401
99,123
O
0
28,584
0
4,708
9,621
9,626
74,59O
3,978
81,516
3,58O
0
102,519
10,810
17.888
111,489
8,532
36,271
0
56,065
80,163
9,539
2,044
15 , 2O8
21,310
3,870
0
0
ATT
OF TANKS
O
8
37
62
169
43
0
7
0
129
0
19
27
0
116
84
59
92
0
9
26
0
0
116
69
75
25
33
4
O
O
36
118
159
20
28
73
32
0
0
0
22
0
258
16
9
113
55
40
• 124
21
ATT
NO. OF TANKS
FIXED
O
2
10
17
46
12
O
2
0
35
0
5
7
0
31
23
16
25
0
2
7
0
0
31
19
20
7
9
1
0
O
10
32
43
5
8
20
9
O
0
0
6
0
70
4
2
31
15
11
33
6
1,465,171
2331
629
B-25
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(continued)
STATE
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Uest Virginia
Wisconsin
Wyoming
ATT
NO. OF TANKS
FLOATING
PRIMARY SEAL
0
5
23
39
106
27
0
4
0
SI
o
12
17
0
73
53
37
58
0
5
16
0
O
73
43
47
16
21
3
0
0
22
74
10O
12
IS
46
20
O
O
0
14
O
163
10
6
71
35
25
78
13
ATT
NO. OF TAMKS
FLOATING
SECONDARY SEAL
O
1
4
6
17
4
0
1
0
13
0
2
3
0
12
8
6
9
O
1
3
0
0
12
7
8
3
3
0
0
o
4
12
16
2
3
7
3
O
0
0
2
O
26
2
1
11
6
4
12
2
NA
NO. OF TANKS
97
0
35
7
413
37
66
9
14
127
144
O
0
232
31
0
12
8
104
18
74
12
212
O
0
61
0
10
21
21
159
8
174
8
0
219
23
38
238
IS
78
0
120
171
20
4
32
46
8
0
0
NATIONWIDE TOTAL
1469
233
3131
B-26
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(continued)
STATE
54 OF
W/
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
NA NA NA
FLOATING x OF FLOATING NO. OF TANKS
PRIMARY W/ SECONDARY FLOATING
SEAL SEAL PRIMARY SEAL
O.9
0.0
0.9
O.O
O.O
0.0
O.9
O.O
0.0
0.0
0.0
O.O
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.9
0.0
0.9
0.0
O.O
0.0
0.0
O.O
0.9
0.9
0.9
O.O
O.9
0.9
O.O'
0.0
0.9
0.9
0.0
0.0
O.O
0.0
0.0
0.0
0.0
O.O
0.9
0.0
0.0
0.9
O.O
0.0
0.1
0.0
0.1
1.0
1.0
1.0
0.1
1.0
l.O
1.0
• 1.0
O.O
0.0
l.O
l.O
0.0
1.0
l.O
l.O
0.1
l.O
O.I
•l.O
O.O
0.0
1.0
O.O
0.1
0.1
. O.I
1.0
0.1
0.1
1.0
0.0
0.1
0.1
1.0
1.0
1.0
l.O
0.0
1.0
1.0
l.O
0.1
1.0
1.0
0.1
0.0
0.0
87
0
31
0
0
0
59
0
o
0
0
0
0
0
0
0
o
0
0
17
o
10
0
o
0
0
o
9
19
19
0
&
157
0
O
197
21
0
O
a
O
0
0
6
0
4
0
0
7
0
O
NATIONWIDE TOTAL
645
B-27
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(continued)
STATE NA ATT ATT
NO. OF TANKS STRG LOSS STRG ' LOSS
FLOATING PRIM SEAL SEC SEAL
SECONDARY SEAL CMg/yr> CMg/yr)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Verwont
Virginia
Washington
West Virginia
Wisconsin
Uyoning
10
O
3
,7
413
37
7
9
14
127
144
O
0
232
31
0
12
8
1O4
2
74
1
212
0
0
61
0
1
2
2
159
1
17
8
0
22
2
38
238
18
78
0
120
171
20
0
32
46
1
0
0
0.0
49.5
224.5
377.2
1023.9
258.7
0.0
40.4
0.0
.780 . 5
0.0
113.2
161. a
0.0
704.1
507.7
357.0
560.3
0.0
52.2
157.1
0.0
0.0
706.5
416.0
455.3
152.3
2O3.0
25.3
0.0 .
O.O
215.7
714.1
961.9
118.3
172.0
445.2
192.9
0.0
O.O
0.0
133.0
0.0
1567.7
95.1
53.9
687.0
335.8
244.9
751.5
125.5
0.0
2.6
11.9
20. 0
54.3
13.7
0.0
2.1
0.0
41.4
0.0
6.0
8.6
O.O
37.3
26.9
"•' 18.9
29.7
0.0
2.8
8.3
O.O
O.O
37.5
22.1
24.1
8.1
10.8
1.3
0.0
0.0
11.4
37.9
51.0
6.3
9.1
23.6
1O.2
0.0
O.O
0.0
7.1
0.0
83.1
5.0
2.9
36.4
17.8
13.0
39.9
6.7
NATIONWIDE TOTAL
2486
B-28
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(continued)
STATE ATT ATT ATT
STRG LOSS WRKG LOSS WRKG LOSS
FIXED ROOF FLOAT ROOF FIXED ROOF
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana ,
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
0.0
1.9.4
87.8
147.5
4OO.5
101.2
0.0
IS. 8
O.O
305.2
0.0
44.3
£3.3
O.O
275.4
198.6
139. &
219.1
0.0
20.4
61.5
O.O
O.O
276.3
162.7
178.1
59.6
79.4
9.9
O.O
O.O
84.4
279.3
376.2
46.3
67.3
174.1
75.4
0.0
0.0
0.0
52. 0
O.O
613.2
37.2
21.1
268.7
131.4
95.8
293.9
49.1
0.0
0.1
0.6
1.0
2.7
0.7
0.0
0.1
O.O
.2.O
O.O
O.3
0.4
O.O
1.8
1.3
0.9
1.5
O.O
O.I
0.4
O.O
O.O
1.8
1.1
1.2
0.4
0.5
O.I
0.0
0.0
0.6
1.9
2.5
O.3
0.4
1.2
0.5
0.0
O.O
0.0
0.3
0.0
4.1
0.2
0.1
1.8
0.9
0.6
2.0
0.3
O.O
75.4
342. 0
574.5
1559.3
394.0
O.O
61.5
0.0
1188.6
O.O
172.4
246.4
- o.o
1072.3
773.1
543.7
853.3
O.O
79.5
239.3
0.0
O.O
1O75.9
633.5
693.3
231.9
3O9.1
38.5
0.0
O.O
328.6
1087.4
1464.9
18O.1
261.9
677.9
293.7
0.0
0.0
0.0
202.6
O.O
2387.5
144.8
82.0
1046.2
511.5
373.0
1144.4
191.2
B-29
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(continued)
STATE NA MA NA
STRG LOSS STRG LOSS WRKG LOSS
PRIM SEAL SEC SEAL FLOAT ROOF
CMg/yr)
-------
TABLE B-8. STORAGE TANK BASELINE EMISSIONS (Mg/yr)
(concluded)
STATE
Alabama
Alaaka
Arizona
Arkansas
California
Colorado
Connect, i cut.
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Hevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsy 1 van i a
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Weat Virginia
Wisconsin
Wyoming
TOTAL CNTRLD . CNTRLD FIXED ROOF
STATE STRG LOSS WRKG LOSS EMISSION
LOSSES FIXED ROOF FIXED ROOF REDUCTIONS
(Mg/yr) (Mg/yr) (Mg/yr)
-------
draining losses - 460 mg/liter. Uncontrolled emissions from loading of
tank trucks were separated into submerged loading (600 mg/liter) and
splash loading (1440 mg/liter).. Unless submerged loading was specified
by a State, uncontrolled bulk plants were assumed to practice submerged
loading in 75 percent of the cases and splash loading in 25 percent of
the cases.
Based upon EPA test data, it was assumed that controls on bulk
plant storage tanks would reduce filling losses by 95 percent, and
draining losses and tank truck loading losses by 90 percent.12
These controls had no affect on breathing losses. Tank truck loading
losses were estimated using 90 percent reduction of emissions from
trucks in "balance" service (96 mg/liter).
Gasoline throughput data was obtained from the National Emissions
Data System for the year 1982, on a State-by-State and county-by-county
basis. However, only a portion of the total gasoline consumed within a
State passes through a bulk plant. The percentage of total gasoline
throughput that is loaded at bulk plants is calculated for each State
by the following relationship:
Percent of gasoline throughput at bulk plants for State A in 1977 =
Total gasoline throughput at bulk plants for State A in 1977
Total gasoline throughput for State A in 1977
The 1977 Bureau of Census provided the gasoline throughput data which was
required to complete this calculation for all of the States.13 This
percentage was then applied to the 1982 gasoline throughput for each
State and county, to derive the gasoline throughput loaded at bulk
plants in 1982.
Table B-9 contains the complete calculations for the baseline
emission estimate. Benzene, EDB, and EDC baseline emissions were
calculated in the same manner as they were for bulk terminals.
B.2.4 Service Stations
Again, the approach to calculating baseline emissions from service
stations was the same as that for bulk terminals and bulk plants. All
B-32
-------
: p §
i is|
IS
'O
ste us.S
is ^i
S |o
jo A o
cn^>i
lli
—
a
O
— ore — si
"»5 5£ ~ai
2S su rs!
.-" l|| -
f-.
"
os 5 5*2 = "Is'
| o o 5 j< o «>|
CQ "S-ISSI
«
03Cj
u> • 10
09 CM
3 SS ' 2 3 S
• -
— -51
OJ
.a
ns
3 iS
J=1C
all
§ i s
$1
1^
581
! » S
!u»W
*5
{ M> •]
,1 is
-
' ™
! =sl a
-i- '
Sf =,-1
5g^o.a-
'3 111
(MC3>OU) ^IP^i O— -•«» O%^ a3-»»O
-------
£ 3
>= S!
s s
•r in
i V!
!S i| S s
3 |ll -
I
ills *s
•Piss
si
3 s a
III "
^ at n
a.^-) —• r^ m CTI •«• *
5S -S S-S2 S?SS2S §S
Sra —• e
ui —• c
;"l sS SssS=s IS SSSS
en jo
— I
(C
I
B-34
-------
a a
i3 -i
is i!
sl
ill ;
i is
P a*"'
I "A
1= 51
= a
M -S
;« a «rl
— u* c)
SIS r Bl
a =3« S— S — — t
a. in «r e
CTT3 SI "CT>*—| coo ^f^c^e
*ac a tori IOP^<*»C
O i— -o *
^ o «
d!
c
•r^
-M
O
*
•01
a
r—
J2
as
SO)>~.
33
g 8
S
SI
S S
• =-;
-
Is
S S M
O jo —
'£ -i
iit 5fj S S 2
if! =g| I- '
o S> u. ^-1
ill -
—t CM O
«.r» ?»J — ^in ^ — • (-» — «— >--,o
MO mm O^ ,-MCy m — • CM .^
M r-i 10 — 3 vo — r- — to LO ia -".^i ^.IA
o — \0r^. tOO O O ."-. ^ to CO O f*l 33 «j*
Mf^ 'J)C\i -^ ID 3%tO ^fflior^. S JTt *°
- "
O) 3 3!
V O C3I
!! i -
B-35
-------
Ill
1
I J
Is
li
581
a Si
sft s
,.s a"
!;» =si s
i ^=!
" r-
3,
\
£
1 112 =-^= MV3IM =^>tv
^—^•ilr lsS5 35?s -Z~
=S? 2333 =5
r»eM(-i--. ™ f^in_i«)_ J\*^
tu -:
o
o
o
sK,
•s-ss:; as g s s s
i. ^ ft ~* -* — —
cn
i
CQ
01
ja
£
2
^. =
B-36
-------
Footnotes for Table B-9
Twenty-five percent of gasoline throughput was assumed to be loaded by
splash fill.
Seventy-five percent of gasoline throughput was assumed to be loaded
by submerged fi11.
'storage tank filling losses were assumed to be 95 percent controlled.
Storage tank draining losses were assumed to be 90 percent controlled.
a - • •
"State regulations require 77 percent vapor control on storage tank filling
and draining losses and tank truck losses.
The percentage of gasoline throughput that is loaded at bulk plants is
calculated for each State by the following relationship:
Percent of gasoline throughput at bulk plants for State A in 1977 =
Total gasoline throughput at bulk plants for State A in 1977
Total gasoline throughput for State A in 1977
The 1977 Bureau of Census provided the gasoline throughput data which was
required to complete this calculation for all of the States.
g
Storage tank filling and draining losses are 95 percent controlled as
required by the Bay Area regulation.
i
State regulations require bottom or submerged fill for bulk plants with
throughput <_ 4,000 gal /day.
State regulation requires vapor balance and an emission limit of
80 mg/liter for bulk plants with a throughput >4,000 gal/day.
Therefore 58 percent of throughput was assumed to be loaded at plants
with a throughput >4,000 gal/day.
Represents the total gasoline throughput and the percent of gasoline
throughput loaded at bulk plants for the county listed. The total
State throughput is obtained as the sum of the county throughput and
the throughput shown for the remainder of the State. The total State
throughput from bulk plants is obtained as the sum of the county
throughput attributed to bulk plants and the throughput from bulk plants
shown for the remainder of the State.
The designation "county (etc.)" represents the total gasoline throughput
and the percent gasoline throughput loaded at bulk plants for all
counties subject to the CTG requirements. The total State throughput
is obtained as the sum of the county throughput and the throughput shown
for the remainder of the State. The total State throughput from bulk
plants is obtained as the sum of the county throughput attributed to
bulk plants and the throughput from bulk plants shown for the remainder
of the State.
B-37
-------
gasoline, with the exception of agricultural accounts, was assumed to
pass through service stations, which includes retail outlets and private
outlets (see Section 4.2.1).
Emission factors for emissions associated with the service station
underground storage tanks were obtained from AP-42 and were all based
on gasoline throughput. Uncontrolled underground tank filling would be
performed either by splash filling (1380 mg/liter) or submerged filling
(880 mg/liter). Unless otherwise specified, it was assumed that 50
percent of the service stations practiced submerged fill and 50 percent
practiced splash fill.14 Where underground tank filling was controlled
by a balance system, an emission factor of 40 mg/liter was used.
Breathing losses were estimated using a factor of 120 mg/liter.
The majority of all service station automobile refueling is
uncontrolled. The AP-42 factor used for the operation was 1080 mg/liter.
Data from the State of California indicates that systems installed to
control automobile refueling are at least 95 percent efficient.15 The
emission factor used for controlled automobile refueling, therefore,
was 54 mg/liter. Automobile refueling controls are installed only in
portions of California and in Washington, D.C. An emission factor for
spillage (84 mg/liter) was also used. This was used regardless of
whether refueling controls were used.
Table B-10 contains the calculations used for determining baseline
emissions from service stations. The factors used to estimate benzene,
EDB, and EDC, emissions from the service station underground tanks were
the same as for bulk terminals. The factors calculated for automobile
refueling were higher, however, because of the higher temperature associ-
ated with automobile fuel tank (tank temperature assumed as 70°F). The
factors used for automobile refueling were as follows: benzene - 0.0066,
EDB - 0.000052, and EDC - 0.00053.
B-38
-------
*z
s
E
^S.
o
CO
00
•i:
UJ
UJ
3£
_J
UJ
^
CD
CD
t— l CM
_J CO
UJ O\
3 *^
U.
UJ Ql
OS «C
LU
lu
«r co
O CO
g —
13
•^
%
0
_I
•z
o
1—
<
CO
UJ
U
>
o:
UJ
co
O
r™
1
OJ
(O
f—
UJ _J i-
< "*"-^
1 — o en
00 t- E
rf)
CO
¥
t- J3
0) -O C
3 flj QJ ^
**~ r— • E ^*,
o> f— o» en
01 ° re E
O +J r2 «3-
•*-> c CLLO
3 o f—
ttl C7
O E
10
i— O
Q.CO
a
UJ _J i.
i— o^>
„,_
•r- en
.c E
10 O
(U CM
en m
e
•o
J ^
c If
t— ^— O
+•> CLCO
to to co
CO
a
u
>
"cn^
3
O (/>
jfg
s
>^
c
3
O
C_3
(O
4->
CO
§
CO
10
O
LO
r-.
r-*
LO
«— 1
t-H
to
to
CO
en
CM
cn
^
CM
^
•o
2
4-1
CO
(TJ
E
ta
-Q
n3
<
LO O
O «3-
i& r^
to
^ <•
^~ f^
tO CM
LO CO
LO
SLO
to
to
CM CM*-*
LO 9
to
z
CO ^
* c
CO •»-
tn
*x
i-
rt3 fO
— S
ta
to c:
JK; o
W N
fO «^-
< <
LO
CM
O
LO
r-. to
CO
CM O
r-. en
^*
cn
LO
CM
LO
CM U9
^«
CM
0
to
CM
Oi CD
LO CO
O CO
!^ CO
,O LO
r^. to
co"
,ii
in
re
3
in D_
fO
in
. c
rt3
<:
o
10
•o
£
^ co •— i LO en
r^ r-. CM to ^H
<- r— i , C 3 CO
•i- (O (O O
C CQ CO CO i-
i- OJ
O .C
U- . 4J
i- O
• ro
0
CO
*r>
to
CO CO
CM CO
CM CM
CO CO
co cn
cn cn
CM CM
$
CO
CO
to co
CM CO
CO CO
CO
LO
en
•—i
ft
•— i
CM CO
CO ^
CO CM
LO ^
r^ i —
CM CM
1 —
U
4->
•»"
"•a
•3
£
t/i
4->
3
CJ
ZJ
U
0)
c
c:
o
CJ
B-39
-------
Ul _J i
v-< :
.— \o
CO «3- (-.
§
* !
J
-
coo ^ -* -^
co en ^f •—• ro
CM CO CO CO "3"
—t crt «H CM -H
O r- C
si1
CO r^ —4
CVJ S
T3
O)
:§
« CO 01 O -< —
O
•r- U
•i
QQ
O)
(O 1—
£ £
2- "5
(A O O3:
i— i~ (U rO
o u. cs =r
B-40
-------
F— -O C
OJ QJ OJ ^
^ OJ cn
O o E
U (15
0.0
(AC
S"
CM CM ~-t Ch CO O
O\ CO CO r- O CM
1-4 —« LO o* r*.
^M CO
CM _4
^1 ^4 O
^4 ^*4 \Q O
LO tO _4 f
B
5
o • »— <
g*-
=3 LU
«s
O C3
-H CO
CM O
cn «—i
—4 CM
•a
O)
r*- LO -*f
d LO a\
CM LO CM
co o c
er> en
to —i r
4-J
£=
O
•*- to .—. • —
I
CD
-------
PiS-1
3 U •*-
5
£
§—
S- CM
_J O3
S2
s§
e-
3K
§
-H CO
en en
dec'
J Cn O
j r*- in
en to Oco o o en coco ^j- cMcn
CM en en tn ro o co ^••er o into
^~t en ~H to in m co •—' »•* co
I
—t - If)
en-< ^- ^^\
g
5
O r- <
in 1-4 co
«-H in" ^-
•o
O)
ZJ
fc.
vt
«
S«* o
0 -H
to vo co
to co en
cu en in
8
0
^
en
CO
in
CO
£
tn 2
to co
CM tO
CM
— < CM in
CO CO CM
en m o
^r en o
r-. in r*-
^- CM
en to
o in
O
»— i
^- CM 00
^ CO CO
^* o en
CM CO CM
f-t ^t to
e
o
o
I
ca
(O
•w X
irt
J£f— i. —
i. cn t-
^S o vi
o o a.
3 tt
i :
—• o
« i—
T- *— (O
O F—
B-42
-------
: g -
e><
as en
^J -«
co r»-
-
•5.S
too
«O IO \O .-< CO
SS SS S
- - -, <~< IO CO *-t
1 E to O V V
j \ ^CM c^ro ^
t**l
CM CO IX)
os vo * r.
V ^ CM CM
O CO
T** ^-t
O
O
1
CQ
JQ
,
W (U ,C
C -O 4J
B-43
-------
i
1?
'^
I
01
S
13
^
Jg*
£
•HM
C/)
g
I/}
5
UJ
s:
-j cn
LU K
I< a
CO «
o
S?7 c
uj en •**
Of wC /)
— * it* «.
p >- a»
fr— U
= UJ •-
•C to >
C^ CO 0)
g— «
g
«•!
g
1
>—
V)
.
CJ
•5
cc
UJ
CO
1
c
!Jj
c
C3
«— •
*
O
ca
(U
r—
(ft
t—
i
UJ r t tn 1 t£)
< Q "S *~i
C/) J— jj-| ,-4*
Ini
-<
XI CO
SCO
(O
a>-Z
c ^
f^O
Q *T
CO
in
fO) CM O
4J rotn
•»- CM Crt
§tn «a-
tn to CM
u- to co
i— §>
^^
j
cu
c
0
•o
c
0)
>s *•*
o u
U J->
3; I
4^ CU
CO >•
f^^
o
CM
•-4
to co
0 0
to CM
CO — <
r»- CM
o
o
oT
in o
to cn
CO CM
in
cn_
CO
CO
^
en
CO CM
»H ro
-4 CO
CM *r
r»- CM
*^
o!
c
o
cn
c
10
i —
o
CO
cn co
CM CM
r— 01
o in
co to
ro ro
Ol
^T
ro r~
CM O
CM
CM
_
U)
in
*
vo
CO
•~4
gs
0 0
in co
CM CO
to co
CO CO
^^
S
cu
if
c re
cn
•»-
.e
5
X
o
r-
ro
to CM
— i in
CM
r- in
r- vo
CM
^
CO
CO
cn «f
o r»
CO
«T in
p~ CM
_-,
ro r-
-. CM
-"
in co
ro CM
cn co
o in
in to
CM
•"^
10 i
•*— TO
C C
f- 4J
cn 3
i- 0.
I/)
cu
2
o *r
co PO
»-* Lfl
cn i-<
to ^o — i
CM *r ^i
IO O" CM
»-< CO «^
CO L(1 1-4
-4
CM (O
** «
CM ^r co
LO cn in
CO LO »-<
en
O
o%
0
g
r-- CO
»H cn
~H rt
coco cn
*~4 to CO
co in in
ro CM r~
ro in ~*
o» cn co
CM *r ^
^.
-2i ?
C 4^
g 1^
C i- -—
i- ca
c/i cn
c c
O *i—
U S
in o
tc >»
2 2B
CM rx.
tfi cr»
cn ^o
^ o
«er
o
CO
2=
LU
0 ^
CO LU
«— « oo
S
3
CL
JZ
en
|
JE
4->
CU
c
o
m
TO
cn
4->
ro
jr
4-)
*
T3 l/l
aj
0 10
01 "S
*— 4.J
'o E
4J 'Z
C to
O 01
u
V)
>,f-
l&
i_ aj
c. ."
5 U
U f-
01 u_
I- 0)
c/> o
01 t.
co c
0 0
•— o
c .
=3 £_
«- 10
Ol Q.
•O
Ol C
— re
tj
•i- 1C
-C CU
cu <-
5» 10
CU
t_
.01
.-§
t/}
s
(_
TO
c
i
o
«sr
c
TO
4-1
cn
>t
TO
u
)J^
u
CU
CU
cn
cu •
u ^
TO CU
•Q .f«
CU J_)
4-> U
.^2
F— ^
(/) VI
c "*"
£- CU
3 TO
CT4J
0) CO
O C
4-> 'i-
0 U
E TO
OX?
in o
•r~ Q.
S TO
UJ >
0
O
4J
J_>
C
CU
^
cu
a.
CO
£
"O
ai
at
TO
cu
£. •
u in
CU J-»
•^3 c
3
(U O
t. 0
cu o
3E TO
m i—
c re
o c.
*•" 3
(/) 4->
in i—
f- 3
(_
cu cn
C TO
t— o
Q) ^J
in
cn..-
c: -^
t— (U
r— L.
a c-
4-> O
u c
'S o
<- u
cu u
CO (O
•a
B-44
-------
B. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14,
GCA Corporation. Status Summary of State Group I'VOC RACT Regulations
as of March 10, 1980. Report to U.S. Environmental Protection
Agency. Research Triangle Park, N.C. Contract No. 68-02-2607, Task
No. 44. May 1980. 168 p.
Reference 1.
Peterson, P.R. et al., Evaluation of Hydrocarbon Emissions from
Petroleum Liquid Storage. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Publication No. EPA-450/3-78-012.
March 1978.
Pacific Environmental Services, Inc. Estimated Nationwide Petroleum
Storage Tank VOC Emissions for the Years 1983 and 1988. Report to
TRW Environmental Engineering Division, Research Triangle Park, N.C.
Contract No. M23399JL3M. April 5, 1983.
Bulk Gasoline Terminals - Background Information for Promulgated
Standards. U.S. Environmental Protection Agency. Research Triangle
Park, N.C. Publication No. EPA-450/3-80-038b. August 1983.
Transportation and Marketing of Petroleum Liquids. In: Compilation
of Air Pollutant Emission Factors. U.S. Environmental Protection
Agency. Research Triangle Park, N.C. July 1979.
U.S. Environmental Protection Agency. Federal Register. Vol. 47
No. 210. October 29, 1982. p. 49323.
Norton, Robert L. Evaluation of Vapor Leaks and Development of
Monitoring Procedures for Gasoline Tank Trucks and Vapor Piping.
U.S. Environmental Protection Agency. Research Triangle Park, N.C.
Publication No. EPA-450/3-79-018. April 1979.
Reference 6.
Reference 1, p. 3-24.
Storage of Organic Liquids. In: Compilation of Air Pollutant
Emission Factors. U.S.. Environmental Protection Agency. Research
Triangle Park, N.C. April 1981.
Pacific Environmental Services, Inc. Compliance Analysis of Small
Bulk Plants. Report to U.S. Environmental Protection Agency, Region
VIII. Denver, Colorado. Contract No. 68-01-3156, Task 17. December
1976.
U.S. Department of Commerce. 1977 Census of Wholesale Trade-Volume
I, Subject Statistics. Petroleum Bulk Stations and Terminals.
August 1981.
Standard Support Environmental Impact Statement for Control of
Benzene From the Gasoline Marketing Industry. U.S. Environmental
B-45
-------
Protection Agency. Research Triangle Park, N.C. (Draft) June
1978; p. 2-17.
15. Memorandum from Norton, R.L., Pacific Environmental Services,
Inc., to Shedd, S.A., Environmental Protection Agency. December
20, 1983. Trip Report to California Air Resources Board.
B-46
-------
APPENDIX C
ASSESSMENT OF ONBOARD CONTROLS
C-l
-------
-------
EPA-AA-SDSB-84-01
Technical Report
The Feasibility, Cost,
and Cost Effectiveness of
Onboard Vapor Control
Glenn W. Passavant
March 1984
NOTICE
Technical Reports do not necessarily represent final EPA
decisions or positions. They are intended to present
technical analysis of issues using data which are
currently available. The purpose in the release of such
reports is to facilitate the exchange of technical
information and to inform the public of technical
developments which may form the basis for a final EPA
decision, position or regulatory action.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Sources
Office of Air and Radiation
U. S. Environmental Protection Agency
C-3
-------
Table of Contents
Page
I. Introduction 1
II. Technological Feasibility 1
III. In-Use Performance of Onboard Systems 7
IV. In-Use Emission Control Effectiveness 10
V. Costs of Onboard Vapor Recovery 13
VI. Cost Effectiveness 21
VII. Leadtime Requirements 24
VIII. Onboard Control Versus Time 27
IX.
Conclusions 33
References
Appendix A: "Recommendation on Feasibility for
Refueling Loss Control," February 1980.
. . 35
Onboard
Appendix B: "LDV and LOT Operation and Usage Characteristics"
Appendix C: Tables from "Manufacturing Costs and Automotive
Retail Price Equivalent of Onboard Vapor Recovery
System for Gasoline - Filling Vapors"
C-4
-------
I.
Introduction
This report updates the previous analysis of the
technological feasibility, in-use effectiveness, cost, and cost
effectiveness of an onboard vapor recovery system for
controlling refueling emissions from gasoline-fueled motor
vehicles. The last report in this area is dated February
1980. In that report it was concluded that onboard vapor
recovery was feasible for light-duty vehicles (LDVs). However,
some question remained about the feasibility for other types of
gasoline-fueled motor vehicles and the cost effectiveness of
controlling refueling vapors through the use of an onboard
system.
Therefore, this report addresses the feasibility of
control for other gasoline-fueled motor vehicles (light-duty
trucks_ (LDTs) and heavy-duty gasoline-fueled vehicles (HDGVs))
in addition to LDVs, and also examines those factors related to
cost effectiveness. The feasibility is examined for HDGVs, but
the cost and emission .reduction impacts a-re not determined.
However, cost-effectiveness values similar to those calculated
for LDVs and LDTs would be expected.
This report begins with a discussion of the technological
feasibility of onboard control, and this will be followed by a
calculation of the in-use effectiveness of onboard control
systems. After reviewing and updating the previous estimates
of the costs of control, the cost effectiveness of an onboard
strategy will be calculated. In addition, a fifth section of
the report estimates the leadtime necessary to implement
onboard controls, and the last section estimates the time
required for an onboard strategy to achieve control of a
substantial portion of in-use refueling emissions. A summary
of the overall conclusions closes the report.
II. Technological Feasibility
A.
Introduction
The bulk of the experimental work in the area of onboard
vapor recovery has been performed by the American Petroleum
Institute (API) and their contractors, Exxon, Mobil, and
Atlantic Richfield (ARGO). They completed a vehicle
demonstration of onboard vapor recovery in October of 1978[1].
The results of that study strongly suggest that onboard
controls are feasible and effective in controlling gasoline
refueling losses from low- to mid-mileage LDVs, with only a
negligible impact on a vehicle's ability to comply with current
exhaust or evaporative emission standards.
C-5
-------
-2-
Following the release of the API work in 1978, EPA
solicited comments from the motor vehicle industry concerning
the cost and technological feasibility of onboard controls for
LDVs and LDTs. These comments were incorporated in EPA's
analysis of the API vehicle demonstration program _ (This
report is presented in Appendix A, "Recommendation on
Feasibility for Onboard Refueling Loss Control," dated February
1980.) [2] The judgment that onboard vapor recovery is
technologically feasible for gasoline-fueled motor vehicles is
based largely on this analysis of the API work and the
technological feasibility comments submitted by the motor
vehicle industry. In the remainder of this section, the
information leading to the conclusion that onboard control is
feasible for LDVs is reviewed, and the feasibility of
controlling LDTs and HDGVs is discussed.
B. Review of LDV Feasibility
1. New System Performance
The onboard contol effectiveness of new systems is based
on the results of the API vehicle demonstration program. This
program consisted of SHED tests of. the entire system minus the
filipipe seal (the fillpipe was plugged) and bench SHED tests
of the ARCO rotary fillpipe seal. These tests showed that
refueling emission control efficiency ranged from 98.2 to 99.3
percent for both total HC and benzene, with an average value of
98.9 percent.[3] Based on these results, a control system
efficiency of at least 98 percent is judged to be
representative of potential new vehicle control for the
canister/modified fillpipe and seal system evaluated by API. A
diagram of the system evaluated by API is given in Figure A-l
of Appendix A.
2.
Mechanical Durability
EPA's 1980 report summarized the ARCO API durability data
on the nozzle/fillpipe seal effectiveness. These data were of
necessity derived from an accelerated test program, and
therefore concerns about seal durability over time could not be
addressed. After completion of the original work for API, ARCO
installed a fillpipe cone seal (Figure A-6 of Appendix A) in a
company vehicle and monitored seal effectiveness over 26 months
and approximately 54,000 road miles near their Harvey, Illinois
facility. During the 26 months, the seal was exposed to
environmental extremes representative of most of the
continential United States.
ARCO tested the seal effectiveness by measuring the HC
concentration at the fillpipe/nozzle interface each time the
C-6
-------
— 3 —
vehicle was refueled. At periodic intervals the seal
effectiveness was checked by measuring the leak rate past the
seal using a specially designed nozzle to pressurize the system
with nitrogen. The pressure check tests were performed at
system pressures corresponding to 5, 10, 15, and 20 inches of
water, at which the seal effectiveness was still 99 percent. A
pressure of 4-5 inches of water is typical for a normal fuel
fill. Therefore, the ARCO data indicate excellent sealing
capability over time. Overall, seal effectiveness was better
than 99 percent after two years of service using unleaded fuel,
and 99 percent effective after an additional 11,000 miles using
a high concentration (20 percent) methanol/gasoline blend.[4]
The ARCO in-use data suggest that effective, durable,
low-cost fillpipe seals (rotary seals or cone seals) are
feasible for LDVs for in-use service to at least 50,000 miles
over a two-year period. The available data is not conclusive
as to which type of seal is preferable. The important
question, which has. not been fully addressed, is whether the
fillpipe seal will be effective throughout the full useful life
of a vehicle. Remaining effective implies no significant
deterioration, contraction, or expansion problems under normal
environmental conditions such that the seal fails to achieve a
leak-free connection with the fuel nozzle, or the nozzle cannot
be inserted through the fillpipe seal at all. At this time,
durability data do not exist out to the full average life of a
typical LDV, 100,000 miles (10 years), or 120,000 miles (11
years) for LDTs. However, the durability data up to 65,000
miles indicate no reason why the seal would not continue to
perform over its full life. Given this durability data to
65,000 miles, the relative simplicity of the system design, and
the nature of its use, it is reasonable to project that
full-life performance should occur.
3.
Effect on Gaseous and Evaporative Emissions
The work conducted by API indicated that purging the
refueling vapors had no significant effect on exhaust
emissions. However, it should be cautioned that the tests were
conducted on 1978 and earlier model year vehicles which had
emission levels higher than today's new vehicles. Also, test
procedures for measuring refueling emissions have not yet been
fully developed and it should be recognized that the test
procedure requirements for purging the vapor recovery canister
could impact exhaust and evaporative emissions. However, it is
expected that through proper design of the onboard control
systems (taking into consideration appropriate purge
requirements), increases in exhaust or evaporative emissions
can be avoided.
C-7
-------
c.
-4-
Feasibility for LDTs and HDGVs
Even though the API feasibility evaluation project
involved only LDVs, onboard control technology should be
directly and fully adaptable to LDTs,, LDV and LOT fuel systems
are practically identical, and both use similar hardware to
comply with the 2.0 g/test evaporative emission standard. The
primary difference between LDVs and LDTs is in the fuel tank
specifications. . Analysis of 1984 certification information and
discussions with the manufacturers indicate that, on average,
LDT fuel tanks are about 25 percent larger than LDV fuel tanks,
and about 20 percent of LDTs use dual fuel tanks. A larger
volume fuel tank would require more charcoal in the canister to
accommodate the increased volume of refueling vapors, and LDTs
using dual fuel tanks may require a separate onboard control
system for each tank. However, neither of these differences
has an effect on the conceptual design or technological
feasibility of an onboard control system.
The above considerations apply equally to many of the
smaller HDGVs (those less than 14,000 Ibs gross vehicle weight
(GVW)). Approximately 65 to 70 percent of all HDGVs are
essentially LDT derivatives.[5] These HDGVs are essentially
the same as their parent LDTs in their basic chassis, body and
powertrain designs, but have been classified as, HDGVs for
purposes of emission control because their GVW, frontal area,
or curb weight are just above the LDT/HDGV cut-off points.
Although these characteristics would have an effect on exhaust
emissions, they would have no effect on the ability to comply
with an onboard vapor recovery requirement. The key parameter
which influences feasibility is fuel tank volume. Most of
these smaller HDGVs have fuel tank sizes similar to the heavier
LDTs, so the onboard systems used on LDTs could be applied
directly to the smaller HDGVs. For those smaller HDGVs with
larger fuel tanks, larger charcoal canister volumes could be
utilized.
The application of an onboard control requirement to many
of the larger, heavier GVW HDGVs is somewhat more complicated.
HDGVs in this group are sold in many different configurations
with different fuel tank sizes and locations. Also, many of
these HDGVs are sold initially as incomplete vehicles by the
primary manufacturer to a secondary manufacturer. In the most
common case, the primary manufacturer produces the chassis and
the secondary manufacturer adds a payload-carrying device. In
C-8
-------
-5-
some cases, the overall vehicle fuel capacity is increased by
the secondary manufacturer. In these cases a problem might
arise because the primary manufacturer would have to certify
the onboard control system before it was sold to the consumer,
but the secondary manufacturer could affect the integrity of
the system. Many of these problems are similar to those
encountered and. resolved in the recent HDGV evaporative
emissions final rule, which suggests that implementation
problems can be solved. Also, for the foreseeable future,
these heavier GVW HDGVs will be using leaded fuel and will not
have a filler neck restrictor. Thus, the onboard control
system for these HDGVs would require the additional hardware
associated with the filler neck restrictor already present on
vehicles using unleaded fuel if they were to use fillpipe seals
similar to those used on LDVs and LDTs.
Although application of an onboard control requirement to
the heavier HDGVs is not as straightforward as for the lighter
weight HDGVs and may be more costly, there does not appear to
be any technological reason why onboard control would not be
feasible for the heavier GVW HDGVs. One possible approach for
applying an onboard control requirement to HDGVs if costs were
excessive, would be to require control for the lighter weight
HDGVs (under 14,000 Ibs GVW) and defer control for the heavier
weight HDGVs (over 14,000 Ibs GVW).
D.
Safety Considerations
In addition to concerns about the performance and
durability of onboard control systems for LDVs, LDTS, and
HDGVs, there are some potential safety considerations which
require evaluation. If a blockage of some type occurs in the
line from the fuel tank to the charcoal canister, pressure
buildups within the system may lead to damage of the fillpipe
seal and possibly a spurt of gasoline from the fuel inlet.
Second, if the automatic shut-off of the gasoline nozzle fails
to operate properly, an overfill of the tank could occur which
also could result in damage to the fillpipe seal and a spurt of
fuel. Third, there is also the possibility that a failure of
the vapor/liquid separator and rollover check valve in the line
from the fuel tank to the canister and an improperly operating
automatic gasoline nozzle shut-off could lead to a tank
overfill and fuel flowing up the line and poisoning the
canister.
C-9
-------
-6-
These problems could likely be resolved with a pressure
relief valve which would vent vapor or gasoline overpressure to
the environment should problems occur. However, prototype
pressure relief systems have not been fully developed and
failure modes have not yet been adequately identified and
evaluated. Also, increasing the diameter of the vapor line
from the fuel tank to the charcoal canister and increasing the
vapor flow capacity of the vapor/liquid separator and the
rollover check valve should decrease overpressure problems.
Thus while some questions relative to the safety of an onboard
system remain unanswered, any problems should be solvable with
direct engineering effort.
E.
Summary
The work conducted through API and later by ARCO suggests
that onboard control is technologically feasible for LDVs, and
evidence is that in-use durability of these systems should be
excellent. Due to the fundamental similarities between LDVs
and LDTs, onboard control should also be feasible for LDTs. In
fact, the onboard systems would likely be nearly identical with
the possible exception of charcoal canister size.
The demonstrated onboard technology also appears adaptable
to HDGVs. For those HDGVs of less than 14,000 Ibs GVW (65-70
percent of all HDGVs), the application of onboard technology
would in all likelihood essentially be accomplished through an
extension of LDT systems. The only major difference might be
larger canister sizes to accommodate the larger fuel tanks used
on some of these HDGVs. Onboard systems for the heavier HDGVs
(those whose GVW exceeds 14,000 Ibs) would be somewhat more
complicated and costly, but nevertheless appear practicable.
It should be possible to minimize any effect of an onboard
vapor recovery requirement on exhaust and evaporative emission
levels through the proper design of the onboard system.
There are some potential safety considerations which must
be identified, evaluated, and resolved. However, it is likely
that these can be adequately addressed through the use of a
pressure relief valve within the fuel delivery system.
Although control systems could be applied to HDGVs., we
have not quantified the costs, benefits, and cost effectiveness
for these vehicles at this time. HDGVs comprise only about 3
percent of the gasoline-fueled vehicles produced each year and
represent on the order of 5 percent of annual nationwide total
gasoline consumption.[6] The remainder of this paper
concentrates on the costs, benefits, and cost effectiveness for
LDVs and LDTs.
C-10
-------
-7-
III. In-Use Performance of Onboard Systems
A.
Introduction
Losses in the effectiveness of in-use onboard systems can
occur through two mechanisms: tampering or deterioration of
the efficiency of the system. Tampering occurs when
individuals purposely disable part or all of the onboard
control system. Tampering could occur with the fillpipe seal
and the charcoal canister and related hoses. System
deterioration occurs when control efficiency of the onboard
system declines with mileage and/or time. Either mechanism
renders the onboard vapor recovery system partially or
completely ineffective. The projected effects of these
mechanisms on onboard system performance are discussed below.
B'. Tampering
1. Fillpipe Seal Tampering
It is possible that fillpipe seals could be subject to
tampering similar to that reported for the tampering with
fillpipe restrictors in vehicles using unleaded fuel, since
violation of the leaded fuel restrictor would also destroy the
vapor recovery seal. Fillpipe tampering data is available from
the National Enforcement Investigations Center (NEIC).[7]
These data show substantial differences in fillpipe restrictor
tampering in areas which have inspection/maintenance (I/M)
programs versus non-I/M areas and different levels of fillpipe
tampering for LDVs. and LDTs. (See Figure B-l of Appendix B.)
A linear regression of this fillpipe tampering data versus
mileage for 1982 produces the following results:
LDVs:
LDTs:
I/M Areas:
Non-I/M Areas:
I/M Areas:
Non-I/M Areas:
TAMP = -1.43 + 1.14(M)
TAMP = -0.78 + 1.65(M)
TAMP = 3.55 + 1.14(M)
TAMP = 10.6 + 1.65(M)
Where:
TAMP = Tampering incidence expressed in percent at a
particular vehicle mileage.
M = Mileage/10,000 miles.
It should be noted that the tampering increase rates (the
change in tampering incidence with mileage) for LDVs and LDTs
are the same. This was taken to be the case because the size
of the LDT sample was too small (323 LDTs versus 1,999 LDVs) to
C-ll
-------
-8-
allow meaningful rates to be.determined. However, the LDT data
were used to derive a mean tampering level for LDTs at the
average LDT mileage, with the LDV tampering rate being applied
to that single point.
However, these tampering rates are conservatively high for
the late 1980's and beyond when an onboard vapor recovery
requirement might be implemented. The primary motive for
fillpipe tampering is to permit the use of somewhat less
expensive leaded fuel in LDVs and LDTs designed to use unleaded
fuel. The tampering rate itself depends on the availability of
leaded fuel, the leaded to unleaded fuel price differential,
and the actual difficulty and" other effects of the tampering
process itself. The tampering rates given above are based on
data gathered in the Summer of 1982, when leaded fuel was
readily available at a differential of about five cents per
gallon and fillpipe tampering was a relatively simple process,
usually with no effect on the integrity of the filler neck
itself.
However, in the late 1980's and beyond, leaded fuel will
be generally less available due to lower overall demand, and
with less demand it is possible that the leaded to unleaded
fuel price differential would decrease. In addition, there
were only a handful of I/M programs in place in 1982 when this
data was gathered. As more I/M programs are implemented over
the next few years tampering should decrease. Perhaps most
importantly, the onboard control requirement could be
implemented with a certification performance standard such as
the parameter adjustment requirement for carburetors on
gasoline-fueled vehicles. This requirement would force the
design of filler neck restrictors and fillpipe seals which are
a more integral part of the fillpipe, thus reducing the
accessibility and success of tampering. Therefore, it is
reasonable to project that fillpipe tampering will decrease
markedly by the later 1980's. After briefly considering the
rate of tampering with charcoal canisters and hoses, a
composite tampering rate will be determined for LDVs and LDTs
if fillpipe tampering is reduced by 50 percent due to the
reasons discussed above.
2.
Charcoal Canister and Hose Tampering
Tampering with the charcoal canister and related
connecting hoses would also destroy the effectiveness of an
onboard vapor recovery system. Since the control approach
expected by EPA assumes an integrated onboard/evaporative
emissions control system, currently available data on tampering
with evaporative emission systems (canisters/hoses) would be
directly applicable to onboard controls as well. Evaporative
C-12
-------
-9-
emission control system tampering rates are also available from
the National Enforcement Investigations Center (NEIC) for
1982. [7] The tampering rates are different for LDVs and LDTs,
but not different in I/M versus non-I/M areas, since- the
evaporative emissions control system is normally not checked
during I/M. (See Figure B-2 of Appendix B.) A linear
regression of this most recent (1982) NEIC evaporative
emissions control system tampering data provides the following
results:
LDVs;
LDTs:
TAMP = -0.55 + .360(M)
TAMP = 2.85 + .360(M)
TAMP and . M are as described previously above, and the
explanation regarding the derivation tampering incidence and
tampering rates for LDVs and LDTs is also applicable.
3. Composite Tampering Rates
It would be convenient to have composite tampering rate
equations for LDVs and LDTs for computing the in-use emission
reductions expected from an onboard vapor recovery system.
Since tampering with either the fillpipe seal or the
canister/hoses would disable the vehicle's onboard system, the
slopes and intercepts of the different tampering equations
given above could simply be added for LDVs and LDTs
respectively. However, this would overstate the total effect
of tampering, because some vehicle owners tamper with both the
fillpipe and the charcoal canister and hoses. Since disabling
either would eliminate the effectiveness of onboard control,
just adding the equations would lead to some double counting.
To determine the degree of overlap in tampering, the
National Enforcement Investigations Center data discussed above
was analyzed. After the overlap tampering was accounted for, a
linear regression of the combined data sets was conducted and
the following regression equations were obtained:
LDVs:
LDTs:
I/M Areas:
Non-I/M Areas:
I/M Areas:
Non-I/M Areas:
TAMP = -1.47 + 1.442(M)
TAMP = -1.52 + 2.114(M)
TAMP = 6.42 + 1.442(M)
TAMP = 13.67 + 2.114(M)
The tampering levels in I/M and non-I/M areas can be
weighted (40 percent I/M and 60 percent non-I/M) according to
the fractions of the U.S. population residing in the two types
of areas. The results of this tampering rate weighting are
shown below.
C-13
-------
-10-
LDVs; TAMP = -1.5 + 1.8452(M)
LDTst TAMP = 10.77 + 1.8452(M)
The weighted composite tampering rate equations given
above are based on current tampering rate data. As was
discussed above, it is likely that tampering will decrease in
the future, thus improving the overall in-use effectiveness of
an onboard vapor recovery program. To estimate this potential
decrease in tampering, the portion of the above composite
tampering rates associated with the fillpipe will be reduced by
50 percent. The portion of the composite tampering rate due
solely to tampering with the fillpipe was taken to be the
difference between the composite rate and the tampering rate
associated with the evaporative HC control system. The result
of decreasing this difference by 50 percent is shown below.
LDVs: TAMP = -1.0 + 1.1026 (M)
LDTS: TAMP - 6.81 + 1.1026 (M)
These projected weighted composite tampering rate
equations will be used in calculating the in-use emission
reductions for onboard vapor recovery control. These equations
will be taken as applicable to 1988 and later model year LDVs
and LDTs.
C.
Deterioration
As was discussed above ,in Section II.B. 2., the ARCO seal
durability data show no deterioration of the onboard vapor
recovery system effectiveness with mileage. This is consistent
with historical EPA certification information which shows that
the efficiency of evaporative emission control systems (which
are similar to vapor recovery systems), do not deteriorate with
mileage. Limited in-use testing of LDV and LOT evaporative
emission systems shows that these systems do function as
designed. At the same time, some small loss of effectiveness
with mileage, on the order of a few percent, would appear
reasonable due to contamination of the charcoal, channeling,
aging, leaks, etc. With no data, it is not possible to
estimate this loss quantitatively. However, the tampering
rates of the previous section .appear large enough to overwhelm
any expected loss in efficiency due to deterioration. Thus,
the losses in system effectiveness due to tampering will be
taken to include any losses due to deterioration.
IV. In-Use Emission Control Effectiveness
An estimate of the annual or lifetime HC emission
reduction potential of vapor-controlled LDVs and LDTs is a
C-14
-------
-11-
function of annual or lifetime mileage, the vehicle's fuel
economy, the uncontrolled refueling emission factor, the
control system effectiveness, and an adjustment factor that
accounts for a loss of effectiveness in-use, (i.e.,
tampering). This relationship on an annual basis is expressed
below:
HC = (VMT)(EF)(NSEFF)(NTAMP)
MPG
Where:
HC =
VMT »
MPG =
EF =
NSEFF
Average annual HC emission reduction per vehicle,
grams.
Average annual mileage, miles. '""
Average in-use fuel economy, miles per gallon.
The uncontrolled refueling loss emission factor, or
4.54 g of HC per gallon of dispensed gasoline.
Onboard control system efficiency of new vehicles,
or 0.98.
NTAMP = An adjustment factor which discounts for in-use
tampering. NTAMP equals (1-TAMP) for any given year.
Estimates of in-use (over the road) fuel economy for new
LDVs and LDTs are based on projections of fleetwide
improvements for 1988 and later years. Annual vehicle miles of
travel estimates are those used in the EPA emission factors
program.[8] This information is contained in detail in
Appendix B. The new vehicle control 'system efficiency of 0.98
and the range of tampering rates as a function of mileage were
discussed above.
The uncontrolled refueling loss emission factor (4.-54 g
HC/gal) is based on recent work conducted by the California Air
Resources Board (CARB).[9] This figure is 11 percent larger
than the emission factor contained in the EPA emissions factor
document (AP-42).[10] The CARB emission factor was selected
over the AP-42 emission factor for three reasons. First, the
EPA emission factor document expressed uncertainty about it's
emission factor value. Second, the CARB factor is based on
data at least 5 years more recent than the AP-42 • emission
factor. And third, an increase in the emission factor can be
explained by the steady increase in gasoline volatility
(expressed as Reid Vapor Pressure) over the past 10 years.[11]
In fact, information recently submitted to EPA by General
C-15
-------
-12-
Motors indicates that.4.54 g/gal may be conservatively low for
today's commercial gasolines.[12]
This equation and these factors may be used to estimate
the lifetime emission reductions for LDVs and LDTs. For any
given model year, the only variable would be the fuel
consumption. The remaining factors for each year of the
vehicle life can be determined and summed to get a single
factor representative for the entire vehicle average lifetime.
Using an average lifetime of 100,000 miles for LDVs and 120,000
miles for LDTs, and assuming that the tampering occurs at the
midpoint in each year, the lifetime HC reduction can be
calculated for any model year.
HC (tons) =
AL
(NSEFF)(EF)
( 453 .6 ) ( 2 , 0 0 0 ) ( MPG
(VMT
NTAMP )
Working through the mathematics of this calculation, the
following equations have been determined for calculating the
lifetime tons of HC emission reductions for LDVs and LDTs.
These are based on an average lifetime (AL) of 100,000 miles
for LDVs and 120,000 miles for LDTs.
LDVS: HC * .4683
MPG
LDTS: HC =
.5095
MPG
With these equations, the average lifetime in-use HC emission
reductions from onboard vapor recovery for any model year LDV
or LOT can be determined using the in-use fuel economy
estimates in Table B-3 of Appendix B. For example, for 1988
model year vehicles, LDV and LOT reductions of 0.0178 and
0.0264 tons respectively per vehicle,, would occur. These will
be used in a later portion of the analysis to calculate the
cost effectiveness.
In addition to computing the annual or lifetime emission
reductions on a per vehicle basis, the nationwide annual
reduction in the overall HC emission inventory can also be
estimated. Determining the reduction in the annual HC
inventory for any given year is a relatively straightforward
calculation involving the annual -gasoline consumption of
vehicles employing onboard controls, the emission factor, and
the in-use control efficiency of those vehicles. The annual
C-16
-------
-13-
gasoline consumption of controlled vehicles is a function of
their total registrations, fuel economy, and the annual miles
of travel. These can be expressed mathematically as shown
below.
IR = (EF)(NSEFF)
(453.6) (2,000)
REGxzSRxZNTAMPxZVMTxz
z=.
MPG
xz
REGyz SRyzNTAMPyzVMTyz
MPG
yz
The variables are the same as identified above, and as noted
below:
x = LDVs
y = LDTs
z = time (years)
IR = annual HC inventory reduction (tons)
REG = new registrations of gasoline-fueled LDVs or LDTs in
each year z=l,n
SR = new vehicle survival rate of gasoline-fueled LDVs or
LDTs in each year
The values for these variables are given in Appendix B.
Working through the calculations above, the following
annual inventory reductions are projected from all in-use LDVs
and LDTs with effective vapor recovery systems.
1988
41,200
Annual Reductions (tons)
1989 1990 1995
2000
77,500
108,400
213,300
257,500
One can see that as a greater portion of the LDV and LDT fleet
employs onboard control, the annual reduction in refueling
emissions becomes substantial.
V.
Costs of Onboard Vapor Recovery
Two new sources of information on the costs of onboard
vapor recovery hardware have become available since the
C-17
-------
-14-
preparafcion of the last estimates shown in Appendix A. The
first is a June 1983 draft report entitled "Manufacturing Costs
and Retail Price Equivalent of On-Board Vapor Recovery System
For Gasoline-Filling Vapors,n prepared by LeRoy Lindgren under
contract to API. The second is a January 1984 cost estimate
presented by API in their final report on the cost comparison
for Stage II versus onboard control of refueling emissions.
The information contained in Lindgren's report was one input
used by API in their most recent cost estimates for onboard.
No updated cost .estimates from the auto industry were available
for this analysis.
In this section of the report, the Lindgren hardware cost
estimates will be reviewed and discussed first. This will be
followed by a discussion of the onboard cost estimates
developed by API and an update of the estimate of the cost of
an onboard vapor recovery system for a current technology LDV
or LDT. This section will close with a discussion of the sales
impact of an onboard control requirement.
A.
Lindgren Report
The Lindgren report to API provides an estimate of both
the manufacturer (or vendor) cost and retail price equivalent
(or customer cost) of a complete onboard control system. The
estimates of these two costs are $12.95 and $29.85,
respectively. Tables 1 through 6 of Appendix C (taken from
Lindgren's draft report) contain the bases for these costs.
Lindgren estimated hardware costs for the system
demonstrated by API in 1978. This system is shown in Appendix
A/ Figure A-l. This system was a fillpipe seal, additional
charcoal canister, and separate plumbing for the evaporative
emissions and onboard recovery systems. Lindgren attempted to
update these designs for changes in LDV engine and emission
control technology which have occurred since 1978. However,
this was not done properly in every case, and costs for
components already on current technology vehicles were
attributed to the cost of an onboard vapor recovery system.
For example, costs were included for a leaded fuel restrictor
and modifications related to the electronic control unit, both
which would be present on current vehicles.
Although most of Lindgren's component manufacturing costs
appear reasonable, there are two other major deficiencies in
the analysis. First, arithmetic errors were made in several
places in the analysis, and an error was made in calculating
the costs after corporate and dealer markups were added. A
markup factor of 2.3 was used instead of 1.8 as specified by
Lindgren in his report. As was mentioned above, Lindgren
C-18
-------
-15-
estimated a customer cost o-f $29.85. Correcting these errors
and applying Lindgren's 1.8 standard markup factor brings the
customer cost down to $24.06. Second, based on previous
analyses of Lindgren's cost methodology, standard absorbed
overhead and profit absorption rates appear to be used at the
corporate and dealer levels, rather than incremental rates.
This results in a substantial overestimation of the
contribution of overhead and profit to onboard control costs.
As will be discussed below, it is believed that an incremental
approach to corporate and dealer overhead and profits is
appropriate for emission controls, resulting in a markup factor
of 1.27 rather than 1.8. Using this incremental markup factor
brings Lindgren's estimate to $17.72.
Thus, Lindgren's cost estimates cannot be used directly
here, but will have to be modified to include only the costs of
components incremental to those already on current technology
LDVs and to more accurately reflect appropriate corporate and
dealer markups. This process'is described in the next section,
after a review of the cost estimates released by API.
B.
API Cost Estimates
In their recent final report comparing Stage II and
onboard costs, API presented their updated cost estimates for
onboard controls.[13] API did not present
component-by-component cost estimates, but only a fleetwide
average cost of $13.43. This estimate included different
canister sizes for LDVs and LDTs and the need for two canisters
on some vehicles. The fleetwide average estimate was
calculated using a cost of $12.07 for LDVs or lighter LDTs with
one canister, $14.47 for LDVs or lighter LDTs with two
canisters, and $20.87 for all heavier LDTs. These costs were
then weighted 70 percent, 20 percent, and 10 percent,
respectively, representing the projected portions of the total
vehicle population. When system development and certification
costs are added, this cost rises to $15.26 per vehicle ( 1983
dollars).
The API cost estimates did not include a retail markup
because they were not certain about what markup figure was
appropriate or how the vehicle manufacturer or dealer might
choose to absorb or pass on costs. If the markup factor of
1.27 is applied, a fleet average cost of $19.38 per vehicle is
obtained. In the section which follows directly, it will be
seen that the marked up API figure is in the range of the
updated estimate of this report.
C-19
-------
-16-
C. Updated Estimate
1. Hardware Costs
The cost estimates here are based on an integrated
evaporative emissions and onboard control system as opposed to
the separate systems on which the Lindgren and API cost
estimates are based. This design is expected to be the
approach preferred by the manufacturers because it makes
optimal use of limited underhood space, simplifies the design,
and reduces cost. The key feature of this design is that one
large charcoal canister can be used for evaporative emissions
and onboard control rather than two separate canisters.
The first step in developing the updated cost estimate was
to decide on what components would make up the system. The
components selected were mostly the same as those used in the
API demonstration program and priced by Lindgren. However,
there are several important differences. First, as was
mentioned above, an integrated onboard/ evaporative emissions
control system was assumed, thus eliminating obvious
redundancies , between the two systems. Second, the cost of a
pressure relief valve was included which might be necessary as
discussed previously in Section II. D. And, third, the
components which are present on current vehicles but were not
present on the 1978 vehicles used by Lindgren were excluded.
Once the components of the system were determined, vendor and
retail price equivalent cost estimates were developed using,
and in some cases modifying, the manufacturing cost estimates
provided by Lindgren.
The expected components and their costs are summarized in
Table 1. The vendor costs include material, direct labor, and
direct overhead and have been multiplied by a factor of Ii4 to
account for indirect overhead and profit at the vendor levels.
The 1.4 factor for vendor allocation and profit was taken from
Lindgren's methodology and represents a standard absorbed
overhead rate and rate of return for this industry. These full
rates are appropriate here because the production of the
emission control equipment is the primary business activity for
the vendor and is not incremental in nature.
These vendor costs were then multiplied by 1.27 to
estimate the corresponding retail price equivalent, accounting
for corporate and dealer overhead and profit. Lindgren applied
a factor of 2.3 to account for these factors, though this
appears to be an error, since his own methodology specifies a
factor of 1.8. The 1.8 factor appears, again, to include a
standard absorbed overhead rate for both manufacturer and
dealer and standard profit margins for both. These figures are
C-20
-------
-17-
Table 1
Onboard Vapor Control Hardware Costs
(1983 dollars)
Component or Assembly
Charcoal Canister LDV/(LDT)
Purge Control Valve
Liquid Vapor Separator
Fillpipe Seal
Pressure Relief Valve
Hoses/Tubing
Miscellaneous Hardware
Vehicle Assembly
Systems Engineering/Certification
Incremental Costs
Vendor
$3.99/(7.83)
0.74
0.71
1.12
0.44
1.90
0.40
—
--
Retail Price
$5.07/(9.94)
0.94
0.91
1.42
0.56
2.41
0.51
IsOO
0.50
LDV Totals:
LDT Totals:
Vendor
Vendor
$9.30 Retail $13.32
$13.42 Retail $18.19
C-21
-------
-18-
not appropriate here because adding emission control equipment
is only incremental to the primary business of assembling
automobiles, and overhead and applied assets are not entirely
variable with respect to vendor cost, but have significant
fixed components. This is particularly true for the dealer,
who would experience almost no effect due to the added
equipment. The 1.27 factor is the result of an incremental
analysis of corporate and dealer overhead and profit which was
performed as part of a recent EPA mobile source regulatory
analysis for LDVs and LDTs.[14]
The size of the carbon canister in Table 1 is that
associated with a fuel tank which would give an in-use driving
range of about 300 miles. Using the in-use fuel economy
projections of Appendix B (Table B-3) for 1985-90, LDVs would
require an average fuel tank size of 10-13 gallons, LDTs would
require an average fuel tank size of 14-18 gallons. To be
conservative, in each case, the higher end of the ranges in
fuel tank sizes was used to size the canisters.
As shown in Table 1, an onboard vapor recovery system is
expected to carry a consumer cost of $13.32 for LDVs and $18.19
for LDTs. Those LDTs using dual-fuel tanks (approximately 20
percent) may require two separate onboard control systems for a
total cost of $36.38. This is a conservative assumption since
costs could likely be reduced by using one large charcoal
canister rather than two separate canisters.
A fleetwide estimate for all LDVs and LDTs can be
determined by sales weighting the costs given above. Using the
projected sales for 1988 from Appendix B (Table B-3), and
assuming 20 percent of LDTs have dual-fuel tanks, the fleetwide
average cost is calculated to be $15.08 as shown below. For
future calculations this cost will be rounded to $15 per
vehicle.
(10.582M) ($13.32) + (2.768M) ( (.8) ($18.19) + (.2) ($36.38))=$15.08
13.35M
This estimate of $15.08 is comparable to API's estimate of
$19.38 after application of the 1.27 markup factor and
Lindgren's estimate of $17.72 after corrections and using the
1.27 incremental markup factor. The main reason for the
difference between this estimate and those developed by API and
Lindgren is because EPA assumed an integrated
onboard/evaporative emission control approach as opposed to two
separate systems.
2. Differences Between Past and Current EPA Cost
Projections"———
EPA's February 1980 report projected a fleet average cost
of $19.70. When inflated to 1983 dollars, this cost becomes
C-22
-------
-19-
$26 per vehicle. There are three reasons for the overall
decrease of $11 between this cost estimate and the previous
estimate.
First, the 1980 projection used a 1.8 retail markup
factor. As discussed above, it is now believed that a 1.27
markup factor is more appropriate; this change alone accounts
for approximately 70 percent of the cost difference.
The second reason is related to changes in system mixes.
The 1980 projection assumed higher costs in some cases due to
the use of two canisters rather than one larger canister, or
due to a manufacturer not choosing an integrated
onboard/evaporative emissions control approach. This accounts
for another 23 percent of the difference.
Third, there have been changes in the components
anticipated to make up an onboard control system and the prices
for some components (notably the fillpipe seal). This accounts
for the remaining 7 percent of the cost difference.
3.
Fuel Economy Impacts
As was stated in the previous EPA report (Appendix A), the
implementation of an onboard vapor recovery requirement would
not be expected to impact LDV or LOT fuel consumption.
Hydrocarbons retained by the onboard canister represent about
0.1 to 0.2 percent of vehicular fuel consumption. The use of
this fuel by the engine could thus be expected to decrease fuel
consumption by this amount. However, the additional fuel
needed to transport the added weight of the onboard system is
also in this range. Thus, no net change in fuel consumption is
expected.
4.
Overall Cost Estimate
As discussed in the previous two sections, the updated
LDV/LDT cost estimate is about $15 per vehicle and there is
adequate explanation as to why this cost is well below that of
February 1980. However, there are still reasons , to believe
that the total cost of onboard control could be somewhat
greater than $15 per vehicle.
One, this figure includes primarily hardware cost and
excludes any costs associated with possible fuel tank
modifications, modifications to the vapor line and rollover
check valve between the fuel tank and the vapor canister,
modifications to make the fillpipe more tamper-resistant, and
general packaging costs to fit the integrated
onboard/evaporative emissions control system into the vehicle.
C-23
-------
-20-
Also, there may be some cost related to manufacturer-specific
electronic control unit (ECU) modifications. For example, some
manufacturers may desire to use specific canister purge cycles
which may require reprogramming or modification of their ECUs.
Finally, since the actual pressure relief valve discussed above
has not been identified, there is some uncertainty in the cost
for that component.
Two, except for the allowance of a dual system for LDTs
with two fuel tanks, the system considered herein is somewhat
ideal. Completely integrated onboard/evaporative emission
control systems are assumed in every case, and this simply may
not be possible. For reasons of canister production economies
of scale, underhood packaging restrictions, or for unique
vehicle models, manufacturers may choose a non-integrated two
canister system similar to those considered by API. As was
discussed above, a non-integrated system would increase the
costs over those shown in Table 1.
Three, in the final analysis, the actual canister size and
purging system will depend on the details of the test procedure
implemented to measure compliance with an onboard vapor
recovery requirement. Factors such as the degree of
interaction between the evaporative emissions and onboard test
procedures and whether the charcoal canister would have to be
purged during the exhaust emissions test will affect the size
of the charcoal canister and the complexity of the purging
system. These in turn would affect the overall cost of the
onboard system.
To account for these and other potential costs, a range of
$15-25 per vehicle will be used rather than the single cost of
$15 per vehicle. While the final cost is expected to be.closer
to $15 rather than $25, the use of $25 as an upper limit will
allow the sensitivity of any subsequent decisions to this cost
to be addressed.
D.
Impact On Sales of LDVs and LDTs
An average purchase price increase of $15 to $25 is
expected to have no discernible impact, on the sales of LDVs or
LDTs and, therefore, no effect on the profitability of the
companies comprising the regulated industry. The "own price
elasticity of demand" for LDVs and LDTs (that ignoring any
crossover purchases in other vehicle classes) is approximately
-1.0, which means that for each 1 percent increase in price,
sales drop 1 percent. With the price of an average new LDV or
LDT now exceeding $10,000, a $15 to $25 first price increase
would be predicted to decrease sales by no more than 0.15 to
0.25 percent. However, there is some question whether the
C-24
-------
-21-
elasticity of demand is even meaningful in measuring the sales
impact of a $15 to $25 increase. Such an increase would tend
to be lost in the annual price increases occurring at the time
of model year introduction.
Furthermore, onboard controls are not expected to affect
operating and maintenance costs, nor significantly affect the
owner's experience of refueling. Thus, there should be no
non-economic resistance which will affect sales or
satisfaction. In the long term, an onboard vapor recovery
requirement should have no perceptible impact on the sales or
profitability of either the manufacturers or dealers.
VI. Cost Effectiveness
The cost effectiveness of onboard control can be
calculated using the LDV and LDT in-use emission reduction
equations developed in Section IV and the range in the average
costs of control calculated in Section V. The in-use emission
reduction varies with each model year's vehicles depending on
the fuel economy, and the average cost varies somewhat based on
relative sales of LDVs and LDTs. The 1988 model year will be
used here, since it is possibly the first model year in which
an onboard requirement could be implemented.
Referring to Appendix B (Table B-3) , the 1988 LDV and LDT
fuel economies are 26.30 and 19.28 mpg respectively, and the
sales are 10.582 and 2.768 million, respectively. Using these
fuel economy figures, the lifetime reduction for LDVs is 0.0178
tons and for LDTs the lifetime reduction is 0.0264 tons. Sales
weighting these figures, the fleetwide average lifetime tons
reduction is 0.0196 tons. Dividing these figures into the
range of fleet average weighted cost of $15-25 per vehicle,
yields an average lifetime cost-effectiveness value of $766 to
$1,277 per ton. As shown in Table 2, this cost-effectiveness
value falls in the range of values for other mobile source
related HC control strategies, though nearer the end.
On an annual basis, the cost effectiveness is somewhat
larger. Using a 10-year vehicle life for LDVs and LDTs, a 10
percent discount rate, and assuming payment in mid-year,
annualization of the $15 to 25 lifetime cost yields an annual
cost of $2.34 to 3.90. Assuming annual mileage is constant for
the ten years, the 0.0196 fleet-weighted lifetime tons
reduction converts to 0.00196 tons annual emission reduction.
The annual cost effectiveness is then about $1,194-1,990 per
ton. The simplifying assumption of constant annual mileage
results in a slight overestimation of this figure, since annual
mileage is higher early in the vehicle's life. Nevertheless,
this provides a valuable additional way of looking at the cost
C-25
-------
-22-
Table 2
Cost Effectiveness of Mobile Source HC
Control Strategies (1983 $/ton)[1]
Control Strategy
Cost Effectiveness
HDGV Evaporative Control
HDGE Useful Life
LDT Useful Life
LDT Statutory Standard
HDDE Statutory Standard
HDDS Useful Life
Interim High-Altitude Standards
Onboard Vapor Recovery (with evap. benefits)
LDV Statutory Standards
Motorcycle Standards
Onboard Vapor Recovery (w/o evap. benefits)
I/M
Auto Coatings
Transit Improvements
$112
$100-200
$406
$207
$319
$323
$416
$435-725
$508
$616
$766-1,277
$943
$1,301
$15,767
[1] Short ton
C-26
-------
-23-
effectiveness of an onboard requirement, especially when the
cost effectiveness of an onboard requirement is compared to HC
strategies where the cost effectiveness is calculated on an
annual basis.
This estimate of the cost effectiveness of onboard 'vapor
recovery_ only considers the emission reductions derived from
eliminating refueling losses. However, preliminary data from
EPA s emission factors program indicates that in-use
evaporative emissions appear to significantly exceed the
lu3?^3.^6 ?? SKandaKd.*K ThlS °CCUrS Pri^rily because in-usJ
fuels typically have higher volatility than the fuel specified
for certification testing and, therefore, produce larger
amounts of evaporative HC which cannot be adsorbed by the
current charcoal canisters. Preliminary estimates of the level
of these excess evaporative emissions can be made, using the
f^anrfYrre^ly availablue. fr°™ EPA's evaporative emission
factors testing program which is now in progress. This program
involves evaporative emission testing using Indolene
(certification) and commercial fuel in carbureted and
fuel-inuected^ vehicles. Based on preliminary data from this
program, it is estimated that LDVs have evaporative emissions
n^f r,fn£e °,f • °'23 to °'44 g/mi Usin9 commercial fuel and
0.16 to 0.24 g/mi using certification fuel, yielding an excess
t?nAe T^ ^.°-07 t°.°-20 g/mi. A best eatimate at ?hiS
time based on this preliminary data is evaporative emissions of
0.33 g/rai using commercial fuel and 0.20 g/mi using
TrSiJ1^!11*?"6!' f?r an excess of °*13 9/mi- Although datl
is not available for LDTs, one would expect results in the same
ranges since LDV and LDT evaporative control systems are very
similar. Simply multiplying the best estimate of these excess
?Ta?°^nJ1Ve Jeinissions by the average lifetime for LDVs and LDTs
(100,000 and 120,000 miles, respectively) and converting to
2S «yi«eAds llfetime excess emissions of 0.0143 tons for LDVs
and 0.0172 tons for LDTs. The fleet-weighted LDV/LDT per
vehicle excess would be 0.0149 tons of HC lifetime or 0.0015
tons annually.
Since refueling only occasionally coincides with the
occurrence of evaporative emissions, the larger charcoal
canister associated with an integrated onboard/evaporative
emission control system could also control these excess
evaporative emissions at little or no extra cost. Adding these
benefits to those from onboard control improves the cost
effectiveness by approximately 43 percent. if all excess
evaporative HC emissions were controlled, the lifetime cost
effectiveness of onboard control would become $435 to $725 per
ton and the annual cost effectiveness would become $678 to
$1130 per ton.
C-27
-------
-24-
As indicated above, the in-use evaporative emissions data
is preliminary as all testing has not been completed. As
additional data become available, it will be possible to make a
firmer estimate of excess in-use evaporative emissions.
However, regardless of the magnitude of excess in-use
evaporative emissions, the onboard control system does have the
potential to control a large portion of these excess
emissions. This additional HC control, if credited towards
onboard control, would improve its cost effectiveness. While
not central to- the issue of controlling refueling emissions
through onboard vapor recovery, this potential for control of
excess in-use evaporative emissions provides an additional
perspective on the value of implementing an onboard vapor
recovery requirement.
VII. Leadtime Requirements
If an onboard vapor recovery requirement were implemented,
it is estimated that approximately 24 months of leadtime would
be necessary before the systems could be required on production
LDVs and LDTs once a rule is promulgated. This leadtime
estimate is based on engineering judgment, and on leadtimes
necessary in similar, previous EPA rulemakings. These include
the original 1978 6.0 g/test LDV/LDT evaporative emission
standard which was implemented in just one year, the 1985 HDGV
evaporative emission standard which will be implemented with
two years of leadtime, and the 1981 2.0 g/test LDV/LDT
evaporative emission standard which was also implemented in two
years. The two-year leadtime estimate to implement an onboard
vapor recovery program is based on the following considerations.
A program to comply with an onboard requirement would
first include approximately six months for the
vendors/manufacturers to develop and optimize working prototype
systems applicable to all of their different vehicle models.
Next, initial verification of the fillpipe seal and pressure
relief valve durability could be conducted in two months or
less under laboratory conditions. However, purge system
optimization and optimization and proveout of the integrated
onboard vapor recovery/evaporative emissions control system
would require some vehicle testing, as would verification of
the efficiency and durability of the fillpipe seal and pressure
relief valve. This vehicle testing would require four to six
months, based on manufacturer estimates for similar in-vehicle
testing programs. Thus, prototype testing and proveout is
estimated to take 12 to 14 months to complete.
Although many of the components of an onboard system would
be "off-the-shelf" or readily fabricated from existing
production tooling, some tooling changes would be necessary for
some components, such as larger charcoal canisters. However,
C-28
-------
-25-
the critical items in terms of production tooling are the
fillpipe seal and pressure relief valve. If the fillpipe seal
and^ pressure relief valve used are some form of currently
available component, then only the question of capacity
exists. Capacity is necessary to meet long term demand in
excess of 13 million units per year. If. the vendors and
manufacturers ultimately settle on prototype designs which
would require significant tooling changes or completely new
production, or if current production capacity is insufficient,
then longer tooling leadtimes may be required.
In any event, commitments leading to production tooling
changes could probably be made after the initial laboratory
verification of the fillpipe seal and pressure relief valve
durability. If vendors/manufacturers are able to use seals
similar to those used in the 1978 vehicle demonstration program
and an acceptable pressure relief valve is available, then
total tooling leadtimes of three or four months would be
necessary. If fillpipe seals and pressure relief valves must
be procured from modified tooling, then leadtimes of- six to
eight months are reasonable. If new tooling must be developed,
then leadtimes for tooling will require approximately 12 months
or perhaps longer. Thus, the range for tooling leadtimes is
three months to one year or more, depending on the source of
the fillpipe seals and pressure relief valve. Assembly line
tooling changes would be handled during normal model year
changeover, and thus would have no effect on this estimate.
Finally, some time would be required to allow for the
normal EPA certification process. -It normally requires a
manufacturer 10 to 12 months to certify its entire product
line.[15]
Given these estimates of the leadtime necessary for
development, laboratory testing, in-vehicle testing, tooling,
and certification, Figure 1 shows how these different estimates
were put together to arrive at a leadtime estimate of two
years. The critical path on this figure is 6 months for
development, 2 months for laboratory testing, 4 to 6 months for
in-vehicle testing, and 10 to 12 months for certification.
Presuming that tooling commitments can be made after the
laboratory testing is concluded, tooling is not a critical path
even if the fillpipe seal and pressure relief valve required
new tooling. Tooling would only become a concern if
commitments were delayed until after the completion of
in-vehicle testing (12-14 months).
In summary, a leadtime period of two years appears
reasonable to implement an onboard requirement. Of course, the
model year of implementation for an onboard requirement would
depend on when a final rule was promulgated.
C-29
-------
-26-
CN
'4-4
oo
§ i
oo
oo
M "%
0)
00
•H I
.M-l VO
ftO
oo
CO
OS
CM H
CO
0)
a
CO
ao
J- H
60
pit • Q
"3
C-30
-------
-27-
VIII. Onboard Control Versus Time
When considering the implementation of onboard controls,
it^is of value to determine how much time would be required to
gain control of a majority of the annual LDV and LOT gasoline
consumption. This, of course, depends on the vehicle scrappage
and replacement rates, the annual vehicle miles of travel and
the vehicle fuel economies. Consequently, the portion of total
LDV and LOT fuel usage (and accompanying refueling emissions)
which would be controlled as a function of time beginning with
the model year of implementation is estimated below. For this
analysis it is assumed that implementation begins in the 1988
model year.
A. Total Fuel Consumption
To determine the portion of total LDV and LDT fuel
consumption controlled as a function of time the controlled and
total LDV and LDT fuel consumption must be estimated by
calendar year. A total gasoline consumption by a specific
model year's vehicles in a given calendar can be derived using
the expression given below;
GC = (REG)(SR)(VMT)(VMTGR)
(MPG)(ODOM)
where:
GC = gasoline consumption (gallons)
REG = new vehicle registrations for that model year
(function of model year)
SR = survival rate of new vehicles in the calendar year of
interest (function of age)
VMT = average annual mileage of the vehicles (function of
age)
VMTGR = growth rate in average annual mileage of the vehicle
(function of model year)
MPG = new vehicle in use fuel economy (function of model
year)
ODOM = Usage pattern factor to account for the different mix
of urban/rural driving and average daily mileage on
average in-use fuel economy (function of age)
Data for the input parameters described above is provided
and referenced in Appendix B. However, a few explanatory notes
C-31
-------
-28-
are appropriate. First, the approach used here models total
LDV and LDT fuel consumption using twenty model years of LDV
and LDT registrations (e.g., 1988 fuel consumption would be
modeled using registrations from 1988 to 1969 inclusive).
While it' is recognized that there are a small number of LDVs
and LDTs older than 20 years still in-use, their contribution
to total fuel consumption is relatively insignificant due to
their low registrations and average annual VMT. Second, Table
B-2 contains average annual VMT data for LDVs and LDTs. This
data is applicable for pre-1982 model year LDVs and LDTs. For
1982 and later LDVs and LDTs, average annual VMT was projected
to increase at a rate of 0.8 percent per year for LDVs and 0.4
percent per year for LDTS.[16] Last, calculation of fuel
consumption included a usage pattern factor (ODOM) to account
for the fact that the mix of urban/rural driving and the daily
vehicle miles of travel both change as an LDV ages, and this
affects the in-use fuel economy in any • given year of a
vehicle's life. This applies to LDVs only.[16]
Given this data, total LDV and LDT fuel consumption in any
given calendar year can be calculated by simply determining the
fuel consumption of each model years LDVs and LDTs in the year
of interest and summing the consumption from each model year's
LDVs and LDTs to derive a total. This method of calculation is
shown mathematically in the expression given below:
GC
20
5
y
(REG
v
v
(VMT GRV^X)
(MPG,r v)(ODOM
V f X *
,) (VMT GR
(MPG )
T. ,X
v a LDVs, t - LDTs, x = model year, y = years,
z = vehicle age
In this method of calculation y = 1 would be the calendar year
of interest, and all data used would begin with that year and
then going back 20 years. Total fuel consumption would be
determined by summing the consumption of the most recent 20
model years LDVs and LDTs in the calendar year of interest.
B.
Controlled Fuel Consumption
Calculation of the controlled fuel consumption requires
only two additions to the discussion given above. First, since
controlled consumption is not assumed to begin until 1988, the
period over which controlled consumption will be calculated
varies from 1 model year in 1988 to' 13 model years in 2000 (or
presumably longer were more data available with which to
calculate controlled consumption after 2000) . Second, as was
C-32
-------
-29-
discussed previously, tampering with the fillpipe or
evaporative emission system will eliminate the control
effectiveness of the onboard vapor recovery system of those
vehicles. Thus, the fuel consumption of tampered LDVs and- LDTs
must be factored out. This can be accomplished using the NTAMP
factor described previously. NTAMP is a function of mileage
and is different for LDVs and LDTs. For any given mileage in
the life of an LDV or LOT, NTAMP = 1-TAMP, where TAMP is the
percentage tampering calculated using the projected composite
tampering rate equations given in Section III.B.3. The mileage
used in the tampering rate equation for each model years LDVs
and LDTs includes the growth rate decribed above for LDVs and
LDTs.
Controlled fuel consumption in any calendar year is then
the sum of the fuel consumption of each model year's
non-tampered LDVs and LDTs in the calendar year of interest.
This method of calculation is shown mathematically below. The
only difference between this and the previous expression is the
limits on the summation and the inclusion of the tampering
factor.
Controlled =
GC
( z)
x> (°DOMv,
(MPG )
u, x
C.
Discussion of Results
The portion of the total LDV and LOT fuel consumption
controlled in any calendar year, 1988 or later, can now be
calculated. Figure 2 compares LDV and LDT gasoline consumption
which would be controlled by an onboard vapor recovery
requirement to total LDV and LDT gasoline consumption, assuming
onboard controls were first introduced with 1988 model year
LDVs and LDTs. Table 3 is a tabular summary of the graphical
information presented in Figure 2. This data shows that
control of 50 percent of all LDV and LDT fuel consumption would
be achieved 5 to 6 years after introducing an onboard vapor
recovery requirement and control of more than 84 percent of all
LDV and LDT gasoline consumption would be achieved by 2000 (13
years after control is implemented). Ttfithout tampering,
control in the year 2000 would exceed 92 percent; control of
approximately 8.5 percent of consumption is lost due to
tampering.
In terms of the separate LDV and LDT fleets, control of 50
percent of LDV fuel consumption would be achieved in about 5
years and by 2000 89 percent of LDV gasoline consumption would
C-33
-------
-30-
Table 3
Gasoline Consumption of Non-Tampered Vehicles
With Onboard Emission Control
Compared to Total Vehicle Fuel Consumption
LDV Gas Consumption (billions of gallons)
Year
1988
1989
1990
1991
1992
1993
1994
1995
1995
1997
1998
1999
2000
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
Total LDV
Gas Consumption
49.0
48.0
46.9
45.9
45.0
44.0
43.1
42.3
41.6
41.0
40.5
40.2
40.0
LDV Gas Consumption
Controlled Vehicles
6.0
11.3
15.8
19.8
Percent Control
12.2
23.5
33,
43,
,7
,1
23.2
,1
,4
,3
26,
28,
30,
31.9
33.1
34.2
34.9
35.5
51.6
59.3
65.9
71.6
76.7
80.7
84.4
86.8
88.8
LPT Gas Consumption (billions of gallons)
Total LDT
Gas Consumption
24.4
24.0
23.4
23.0
22.7
22.4
22.2
22.1
22.1
22.0
22.0
22.0
22.2
LDT Gas Consumption
Controlled Vehicles
2.4
,4.5
6.3
8.0
9.5
10.9
12.1
13 ,.2
14,2
15.0
15,.8
16,.5
17.0
Percent Control
9.8
18.8
26.9
34.8
41.9
48
7
54.5
59.7
64
68
3
2
71.8
75.0
76.6
C-34
-------
-31-
Table 3-Cont'd
Gasoline Consumption of Non-Tampered Vehicles
With Onboard Emission Control
Compared to Total Vehicle Fuel Consumption
LDV and LPT Gas Consumption (billions of gallons)
Total LDV & LOT
Year Gas Consumption
1988 73.4
1989 72.0
1990 70.4
1991 69.0
1992 67.7
1993 66.4
1994 65.3
1995 ' 64.4
1996 63.6
1997 63.0
1998 62.5
1999 62.2
2000 62.2
LDV &
LOT Gas Consumption
Controlled Vehicles
8.4
15.8
22.1
27.8
. 32.7
36.9
40.4
43.5
46.1
48.2
49.9
51.4
52.5
Percent Control
11.4
21.9
31.4
40.3
48.3
55.6
61.9
67.5
72.5
76.6
79.8
82.6
84.4
C-35
-------
80 -
70 .
60 .
50 .
to
o
•H
H
H
•H
(0
c
o
a
e>
40
30
20
10
-32-
Figure 2
Controlled vs. Total Gasoline
Consumption for LDVs and LDTs
1988 - 2000
Total Consumption
Controlled Consumption
Controlled Consumption
Controlled Consumption
LDV & LDT
LDV
LDT
.-+—'
#
/
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
C-36
-------
-33-
be controlled. For LDTs, 50 percent control would require 6
years and 77 percent control would be achieved by 2000.
This method of determining the time for achieving control
of refueling emissions differs slightly from that used by
Lindgren,[17] which estimated the fraction of the vehicle
population which would be equipped with onboard gasoline vapor
controls over time. The fraction of dispensed gasoline
controlled is more appropriate than the fraction of vehicles
controlled, since refueling loss emissions are a function of
the amount of gasoline dispensed and not simply a function of
the number of vehicles in the fleet.
IX. Conclusions
The data from the API demonstration program and the
manufacturers' previous comments both indicate that onboard
control of refueling emissions from LDVs, LDTsr and lighter
weight HDGVs should be technologically feasible using a
fillpipe seal and an integrated onboard/evaporative emission
control system. Onboard control should also be feasible for
the heavier HDGVs, but the systems used on heavier HDGVs would
be somewhat more complex and costly. The implementation issues
for the control of heavier. HDGV refueling emissions could be
worked out in a manner similar to the approach used in the
recent HDGV evaporative emissions final rule, so control of
virtually all of the gasoline-fueled motor vehicles may be
possible. Implementation of an onboard requirement should have
a negligible impact on the vehicle's exhaust emission levels.
An in-use control efficiency of 98 percent is expected,
with negligible deterioration for a well-maintained vehicle.
Using the tampering rates expected in the late 1980's and
beyond, owner tampering with the filler neck restrictor and the
charcoal canister could reduce the average lifetime efficiency
to 91.8 percent for the sales-weighted fleet of LDVs and LDTs^
Using 1988 projected fuel economies for LDVs and LDTs, the
fleet average lifetime reduction in refueling HC emissions is
.0196 tons per vehicle.
An integrated onboard/evaporative emission control system
is expected to carry a fleet average cost of $15 to $25 per
vehicle, although the average . should be nearer $15.
Implementation of an onboard requirement would not increase
lifetime operating or maintenance costs. At $15-$25 per
vehicle, an onboard requirement would have no perceivable
impact on manufacturer or dealer sales.
Using the costs and emission reduction benefits mentioned
above, the sales-weighted lifetime cost effectiveness for
C-37
-------
-34-
onboard control is $766 to $1,271 per ton of HC controlled.
The annual cost effectiveness is $1,194 to $1,990 per ton of HC
controlled.
The larger charcoal canister of an integrated onboard/
evaporative emissions control system could potentially control
excess in-use evaporative emissions. If the preliminary best
estimate of these benefits is added to those achieved by
onboard control, the lifetime cost-effectiveness values fall to
$435 to $725 per vehicle and the annual cost effectiveness
becomes $678 to $1130 per vehicle.
An onboard requirement could be implemented two years
after promulgation of a final rule. Control of refueling
emissions from 50 percent of the total annual nationwide LDV
and LDT gasoline consumption could be achieved in five years.
Control of refueling vapors from more than 84 percent of total
annual nationwide LDV and LDT gasoline consumption could be
achieved by 2000.
C-38
-------
-35-
References
1. "On-Board Control of Vehicle Refueling Emissions
Demonstration of Feasibility," API Publication No. 4306,
October 1978.
2. "Recommendation On Feasibility For On-Board
Refueling Loss Control," U.S. EPA, OMSAPC, February 1980.
3. See reference 1, p. 25.
4. Letter, F. L. Voelz, ARCO to E. P. Crockett, API,
January 14, 1982, and follow-up telephone conversation between
M. Reineman, U.S. EPA and F. L. Voelz, ARCO, August 18, 1983.
5. "Staff Report, Issue Analysis - Final Heavy-Duty
Engine HC and CO Standards," U.S. EPA, OANR, QMS, ECTD, SDSB,
March 1983.
6. "Transportation Energy Data Handbook," Sixth
Edition, ORNL-5883, Oak Ridge National Laboratory, 1982.
7. "Motor Vehicle Tampering Survey - 1982," U.S. EPA,
National Enforcement Investigations Center, Larry Walz,
EPA-330/1-83-001, April 1983.
8. "Draft Mobile 3 Documentation," data provided by
Lois Platte, U.S. EPA, QMS, February 14, 1984.
9. "A Report to the Legislature on Gasoline Vapor
Recovery Systems For Vehicle Refueling at Service Stations,"
California Air Resources Board, March 1983.
10. "Compilation of Air Pollutant Emission Factors,
AP-42, Supplement 9," U.S. EPA, OAQPS, July 1979.
11. "Trends in Motor Gasolines: 1942-1981", E. Shelton,
et al, U.S. Department of Energy, DOE/BETC/Rl-82/4, June 1982.
12. "Decision: Vapor Recovery Control Strategy,"
General Motors Corporation briefing to EPA, February 3, 1984.
13. "Cost Comparison For Stage II and On-Board Control
of Refueling Emissions", American Petroleum Institute,
January 1984.
14. See for example the Regulatory Analysis and Summary
and Analysis of Comments prepared in support of the light-duty
diesel particulate regulations for 1982 and later model year
light-duty diesel vehicles. Both are available in Public
Docket No. OMSAPC 78-3.
C-39
-------
-36-
References (cont'd)
15. Trap Oxidizer Feasibility Study", U.S. EPA, OANR,
OMSAPC, ECTD, SDSB, March 1982.
16. "The Highway Fuel Consumption Model - Ninth
Quarterly Report, prepared by Energy and Environmental
Analysis, Inc., for U.S. Department of Energy, February 1983.
17. "Manufacturing Costs and Automotive Retail Price
Equivalent Of On-Board Vapor Recovery System For
Gasoline-Filling Vapors," Leroy H. Lindgren, Consultant, Draft
Report, June, 1983.
C-40
-------
APPENDIX A
Recommendation on Feasibility for
Onboard Refueling Loss Control
C-41
-------
Appendix A
February 1980
Recommendation on Feasibility
for Onboard Refueling Loss Control
NOTICE
Technical Reports do not necessarily represent final EPA decisions
or positions. They are intended to present technical analysis of
issues using data which are currently available. The purpose in
the release of such reports is to facilitate the exchange of
technical information and to inform the public of technical devel-
opments which may form the basis for a final EPA decision, position
or regulatory action.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air, Noise and Radiation
U.S. Environmental Protection Agency
C-42
-------
I. Introduction
Gasoline refueling losses estimated to be in the range of 4-5
g/gallon, can be controlled by use of control equipment at the
service station (Stage II control) or by use &f control equipment
in the vehicle (onboard control). As required by the 1977 amend-
ments to the Clean Air Act, the Emission Control Technology Divi-
sion (ECTD) of EPA has reviewed and analyzed available data on the
feasibility and desirability of onboard refueling loss control
which will be discussed in this report.
II. Summary of Conclusions and Recommendations
Several hardware demonstrations and paper studies, Ref. 1, 2,
have been conducted to determine the technical feasibility and cost
effectiveness on onboard refueling loss control. Much of the
current information is from the American Petroleum Institute (API)
onboard demonstration program, Ref. 3. Other current information
was obtained from motor vehicle manufacturers in response to a June
27, 1978 Federal Register (43FR 27892) request for relevant infor-
mation. • These demonstrations and analyses deal with .the state-of-
the-art emission control technology.
Analysis of this information supports the following conclu--
sions:
1. Onboard refueling loss control is feasible for light-
duty vehicles.
2. The most probable control system uses hydrocarbon adsorp-
tion on charcoal (the same strategy that is, used for evaporative
emission control).
3. Control effectiveness can be as high at 97%, but this
depends especially upon the vehicle fillpipe/service station nozzle
interface and upon control technology design.
4. An analysis of data from three fillpipe/nozzle concepts
(fillpipe seals, nozzle seals, and combination fillpipe/nozzle
seals) shows that the effectiveness of all three concepts is
approximately equal. Durability effects have not been extensively
evaluated, especially for the nozzle seal concept.
5. A vapor/liquid pressure relief valve is required to
protect the integrity of the vehicle fuel tank during the refueling
process. The pressure relief valve can be designed to function on
the fuel nozzle, or it may be incorporated as part of the fillpipe
seal mechanism, which would be sealed-off by the fuel cap during
vehicle operation. Durability effects have not been evaluated for
either the fillpipe or nozzle pressure relief. ECTD recommends
that the fillpipe/nozzle seal and pressure relief be located on the
vehicle if onboard controls are required.
C-43
-------
-2-
6. The consumer cost for light-duty vehicle refueling loss
control systems are estimated to be $17/vehicle. The $17 estimate
does not include costs for a seal or pressure relief. Cost for a
seal and pressure relief, if used on the vehicle, is estimated to
be about $2.70. The cost of a seal on the nozzle should be the
same as the cost for a Stage II nozzle. Except for the as yet
undefined durability of the interface seal no maintenance costs are
expected.
7. The feasibility of controlling refueling loss emissions
from gasoline fueled trucks and diesel fueled vehicles has not been
evaluated to date. Technical feasibility and cost effectiveness of
controlling these sources should be determined.
8. Minor increases in CO exhaust emissions can probably be
controlled by minor changes to either the refueling loss control
system or to the exhaust emission control system. The ability to
certify a vehicle to a 3.4 g/mi CO standard to 50,000 miles should
not be seriously impaired.
9. The use of a bladder in the fuel tank appears to be a
viable alternative control strategy, but some problems exist and
technical feasibility is yet to be demonstrated.
*
10. Considering the lead time needed for regulation develop-
ment and review within EPA and the lead time required by the
industry for development and application of technology, implemen-
tation of onboard controls cannot occur before 1984.
ECTD recommends that the choice between onboard control and
Stage II control of refueling loss emissions be based upon air
quality considerations and the relative cost effectiveness of the
two strategies for the same overall level of control.
It is recommended that methods of reducing the cost of onboard
refueling control systems be examined by considering tradeoffs
between control system capacity and cost. It may be possible to
sacrifice some capacity that is only required under '. infrequent
conditions and achieve proportionately more significant cost
savings.
The feasibility and desirability of control of refueling
losses from light and heavy-.duty gasoline fueled trucks and from
diesel fueled vehicles should be considered. EPA should support
the development of the bladder tank alternative for. refueling loss
control strategy. If regulations are to be developed for onboard
refueling loss control, a certification test prpcedure must be
developed.
C-44
-------
-3-
III. Review of Available Information
The data and information summarized in this section are based
on material submitted to EPA by the American Petroleum Institute
and information received in response to a request for information
(43FR 27892) published on June 27, 1978. The API material, Ref. 3,
is the result of their most recent study to assess onboard techni-
cal feasibility and compare the cost effectiveness of onboard
refueling controls and Stage II controls. This study was initiated
at the urging of EPA. Respondents to the Federal Register notice
included^ General Motors, Ford, and AMC. The API, GM, and Ford
information contain data from tests with onboard control hardware.
All respondents, with the exception of AMC, submitted information
on the cost and the desirability of onboard control systems.
1.
API Onboard Study
The API Onboard Control Study was structured to address
questions regarding onboard feasibility which were posed to API in
a December 1977 meeting with EPA. The API study consisted of three
tasks: a vehicle concept demonstration, a fillpipe/nozzle concept
demonstration, and a cost/benefit analysis. Exxon Research and
Engineering Company and Mobil Research and Development Corpora-
tion were the API contractors for the vehicle concept demonstra-
tion. Atlantic Richfield Company was the API contractor for the
fillpipe/ nozzle concept demonstration. Exxon R & E completed the
cost/benefit analysis for API.
The vehicle concept modification task had the following design
objectives:
1) Minimum 90% overall refueling vapor recovery.
2) No significant effect on exhaust emissions.
3) No significant effect on evaporative emissions.
4) Design should be durable, practical, and safe.
The fillpipe/nozzle demonstration had the following objec-
tives:
1) 90% overall vapor control.
2) Compatible with existing vehicle population.
3) Compatible with existing Stage II nozzles.
4) Design should be durable, practical, and safe.
Test procedure guidelines for the API work were discussed at a
meeting with API on March 15, 1978. Important procedural guide-
C-45
-------
-4-
lines which resulted from" that meeting are summarized as follows:
Fuel specification: Indolene unleaded test fuel was used for
all exhaust, evaporative, and refueling loss measurements.
Dispensed fuel quantity: Test vehicles were refueled to 100%
of capacity from a condition of 10% tank capacity.
Fuel tank temperature/Dispensed fuel temperature; The dispen-
sed fuel temperature was selected to be representative of summer
refueling conditions in Los Angeles during the month of August, or
about 85*F. The fuel temperature in the tank was also selected to
be 85*F. Thus., the refueling was isothermal.*
Purge Cycle; For the purposes of the API study, the only
driving cycle which was used for purging the refueling loss can-
ister is the LA-4 cycle.
Individually, these test procedure guidelines are considered
to represent real world situations in a high oxidant forming
location, e.g., Los Angeles during the month of August. Collec-
tively, these guidelines imply that the API vehicles demonstrated
the feasibility of onboard control systems in an approximate worst
case condition. This reasoning is consistent with earlier EPA
recommenda't ions that API err on the conservative side during
their study. For example, Exxon used the following test sequence
to quantify the exhaust emissions interaction between the refueling
control system and the exhaust emission control system:
1) Load ECS (Evaporative Control System) canister to break-
through .
2)
3)
4)
through.
Condition the vehicle by driving 2 LA-4's.
Soak vehicle overnight.
Load RCS (Refueling Control System) canister to break-
5) Condition the vehicle by driving 5 to 6 simulated city
driving days (4.7 LA-4's with one hour hot soaks in between and a
diurnal at the end of the day) to consume 90% of the fuel in the
tank.
*This represents a conservative situation as survey data, Ref. 4,
show that nationwide dispensed fuel temperatures are typically
lower than tank fuel temperatures, thereby representing a vapor
shrinkage situation during the refueling process.
C-46
-------
-5-
6) Drain the fuel tank.
7) Block RCS canister line.
8) Fill tank to 40%, unblock RSC cani-ste* lines.
9) Conduct diurnal evaporative test in SHED.
10) Drain tank to 10%.
t
11) Bring fuel tank liquid and vapor to equilibrium at 85°F
(shake the vehicle to accelerate the equilibrium process).
12) Refuel the vehicle to 100% in SHED with 85°F fuel.
13) FTP
14) Hot soak evaporative test in SHED.
Obviously, these test procedures do not lend themselves to a
routine laboratory certification test procedure. They do, however/
permit an approximation of how an onboard control system would
function in a severe "real-world" situation.
A review of the three API contractor's activities is presented
below.
Exxon
Exxon assumed the responsibilty for modifying four test
vehicles. Their vehicles included the following:
• 1978 Chevrolet Caprice
1978 Ford Pinto
1978 Plymouth Volare*
1978 Chevrolet Chevette
All vehicles were designed to comply with 1978 California
exhaust and evaporative emission standards (.41 HC, 9.0 CO,
1.5 NOx, 6.0 Evap).
* Vehicle subsequently dropped from test program because of high
baseline NOx levels.
C-47
-------
The Caprice is a conventional oxidation catalyst vehicle,
while the Pinto is a three-way catalyst vehicle with feedback
carburetor control. Vehicle descriptions and complete refueling
loss control system descriptions are presented in Table A-l and
Figure A-l of the Appendix. The refueling loss canisters in the
Caprice, Pinto and Chevette are described as follows:
RCS
Vehicle Carbon Volume
Caprice
Pinto
Chevette
5.0P_
3.0JP
3. OP
Carbon Mass
1800 g
1100 g
1100 g
Carbon Type*
BLP-F3
BLP-F3
BLP-F3
Location
Underhood
Underhood
Trunk
* Same carbon currently used for controlling evaporative emissions.
The Exxon exhaust and evporative emission test results which
compare baseline and modified versions of the Caprice, Pinto and
Chevette are summarized in Tables 1, 2 and 3. Engine-out data are
summarized in Tables 4 and 5. Refueling loss effectiveness test
results are summarized in Table 6. All Exxon refueling emission"
tests assum&d a no-leak seal at the fillpipe/nozzle interface. In
laboratory practice this was achieved with leak free connections
from the fuel nozzle to the fillpipe.
Benzene emissions from the Caprice and Pinto were measured
during several of the refueling loss SHED tests. These results are
summarized in Table 7. The Exxon data indicate that benzene
control is directly proportional to refueling loss control effect-
iveness, although current benzene levels in the SHED are at the
detectable limit of the instrumentation.
Table 8 presents Exxon's manufacturer cost estimates for
onboard control systems for the 1978 Caprice and .Pinto. These
estimates do not include the costs for fillpipe sealing devices and
pressure reliefs. This hardware represents an additional cost
of approximately $1.50 (manufacturer's cost) per vehicle. Exxon's
cost estimates assume an estimated $.50 credit for downsizing the
ECS canister, which in the two canister system, controls only
carburetor losses. Exxon estimates the incremental cost of two-
canister refueling control systems to range from $8.25 to $10.53.
This estimate includes the above mentioned $.50 credit but does not
include the $1.50 cost for the fillpipe seal and pressure relief.
The corresponding cost range for single canister refueling control
systems is $6.75 to $9.00. For light-duty trucks, Exxon estimates
a cost range of $12 (large single canister) to $20 (two
separate refueling loss canisters or multistage purge systems).
C-48
-------
-7-
a
Ed
*-*
CO
C
3
Q
X
O
z
^
•4
E
so o
v^ 0
cn
3
eg
.C
X
(d U
. ONOO -,-* r*
CM o en o
i— i « • » > • JZ
C < CO C < W O
c
cy
u
u
ceo;
o o CM
•*4 ,^4
iJ U
to eg
)-c u
3 3
eo oo
U-l U-l
e c
o o
o u
• 11 >• i
C < CO C < CO CJ
4^
e
-------
-8-
en
o
o
eo
H
-
XJ
0)
01
O
1
01
e
o
en
'
•o
c -
CO
CO
3
CO
j:
•*• •
H
Cu
fa .
.
•
eg
O
H
60 CO
^ O
CO
O* 4^
CO O
>MM
iA
Id
in
\O r*** eN "™"'
• • • n
«S 0 — 0
vO «l
co en en o
•* C5 O O O
II
e
0N -^ C^ *^
00 — 1 +
-^ o -< .
eN «) en
r-» en vo e>i i
• • • *
en o en o
P** CS \O W\ ^f
cs o e>i o 1
do' do
V
60
. . . . c
en « Q en eu C> co
n > • H > • jz
c < w e < ui u
. c
CU
u
e e n
o o - cu
•»J -«H
^J 1 if
CO co
h U
3 3
BO 60
••* — ^
1M U-4
C C
0 0
C3 CJ
CU tJ
C eu
13 -tl
CO *^
CO O
« s
en
1
Ij
*j
u
co
xj
03
i i
£
en
+
CU
H
fa
*
C-50
-------
-9-
1)
«
•- H
u
Q.
CO
1
CO
o
"
CO
CO
"I
3
o
a
c
60
c
Cd
X
00
"M*
^
CM
p"*
b
o
z
o
o
u
=
0
I
oo
*
1,
o
0
eo r* ff — i
«3- r» O <• CM
• • • • ^
vO O CM CM
CO in ON O> -*
CM O CM O +
— C3 — C3
CU
00
• • * * C
m cu Q m « O co
n > • u > • r;
c < co c < en u
c
cu
u
e c
m \o +
-
CO vO 1^. 0\
CO O r* O CO
- 0 '
.
• • •
sr §1 a. ^r 5 Q.
II < CO II < 03
c , . c «
00
c
10
6
C C j-t
0 0 C
-*4 . -M* (j;
-------
-10-
01
a*
9)
a
a
to
o\
o\
co
CO
o\
\o
0)
CB
H
w
jj
e
• • • I •
— o\ m
-------
-11-
to
H
at
o
• ~4
0}
-------
-12-
Table 8
COST ESTIMATES FOR ONBOARD SYSTEMS
(1)
Charcoal
(2)
(3)
Canister and Valves
(4)
Tank Modifications
Hoses and Tubing
Assembling and Installing
Q $20.00/hr.
(6)
Credit for Downsized
Evaporative Control System
Caprice
$4.96
2.50
0.30
1.57
1.50
$11.03
$0.50
$10.53
Pinto
$3.03
2.00
0.50
1.72
1.50
$8.75
$0.50
$8.25
(1) Estimates are made for manufacturer's large volume production.
(2) 1800 g for the Caprice canister, 1100 g for the Pinto canister
at $1.25/lbm (Calgon BPL-F3 carbon).
(3) Plastic container and valves.
(4) Larger size float/roll-over valve.
(5) 3/4" vapor line from fuel tank to canister, 3/8" .purge line.
EPDM tubing for vacuum control lines.
(6) Additional 4.5 minutes labor at $20/hour.
(7) Reduced size evaporative control canister.
C-54
-------
-13-
Exxon estimates the average cost for onboard control systems
to be $9/vehicle. This is based on the following assumptions:
1) Onboard systems are designed to control refueling emis-
sions 'from light duty vehicles with an average fuel tank size of 17
gallons refueled to 100% capacity from a condition of 10% tank
capacity. The onboard systems are designed to control hydrocarbon
emissions at a level of 6 g/gal.
2) 70% of light-duty vehicles and single tank light-duty
trucks are assumed to use single canister (evap + refueling)
systems.
3) 30% of light-duty vehicles and single tank light-duty
trucks are assumed to use two canister systems.
4) Light duty trucks with dual or large fuel tanks consti-
tute approximately 10% of the light-duty vehicle and light-duty
truck population.
In summary, Exxon finds that onboard refueling 'controls for
light-duty vehicles are a technically feasible, practical, and cost
effective alternative to Stage II vapor recovery. They are of the
opinion that the same may also be said for light-duty trucks.
Mobil
Mobil R&D has modified a 1978 Pontiac Sunbird for control of
refueling losses. This vehicle has a three-way catalyst with a
feedback carburetor control system, and is certified for complaince
with California exhaust and evaporative emission standards.
This modified vehicle uses a single canister which contains 1550
grams of Calgon BLP-F3 carbon. The complete vehicle and refueling
loss control system descriptions are presented in the Appendix.
Table 9 presents comparisons of exhaust and evaporative emissions
from the Sunbird for the baseline and modified configurations; a
summary of the refueling emission data is presented in Table 10.
Similar to Exxon's findings, Mobil states that their test
results have demonstrated that onboard controls are a feasible and
desirable method of controlling refueling losses from light-duty
vehicles and light-duty trucks.
Atlantic Richfield Company
One of the requirements for the operation of an effective
refueling loss control system is a no-leak seal at the fillpipe
nozzle interface. Atlantic Richfield (ARCO) has designed and
tested three types of sealing systems. They included:
1) Modification of the vehicle fillpipe to achieve a seal
when used with conventional lead-free nozzles.
C-55
-------
-14-
,
T3
"Ja
e
a
CO
i
1 i
•
0)
O
CO
•••4
(0
O.
S
6
O
ON I
CO
O W
~« ti
H
0)
U
o
Q.
>
Ed
•O
C
JJ
0]
eo
i
cu
.—4
Q
1 1
0
H
-N .*
00 eo
~> O
en
Q. 4J
3 5
\ """
1—4
eo
h
• ••4
O
*
^
"^
^—^
^
i
"M c
•n
CO
Ed CJ
.
^t
O si- — < vo
O en — i u"> vo
• • • * ^
(NO CS O
eN en i*. r»-
—, _ CN en
-i d -< d
•
r«. tN en
co O r»- CN
do o o
eN
U -*
c n
CO ^^ ^^ ^^
CT\ O 0*> O »^
O O O O
_ — , \r\ . t
».p ^3 *-Q ^3
ON en o en
en o -TO en
• • * • ^*
o o. o o
ON CU O NO CU O CO
U > • H > • -00
. c < co c < co c
- eo
€
.u
c
e c cu
O O CJ
•H -^ 14
u u CD
co eo • O-i
l-i )i
3 3
00 60
(U Uj
C C
O O
CJ CJ
eu -a
c o
• •-4 -*4
~* U-l
CU **^
CO T3
eo O
CU
(•^
jQ
eo
0
>
eo
3
CU
^
eo
CD
^^
3
co
U
•^
CO
O
CO
4J
o -u
J- U
i c
CO 3
CA
f«4
i
3 to
"•0 0 C
— co
— e
eo CU CU
3 "-1 *•"
T3 ja 3
— i co en
> H RJ
! ^
•f-l CO
ca
cu o
U hJ
M
J 60
3 C
•^
QJ --
60 «
eo 3
<
!-> O
u e
i cu
O -H
O 0
U-l
[d
'^
CO
00
en ~»^
C 60
O ^
**
CO CO
CA JJ
-< C
= (U
CU
00 W
C 3
-i CO
i— * CO
CU CU
3 •£
U-l
U Q
OS W
33
CO
X
Ol
±t
cu
J_
tl.
c.
CO
CJ
C
a
4v
u
cu
t-^
c
CJ
C_J
a
6C
v^'
(
ca
C
cu
a
a
C.
i™"*
<
Z
b
P^ ON -T in
ON O> ON ON
CO CN •£!" NO
— i O -N C
u
• ** M
O 'I— I
a. — (
C Bi CO
3
CO •
e ja a;
O *"* ^J
U eo
U-l ON U
•a •— 60
CU ea C
oo c —
U '-4 r->
3 S «J
a. o 3
C U-4
Id CU
CU uj i_i
• • . • co S
\Q \f\ \Q f"N^ «^^ ^^ Q« •
_ — i — , — i c cu oo
sa 3'
CJ to CO
*
*
C-56 .
-------
-15-
2) Modifications to both the fillpipe and lead-free nozzle.
3) Modification of a Stage II vapor recovery nozzle.
A description of each type of seal ,and., a summary of the
durability data collected with each system are presented below:
Fillpipe seals; Two types of fillpipe seals have been ex-
amined. They are a rotary grease seal (similar to grease seals
used on rotating machinery shafts), and a doughnut shaped seal.
The material types for these two seals are a compounded nitrile and
thermosetting urethane, respectively. More complete descriptions
of these seals, including durability data, are found in Figure
A-5 and Tables A-2 and A-3 of the Appendix. Appproximately thirty
days of durability tests with both types of seals have demonstrated
that the rotary seal is more effective, basically due to the
absence of expansion problems when exposed to gasoline liquid and
vapor atmospheres. The seal effectiveness of the prototype fill-
pipe and nozzle hardware are determined by a bench test apparatus
which pressurizes a particular system and measures the resulting
leak ^rates. Seal effectiveness calculations are determined by
dividing the leak rate by a nominal fueling rate (assumed to be 7.5
.galIons/min.). Durability tests conducted with the rotary seal
have demonstrated that the rotary seal is effective after 700-1000
nozzle insertions, which correspond to the number of fuel fills
expected during the life of the vehicle.
Combination fillpipe/nozzle seals; These systems consist of
connecting parts on both the fillpipe and nozzle. Figure A-6 is an
example of a prototype design evaluated by ARCO. Durability test
results with these systems are similar to results obtained with the
rotary seal..
Nozzle Modification; Working prototypes of vapor recovery
nozzles, modified for refueling loss control, have been developed
by OPW and Emco Wheaton and evaluated by ARCO for effectiveness and
durability. These nozzles are designed to seal on standardized
fillpipes. The modified vapor recovery nozzles incorporate a
pressure relief valve, which is located at the vapor return exit or
cast into the nozzle body and designed to open at approximately
14-17 in. water pressure*, thereby permitting the nozzle to refuel
onboard control vehicles and in-use vehicles. Nozzle durability
data are very limited but one nozzle has been inserted and latched
7500 ^times, representatitve of one year of service at a high volume
station, and showed a seal effectiveness of greater than 99%.
ARCO concludes that the preferred seal techniques are either
the fillpipe seal method or the combination fillpipe/nozzle seal.
* Refueling loss control systems designed by Exxon and Mobil are
designed to operate at fill pressures of less than 4 in. of water
pressure.
C-57
-------
-16-
2. Vehicle Manufacturer' Information
General Motors
General Motors has several reservations concerning the appli-
cability of onboard controls, citing such things as: the uncer-
tainty of the effectiveness of f illpipe/nozzle seals, potential
cost increases associated with exhaust emission control systems
which must be designed to control increased CO emissions, negative
fuel penalties which are the result of this increased emission
control, and the long lead time which is required to obtain a
substantial reduction in atmospheric hydrocarbon and benzene
loading. GM has stated that refueling losses can be controlled on
the vehicle (feasibility for trucks has not been demonstrated) ^ or
at the service station. GM's disagreements with controlling
refueling emissions with onboard controls are primarily based on
the issue of cost/effectiveness.
More recently, (Ref. 13 & 14), GM has expressed concern about
possible inadequacies in the design of the API demonstration
vehicles. Specifically, GM states that a large diameter refueling
vent and^ a small restricted vent are necessary for protection from
overfTTTing the fuel tank and preventing excessive hydrocarbon
loading of the storage canister during high temperature vehicle
operation.
GM's March, 1978 submission to EPA presents a summary of their
work on the- control of diurnal evaporative emissions and refueling
losses using fuel tank bladders.
It is EPA's opinion that the theoretical control effectiveness
of evaporative and refueling loss emissions using bladder tank
technology is high and that problems related to vapor storage,
pressure relief valves, and bladder materials can be solved. It
is recommended that bladder tank feasibility be researched by
funding a. bladder tank hardware demonstration contract.
The March, 1978 submission presents calculations showing that
the additional weight of the components of an onboard control
system will cancel out any potential energy saving which results
from the combustion of the refueling vapors. ECTD agrees with
this analysis.
The June, 1978 submission is basically a cost effectiveness
analysis comparing onboard controls with Stage II controls (balance
displacement and vacuum assist systems). GM estimates that onboard
control systems, effective with the 1982 model year, will range
from §16 to $24. These figures are about $5 to $9 higher than the
March, 1978 estimates due to higher estimates for larger canisters
and a new vapor/ liquid separator. GM assumes, that the seal at the
fillpipe/nozzle interface will be obtained using modified vapor
C-58
-------
-17-
covery. Rather, GM emphasizes certain technical concerns which
they say are not fully addressed by the API study. According to
GM, these include API's unsubstantiated support for the onboard
fillpipe seal and pressure relief (lack of adequate durability
results), an unknown CO penalty for light-duty vehicles (no sensi-
tivity data relating CO to test procedure differences), and un-
proven feasibility for trucks.
GM is of the opinion that accelerated laboratory durability
tests are not sufficient to prove that proposed elastomer type
seals will be effective in the extreme usage and environmental
conditions of the real world, particularly when considering a ten
year average lifetime for a light-duty vehicle.
Ford
_ Ford has submitted test results from four 1978 model year
vehicles (three non-feedback systems and one feedback control
system) modified for refueling loss control. These vehicles are
described in detail in Table A-4 in the Appendix and in their
submission to EPA, Ref. 7. The purge control systems for these
vehicles are shown in Figures A-7 and A-8 in the Appendix.
Ford estimates the cost to the consumer of onboard controls to
range from $15-$20. They note that the $15-$20 estimate does not
include additional expense for such items as: packaging costs,
incremental labor costs, or the costs for additional exhaust
emission control, such as feedback control over a wider air/fuel
ratio range.
Recent Ford material, Ref. 12, suggest that -the cost of
onboard syst.ems may range from $30 to $253. The $30 estimate
includes costs over the original $15-20 estimate, including costs
for such items as vehicle modifications to package onboard systems,
incremental assembly, and material substitution. The $253 estimate
includes the cost for a feedback fuel system and electronic con-
trols for vehicles which are not planned to be equipped with these
control devices.
On the basis of their in-house test results, Ford has conclu-
ded that onboard controls are not technically feasible for light-
duty vehicles.
American Motors
AMC has submitted a letter to EPA, Ref. 9, which states their
concerns with the possible use of onboard controls. They state
that packaging concerns, reduced quantities of purge air from
downsized engines, and compliance with stringent evaporative
emission standards are unresolved technical issues which have not
been addressed by the API work to date.
C-59
-------
-18-
AMC does not find that API has demonstrated light-duty vehicle
technical feasibility.
IV. Analysis of Available Information
1. API Work .
Exxon
Exxon R&E appears to have done a credible job in character-
ising the components of a hydrocarbon adsorption system. An
examination of the results from baseline tests and tests with the
modified Pinto (3-way + feedback carburetor system) show small but
finite increases in engine-out (14%) and tailpipe (8%) CO emis-
sions. HC, CO, and NOx emissions are still well below statuatory
emission levels for low mileage vehicles. Engine-out CO emissions
from the Caprice are approximately 20% higher than baseline test
results; tailpipe CO emissions are approximately 10% higher than
the baseline results. No increase in tailpipe CO was observed
during tests with the Chevette. Exxon suggests that differences in
CO emissions for the Caprice and the Pinto can be further reduced
by minor modifications to the refueling loss control system or the
exhaust emission control system, although this has not been demon-
strated.
*
Figure A-2 shows canister purge as a function of time.
Although the data are bench test results, the results are also
representative of actual control system purge data. It is signifi-
cant to note that the refueling loss canister is essentially purged
to its working capacity after three LA-4 driving days. _ This
implies that the refueling control/exhaust emission interaction is
likely to be less in a typical driving day than Exxon has measured
using conservative test methods, which required running a cold
start FTP immediately after: a 90% refueling.
| j
ECTD expects that refueling loss control systems will result
in slightly higher CO feedgas levels. Exxon estimates that the
average increases in CO feedgas between refuelings will be approx-
imately 5% for non-feedback control systems and less than 3% for
feedback control systems. ECTD has no other data concerning either
the magnitude of the average CO feedgas penalty or the resulting
effect on catalyst durability. It is ECTD's opinion that the Exxon
estimates are reasonable and that these -additional CO penalties
will make it more difficult for vehicle manufacturer's to certify
some engine/families to the 3.4 g/mi CO standard. The higher CO
levels somewhat reduce the margin available to allow for exhaust
system deterioration over 50,000 miles.
ECTD finds that light-duty vehicles equipped with onboard
systems are capable of meeting a 2 gram evaporative emission
standard.
C-60
-------
-19-
An analysis of the control effectiveness"of benzene emissions
during refueling, Table 6, indicates that icharcoal. canisters can
control in excess of 99% of the uncontrolled benzene emissions.
Exxon conducted additional tests with the Caprice and Pinto using
indolene test fuel with a high benzene content <4.2%X. The results
from these five tests suggest that benzene emissions are controlled
in excess of 992 during refueling.
Packaging refueling loss control systems, is :a difficult
problem, but definitely not an insurmontable one. The refueling
loss canister is located behind the rear seat and above the rear
axle in the Caprice, and in the engine compartment of .the Pinto.
It is Exxon's opinion, and ECTD agrees,- ;that it is. possible for
manufacturers to locate a refuelling loss canister on downsized
vehicles without major engine compartment tor, sheetmetal modifica-
tions . .'.-,- -. , ..;;„ i":;, ..'•-,:'• •:.
The feasibility of refueling lossr controls for light-duty
trucks has not been evaluated by Exxon, but they are of the opinion
that refueling loss control is f easib'le* ;f or, light-duty trucks by
using larger control systems and more sophisticated -purging con-
trols (refueling loss control canisters for each tank and/or two
stage purging systems). It-. is ECTD's opinion that the control of
refueling losses from light-duty trucks needs to be demonstrated,
especially the ability to comply with a 2 g evap standard, before
onboard controls are judged to be effective for these vehicles at
the costs Exxon has estimated.
Table 8 shows Exxon's detailed manufacturer's cost estimates
for refueling control systems which have two canisters. ECTD finds
these cost estimates to be reasonable for onboard systems designed
to control. 100% of refueling emissions from 90% fill conditions.
Exxon estimates the average manufacturer's cost for the light-duty
truck and light-duty vehicle population to be about $9. That
number is derived as follows:
One-cansiter vehicles*
Two-canister vehicles*
6,000 to 8,500 Ibs. trucks**
Weighted average
Assumed
Average Cost
$7.88
$9.38
$16.00
$9.00
% of
Population
70
20
10
* Includes light-duty vehicles and light-duty trucks under 6000
GVW - average fuel tank size = 17 gal.
** Average fuel tank size = 35 gal.
The charcoal cost per gallon of tank volume is assumed to be about
$0.20.
C-61
-------
-20-
The $9 incremental manufacturer cost may be translated to a
consumer cost estimate of $16.20 by multiplying the manufacturer's
cost estimate by a factor of 1.8 (Ref. 10, EPA Report "Cost Esti-
mations for Emission Control Related Components/Systems and Cost
Methodology Description" by Rath and Strong-, March 1978). The 1.8
factor is in general agreement with previous EPA studies, such as
the EPA Report, Ref. 11, "Investigation and Assessment of Light-
Duty Vehicle Evaporative Emission Sources and Control," June 1976,
which used a manufacturer to consumer cost factor of 2.0. The
$16.20 estimate is in good agreement with consumer cost estimates
submitted by GM ($16-$24) and Ford ($15-$20). It is possible to
further reduce the cost of an onboard system by trading off some
degree of refueling loss control effectiveness.
Exxon has designed refueling loss control systems based on
conservative criteria, and thus a different set of design criteria
will afford reductions in the cost of onboard control systems.
Texaco has submitted data (Figure A-ll) Ref. 12, which relates the
number of light-duty vehicle refuelings and the percent of tank
fill. A reasonable design criterion is to size the refueling
canisters to control 90% of nationwide refueling emissions.
Calculations (Figure A-12) show that 90% control can be achieved by
designing systems to control 100% of "refueling emissions from fills
to 63% of fuel tank volume. If onboard control systems are de-
signed to control emissions from refueling to 63% of tank capacity
rather than 90% of tank capacity, the Exxon estimate of $9 per
vehicle can be reduced by $1.60 as the result of reduced charcoal
quantity. This cost reduction is proportional to the reduction in
carbon bed volume. The net effect of this design change is a cost
reduction to the consumer of approximately $2.88. Changes in
design specifications such as the 90% fill requirement may afford
additional'cost reductions for other control system components as
well as a general reduction in the problem of packaging onboard
control systems.
*
ECTD estimates the consumer cost of light-duty vehicle onboard
control systems designed for maximum control effectiveness to be
about $17. This estimate does not include an estimate for the cost
of the fillpipe seal or pressure relief valve. The $17 estimate is
based on Exxon estimates, which when translated to consumer costs,
are in agreement with consumer cost estimates provided by GM and
Ford.
Exxon estimates the manufacturer's cost for a fillpipe seal
and onboard pressure relief valve to be approximately $1.50. ECTD
estimates the consumer cost of an onboard fillpipe seal and pres-
sure relief to be $2.70.
Mobil
Comparisons of baseline and modified vehicle test results
indicate that Mobil R&D is able to add refueling controls to the
C-62
-------
-21-
1978 Pontiac Sunbird (3-way + feedback carburetor system) without
adversely affecting exhaust or evaporative emissions. No changes
in engine-out or tailpipe CO emissions are observed. Evaporative
emissions are also unchanged, with both baseline and modified
vehicle test results near the- 2 g evaportive emission level.
It must be emphasized, however, that Mobil and Exxon use
different test procedures for measuring the refueling control/
exhaust emission interaction. Mobil's test procedure consists of
the following sequence of events:
1) Load canister to approximately one-half of working
capacity.
2) Condition vehicle by driving two simulated city driving
days (4.7 LA-4's with one hour hot soaks in between and a diurnal
at the end of the day).
3)-
Drain fuel tank to 10% of volume.
4) Refuel to 90% of volume in SHED. •
5) Conduct hot start emission test.
6) So'ak vehicle for 11 hrs.
7) Conduct diurnal evaporative test in SHED.
8) FTP '
9) Hot soak evaporative test in SHED.
Steps 1, 2, and 5 are the important differences between the
test procedures used by Exxon and Mobil. Mobil starts their test
sequence with a canister loaded to one-half of working capacity,
versus a saturated condition for the Exxon procedure. Mobil purges
the refueling loss canister with two LA-4 driving days, versus the
Exxon method of purging by running a series of LA-4 driving days
until the fuel tank reaches 10% of capacity. Mobil runs a hot
start emission test prior to the FTP; no such additional condition-
ing is used in.the Exxon test sequence. It is ECTD's opinion that
the the Mobil test sequence, particularly the addition of a hot
start exhaust emission test, will result in a less severe refueling
control/exhaust emission interaction. This is due to the smaller
quantity of hydrocarbon which is purged during the cold start FTP
when using the Mobil test sequence. The actual emission sensi-
tivity to various test procedure arrangements has not yet been
determined.
Atlantic Richfield Company
ARCO states that the fillpipe modification approach and. the
C-63
-------
-22-
combination fillpipe/nozzle seal concept are the preferred ^ tech-
niques for achieving a no-leak seal. This recommendation is not
supported from an analysis of leak rate and durability data because
the test results show that seal effectiveness among all three
concepts are equal. Cost estimates for the three designs have not
been submitted. ARCO is continuing to collect field durability
data on their prototypes, but the lack of a more extensive durabil-
ity demonstration under simulated conditions of real world usage
makes it questionable to assume that their seals will function as
well in the field as they have in the laboratory.
In particular, ARCO has not adequately addressed the issue of
onboard pressure relief valves versus liquid pressure relief valves
located on the fill .nozzle. Pressure relief valves are necessary
to prevent over-pressurization of the fuel tank in the event of a
failure of the automatic shutoff on the fill nozzle. For the
purpose of fuel tank integrity in the event of a vehicle crash,
NHTSA recommends that the pressure relief not be located on the
fuel tank. However, a relief valve might be incorporated safely
with a fillpipe seal mechanism, which would be sealed-off by the
fuel cap during vehicle operation.
The achievement of a safe and durable seal at the nozzle
fillpipe interface is critical to the performance of an onboard"
refueling loss control system. ARCO has demonstrated that the
effectiveness of fillpipe seals, combination seals and nozzle
seals are equal; but the design, location, and durability of the
pressure relief valve have not been adequately addressed.
Conceptually, a pressure relief may be designed to function
properly when located on the vehicle or on the nozzle. However, if
refueling losses are controlled on the vehicle, it is recommended
that the fillpipe/nozzle seal and pressure relief valve also be
located on the vehicle.' Locating all parts of an onboard system on
the vehicle will prevent the potentially serious problem of refuel-
ing a controlled vehicle without protection from overpressurization
(no relief valve). Administrative and certification concerns also
suggest that onboard controls are practical only if the seal and
pressure relief are located onboard.
An alternative technique of achieving a seal at the fillpipe/
nozzle interface is the liquid trap or submerged fill. This seal
concept has not been" adequately investigated. Submerged fill
offers the potential for significant advantages in terms of simpli-
city of operation and durability (mechanical, magnetic, or elas-
tomer type seals are avoided). It is ECTD's opinion that the
submerged fill concept should be investigated. • Submerged fill (and
seal techniques investigated by ARCO) must be evaluated in the
context of a complete refueling and evaporative emission control
system. This includes incorporating features to provide^ adequate
thermal expansion capability and rollover protection while still
permitting normal safe refueling.
C-64
-------
-23-
2. Vehicle Manufacturer Information
General Motors
General Motors has several reservations concerning the appli-
cability of onboard controls, citing such things as: the uncer-
tainty of the effectiveness of fillpipe/nozzle seals, potential
cost increases associated with exhaust emission control systems
which must be designed to control increased CO emissions, negative
fuel penalties which are the result of this increased emission
control, and the long lead time which is required to obtain a
substantial reduction in atmospheric hydrocarbon and benzene
loading. GM has stated that refueling losses can be controlled on
the vehicle (feasibility for trucks has not been demonstrated) or
at the service station. GM's disagreements with controlling
refueling emissions with onboard controls are primarily based on
the issue of cost/effectiveness.
More recently, (Ref. 13 & 14), GM has expressed concern about
possible inadequacies in the design of the API demonstration
vehicles. Specifically, GM states that a large diameter refueling
vent and a small restricted vent are'necessary for protection from
overfilling the fuel tank and preventing excessive hydrocarbon
loading of 'the storage canister during high temperature vehicle
operation.
GM's March, 1978 submission to EPA presents a summary of their
work on the control of diurnal evaporative emissions and refueling
losses using fuel tank bladders.
It is EPA's opinion that the theoretical control effectiveness
of evaporative and refueling loss emissions using bladder tank
technology is high and that problems related to vapor storage,
pressure relief valves, and bladder materials can be solved. It
is recommended that bladder tank feasibility be researched by
funding a bladder tank hardware demonstration contract.
The March, 1978 submission presents calculations showing that
the additional weight of the components of an onboard control
system will cancel out any potential energy saving which results
from the combustion of the refueling vapors. ECTD agrees with
this analysis.
The June, 1978 submission is basically a cost effectiveness
analysis comparing onboard controls with Stage II controls (balance
displacement and vacuum assist systems). GM estimates that onboard
control systems, effective with the 1982 model year, will range
from $16 to $24. These figures are about $5 to $9 higher than the.
March, 1978 estimates due to higher estimates for larger canisters
and a new vapor/liquid separator. GM assumes that the seal at the
fillpipe/nozzle interface will be obtained using modified vapor
C-65
-------
-24-
recovery nozzles. GM does not include seal costs in its estimate.
They assume these costs will be the same for either Stage II or
onboard controls and hence, leave these costs out of their analy-
sis of both options. General Motor's onboard cost estimates are
costs to the consumer. These estimates kre based on costs for
hydrocarbon adsorption systems which control evaporative and
refueling emissions with one canister and systems which use two
separate canisters for containing evaporative (diurnal and hot
soak) and refueling emissions. The GM cost estimates are con-
sistent with Exxon's manufacturers cost estimates for onboard
controls. As discussed earlier, it is possible to design cheaper
refueling loss control systems by not providing 100% control _of
refueling emissions under worst case conditions. If the design
criterion of 100% control for a 90% refueling is changed to
100% control for a 63% refueling, it is possible to reduce the
required working capacity of the charcoal canister, thus reducing
the average system cost to the consumer by about $3.00.
GM did not comment on the feasiblity of refueling loss con-
trols for light-duty trucks and heavy-duty gasoline powered ve-
hicles.
Ford
Ford emphasizes that the refueling loss/exhaust emissions
interaction is a function of the test procedure and that the
differences between emissions interactions measured by Exxon and
Mobil are due to test procedure differences. This statement is
correct, although the actual emission sensitivity to the test
procedure is unknown.
Ford attributes the high CO effects which they have observed
with both conventional oxidation catalyst systems and three-way
plus feedback carburetor systems to the presence of refueling loss
controls. However, high CO emissions are likely the result of
using a manifold vacuum purge control system. This system results
in cold-start hydrocarbon loadings that are two to three times
higher than results obtained with venturi vacuum control systems
(Exxon system). This is the reason the Ford results are so high,
particularly engine-out CO emissions. Ford maintains that refuel-
ing loss control systems produce peak enrichment effects equal to
two air/fuel ratios, which is beyond the capability of their
current feedback carburetor control system. Exxon has demon-
strated, however, that venturi vacuum maintains the air/fuel ratio
within the control limits of the feedback control system. Problems
with the existing Ford feedback control system are likely to be the
result of response time problems, not control range problems.
Some of Ford's concerns with onboard refueling control sys-
tems, such as packaging, weight of onboard systems, and the design
of vapor/liquid separators have been examined during the API study
C-66
-------
-25-
and shown not to be significant problem areas. Other concerns with
onboard controls, including system durability, onboard feasiblity
for light and heavy duty trucks, and high altitude feasibility,
have not been adequately addressed in any of the information
submitted to ECTD. It remains ECTD's judgment that these issues
need further examination, particularly before onboard controls are
determined to be feasible for light and heavy-duty trucks. Al-
though onboard durability data are not available, ECTD finds that
onboard control systems should be as durable as current evaporative
emissions control systems, which last for the lifetime of the
vehicle.
Ford estimates the consumer cost of onboard controls for
light-duty vehicles to range from $30 to $253. EPA estimates that
the consumer cost of onboard control systems will be about $20
(includes $2.70 for the cost of an onboard seal and pressure
relief).
American Motors
AMC's concerns with the use of onboard controls are addressed
to the issues of exhaust and evaporative emissions interactions,
feasiblity of vehicles using small engines, costs, and light-duty
truck feasibility. With the exception of feasibility for light-
duty trucks', AMC's concerns have been examined in detail by the API
study. EPA's analysis of that data is that refueling loss controls
are feasible for light-duty vehicles at a consumer cost of approxi-
mately $17.
V.
Conclusions
Feasibility
An Analysis of the available information has shown that
onboard refueling loss controls are feasible for light-duty
vehicles designed to meet low exhaust and evaporative emission
standards (0.41 EC, 3.4 CO, 1.0 NOx and 2.0 Evap.). However, the
feasibility for light-duty trucks, particularly the assurance that
onboard control systems are compatible with a 2 gram evaporative
emission standard, has not been established. Feasibility for
heavy-duty gasoline vehicles has not been established.
An analysis of information and test data presented to EPA
regarding the control of light-duty vehicle refueling emissions
offers the following conclusions:
1. Onboard control systems in laboratory use situations can
control in excess of 97% of the uncontrolled hydrocarbon refueling
losses.
C-67
-------
-26-
2. The same systems in laboratory use situations can control
in excess of 972 of the uncontrolled benzene refueling losses.
3. Test results from two light-duty.- vehicles equipped with
three-way catalysts, feedback carburetors, and prototype refueling
loss systems show that tailpipe CO emissions range from a 0 to 82
increase.
Test results from the same vehicles show that engine-out CO
emissions range from a 0 to 14% increase.
4. Emission data from two conventional oxidation catalyst
equipped light-duty vehicles show that tailpipe CO emissions
range from a 0 to 10% increase.
Data from one of the conventional oxidation catalyst vehicles
show that engine-out CO emissions increase by 10 to 20%.
5. The addition of refueling loss controls to light-duty
vehicles does not significantly affect evaporative emission losses.
6. Minor increases in CO exhaust emissions seen for some
vehicles can probably be controlled by minor change to either the-
refueling loss control system or to the exhaust emission control
system. However, the addition of refueling loss controls will
likely make it more difficult to certify some vehicles to the 3.4
g/mi standard at 50,000 miles.
7. Onboard controls do not affect vehicle fuel economy.
8. Onboard controls do not affect vehicle driveability.
9. Refueling loss control systems for light-duty vehicles
are estimated to add $17 to the vehicle sticker price. ^The^$17
estimate does not include the costs associated with the fillpipe/
nozzle seal or pressure relief valve. The consumer cost of a seal
and pressure relief in the fillpipe is estimated to be about $2.70.
The cost of a seal on the nozzle should be roughly the same as the
cost for a Stage II nozzle. However, ' it is recommended that all
components of an onboard control system be located on the vehicle.
Lead time
Onboard refueling loss control can be implemented for 1984
model year light-duty vehicles, provided that potential problem
C-68
-------
-27-
areas such as the design and development of effective fillpipe/
nozzle seals and pressure relief valves do not require additional
hardware demonstration programs. It is anticipated that the
fillpipe/nozzle seal and the control feasibility for light and
heavy-duty trucks are issues which can be resolved during the NPRM
process.
ECTD estimates that a minimum of two years leadtime will be
required by manufacturers for development (purge system optimiza-
tion, design and verification of fillpipe seal mechanisms) and
production tooling changes (tooling associated with fabrication and
relocation of new evaporative control components). These estimates
are based in part upon data provided by manufacturers relating to
carburetor tooling changes, and in part upon data supplied by GM
relating tOf retooling changes for body panel modifications.
Additional time will be required for EPA to develop a certification
type tes't procedure and issue regulations; however, the certifica-
tion procedure development can overlap the production tooling lead
tLme,'QOnTheref°re' the Pr°Jection ^ that an NPRM can be published
in 1980 with final rules promulgated by 1981 with the earliest
possible implementation date being 1984. (See lead time chart
Figure 1).
Compliance Costs
ECTD.estimates that certifying light-duty vehicles for compli-
ance with a refueling loss standard will require an additonal
nfe7nn ^nerSTT.ar at the EPA-MVEL' T^3 « based on an estimate
of 100-150 refueling loss tests per year. Facility modifications/
equipment procurements will cost from $30K to $80K.
A potentially significant impact on refueling loss compliance
costs is Inspection/Maintenance testing of light-duty vehicles.
EPA has not developed, and is not aware of, a valid I/M test for
determining the performance of evaporative emission control sys-
tems. Monitoring the performance of in-use refueling loss control
systems will be difficult and cumbersome. At this time, it may be
assumed that the onboard compliance costs associated with an I/M
test will be equal to the cost of Stage II enforcement.
VI. Recommendations for Future Work
1. ECTD recommends that additional hardware testing be
conducted to determine the optimal fillpipe-nozzle'seal. Addition-
,?'*,? °Peratlon and durability of a fillpipe or nozzle pressure
relief (including overfill protection for liquid expansion) must be
demonstrated. The use of an onboard liquid trap seal (submerged
till; as an alternative to elastomer type seals should be invest--
igated.
C-69
-------
-28-
Figure 1
Lead Time
Quarter:
Develop Certification
Test Procedure
Continued Study of
Fillpipe/Nozzle Seal
Concepts
Decision on Seal
Concepts
EIS, EIA, NPRM
Preparation
Publish NPRM
Final Rule
Manufacturers
Lead Time
U,
1980
234
Calendar Year
1981 1982 1983 1984
234I1234I1234I1234I
-(Decision to publish service station nozzle
requirements or put seal on vehicle)
"1984 MY
C-70
-------
-29-
2. ECTD recommends that additional hardware testing be
conducted to assess the feasibility of controlling refueling
losses on light-duty trucks and heavy-duty gasoline powered
trucks.
3. ECTD recommends that the need for controlling refueling
losses from diesel powered vehicles be investigated since these
vehicles are predicted to represent a substantial fraction of the
entire motor vehicle population in the 1980's.
4. ECTD recommends that the bladder fuel tank be investi-
gated as an alternative to carbon adsorption technology. It is
ECTD's opinion that the theoretical control of ecvaporative and
refueling loss emissions with bladder tanks is high and that
technical problems can be solved. It is recommended that bladder
tanks feasibility be researched by funding a hardware demonstration
contract.
5. Finally, ECTD recommends that methods of reducing the
cost_of onboard refueling control systems be examined. Such
studies should be directed toward tradeoffs between level of
control effectiveness and cost. It may be possible to sacrifice
control capacity that is required under only infrequent conditions
to achieve a proportionally more significant cost savings. '
C-71
-------
-30-
References
1. "Control of Refueling Emissions," Statement by General Motors
Corporation, June 11, 1973.
2 "Control of Refueling Emissions with an Activated Carbon
Canister on the Vehicle - Performance and Cost Effectiveness
Analysis," Interim Report Project EF-14, prepared for the
American Petroleum Institute, Washington, D.C., October 1973.
3 "On-Board Control of Vehicle Refueling Emissions - Demonstra-
tion of Feasibility," API Publication No. 4306, October 1978.
4. "Summary and Analysis of Data from Gasoline Temperature Survey
Conducted at Service Stations," Radian Corporation, Austin,
Texas. Prepared for the American Petroleum Institute, Wash-
ington, D.C., November 1976.
5. "General Motors Commentary to the Environmental Protection
Agency Relative to On-Board Control of Vehicle Refueling
Emissions," March 1978.
6. "Suppplement to General Motors Commentary to the Environmental
Protection Agency Relative to" On-Board Control of Vehicle
Refueling Emissions," June 1978.
7. "Ford Motor Company Response to EPA Concerning Feasibility and
Desirability of a Vehicle On-Board Gasoline Vapor Recovery
10,
and
System."
8. "Ford Motor Company Position Concerning Feasibility
Desirability of Vehicle On-board Refueling Vapor Control
Systems," November 6, 1978.
9. AMC letter to Paul Stolpman, August 3, 1978.
"Cost Estimations for Emission Control Related Components/Sys-
cems and Cost Methodology Descriptions," Rath and Strong,
Inc., Lexington, Massachusetts. Prepared for the Environ-
mental Protection Agency, Ann Arbor, Michigan, March 1978.
11. "Investigation and Assessment
tive Emission Sources and
of Light-Duty Vehicle Evapora-
Control," Exxon Research and
12.
Engineering Company, Linden, New Jersey. Prepared for the
Environmental Protection Agency, June 1976.
Texaco statement submitted to Paul Stolpman, July 18, 1978.
C-72
-------
APPENDIX
^ Appendix contains detailed descriptions and data from the
test vehicles and fillpipe/nozzle seals which were used in the most
recent testing and evaluation of refueling loss control systems.
Exxon
Table A-l presents a description of all Exxon test vehicles.
Figure A-l is a schematic of the basic control system designed for
the Chevrolet Caprice and the Ford Pinto. The refueling emissions
(RCS) canister controls both refueling emissions and diurnal
evaporative emissions; the evaporative emissions (ECS) canister
controls carburetor hot soak losses. Exxon investigated several
different purge mechanisms, including combinations of manifold
vacuum and venturi vacuum, and two stage purge control valves
controlled by ^ fuel volume, but venturi vacuum, which is propor-
tional to engine air flow, is the most effective purging method.
Exxon's control system is designed to maintain the total purge air
volume (RSC + ECS) equal to the purge air volume of. the unmodi-
fied vehicle's evaporative control system.
The air bleed control valve, shown in Figure A-l, is necessary
because the'RCS canitser is purged more efficiently (higher hydro-
carbon purge per unit volume of air) than the unmodified ESC
system-, thereby resulting in richer A/F mixtures. This air bleed
may not be necessary for other vehicles wich feedback carburetor
controls.
figure A-2 is a plot of the RCS canister purging as a function
of time. These data are based on consecutive LA-4 driving days.
As noted, the RCS system is purged at a rate of about 4 litres/
min., which corresponds to a total canister purge volume of about
40 litres during an LA-4 driving cycle.
Mobil
Specifications for the vehicle Mobil has modified for refuel-
ing loss control are summarized as follows:
Vehicle: 1978 California Pontiac Sunfaird
Engine Size: 151 cu. in. L-4
Interia Weight: 3000 Ibs.
Emission Control System:
Exhaust: 3-way catalyst with feedback carburetor, •
EGR
Evaporative: Carbon canister
C-73
-------
A-2
Fuel Tank Capacity: 18.5 gallons
The production vehicle is modified for controlling refueling
emissions by enlarging the existing carbon canister, (one canister
controls refueling, diurnal, and hot soak loss),- enlarging the
vapor line between fuel tank and canister, redesigning the vapor/
liquid separator, and installing a purge control orifice between
the canister and intake manifold. A schematic of the Sunbird's
control system is shown in Figure A-3. Various flow control
orifices were inserted in the canister purge line but best results
are obtained with an orifice of 0.100 in. diameter. Mobil uses
1550 grams of Calgon BPL-F3 carbon for their control system, which
assumes a 20% safety factor. This quantity ofl carbon is based on a
90Z fill of the 18.5 gallon tank, and assumes ia hydrocarbon loading
of six grams per gallon of dispensed fuel. The working capacity of
the canister is approximately 240 grams. The basic components of
the canister control system are shown in Figure A-4. The ported
vacuum purge control valve is from a 1978 Chevrolet Impala evapora-
tive canister, while the two fuel tank vapor valves (two are
used to reduce the pressure drop during the refueling operation)
are carburetor bowl valves from a 1978 Impala. Using two fuel tank
vapor valves results in fillpipe pressures as low as two inches of
water pressure during refueling. The fuel tank vapor valves are
also controlled by manifold vacuum such that the vapor valves
are closed when manifold vacuum is present at the control port.
Atlantic Richfield Company
Figure A-5- shows the fillpipe seal which ARCO has developed
and tested for durability. Tables A-2 and A-3 are typical of the
durability results obtained with this seal. Figure A-6 is an
example of-a prototype combination fillpipe/nozzle seal which has
been developed and evaluated by ARCO.
•i'
Ford
The vehicles which Ford has used for refueling loss testing
are shown in Table A-4. A single 4.35 I canister is used in the
Mustang, while a duel, canister system, 829 ml and 3.4 1, are used
for controlling carburetor vapors and diurnal/refueling losses,
respectively, in the Pinto. The purge systems for the Mustang and
' the Pinto are shown in figures A-7 and A-8.
Figures A-9 and A-10 are plots of canister loading versus test
procedure sequence. These plots indicate that Ford's refueling
loss control system is quite sensitive to the particular test
procedure which is used to quantify the refueling control/exhaust
emission interaction.
C-74
-------
A-3
Table A-l
Vehicle Descriptions
Make
Engine Displacement/
Model Configuration
Control Systems
Fuel Tank
Capacity
(gallons)
Chevrolet Caprice 5.0 litre (305 CID)/V-8
Ford Pinto 2.3 litre (140 CID)/L-4
Plymouth Volare 3.7 litre (225 CID)/L-6
Chevrolet Chevette 1.6 litre (98 CID)/L-4
Ox. Cat., AIR, EGR
3-Way, Ox. Cat.,
AIR, EGR
Ox. Cat., AIR, EGR
Ox. Cat., AIR, EGR
21.0
13.0
18.0
12.5
C-75
-------
Table A-2
FILLPIPE MODIFICATION
ROTARY SEAL-CR 7538
LEAK RATE AS AFFECTED BY
FILLNECK PRESSURE AND WEAR
NO. OF SPOUT
INSERTIONS
0
100
100
100
100
100
100
100
100
100
100
TYPE
SPOUT
Smooth
Rough
Smooth
Rough
Smooth
Rough
Smooth
Rough
Smooth
Rough
CUMULATIVE
INSERTIONS
0
100
200
300
400
500
600
700**
800
900
1000
FT3/MIN
@ 5" W.C.
0
0
0
0
0
0
0
0
0
0
0
LEAK *
(§'15" W.C.
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001 '
0.002
0.00.2
0.002
* Leak rate average of six nozzle insertions.
** Expected number of insertions during vehicle life.
RGJ:ip
7/13/78
C-76
-------
HOURS OF
LIQUID
SOAK
0
16
35
Table A-3
FILLPIPE MODIFICATION
ROTARY SEAL-CR. 7538
EFFECT OF LIQUID
SOAK ON SEAL ID-
TOTAL WEEKS
OF VAPOR
SOAK
0
0
0
2
3
4
5
6
7
8
AND VAPOR
AND' LEAK
SEAL
ID, IN.
.712
.712
.711
.705
.699
.701
.703
.698
.693
.691
GASOLINE
RATE*
FT3/MIN LEAK**
w J yv*~ (9 J.3 we
o o
o o
0 o
0 o
0 .001
0 .001
•001 ..001
0 o
o .001'
•001 .002
* Vapor and liquid soak at 72°F.
** Leak rate average of nine nozzle insertions.
RGJrip
7/13/78
C-77
-------
A-3
Table A-l
Vehicle Descriptions
Make
Model
Engine Displacement/
Configuration
Control Systems
Fuel 'Tank
Capacity
(gallons)
Chevrolet Caprice 5.0 litre (305 CID)/V-8
Ford Pinto 2.3 litre (140 CID)/L-4
Plymouth Volare 3.7 litre (225 CID)/L-6
Chevrolet Chevette 1.6 litre (9.8^ CID)/L-4
Ox. Cat., AIR, EGR
3-Way, Ox. Cat.,
AIR, EGR
Ox. Cat., AIR, EGR
Ox. Cat., AIR, EGR
21.0
13.0
18.0
12.5
C-78
-------
I
CO
s
UJ
03
03
J
O
cc
2
o
o
g
O3
03
2
LL!
o
3=
LU
LL
LU
CC
Q
LU
<
CC
o
CL
<
LU
C-79
-------
*>
a
O
s
-------
CO
O
to
UJ
O
•2.
__i
UJ
<
u
3
00
o
o:
O
O
C/l
O
CO
O.
-------
3» M » *• •
Q- t/N ro
_l
S .2>E
(O
oo *"
"
I ™
00
o>
-------
FILL PIPE MODIFICATIONS
ROTARY SEAL , ..
ROTARY SEAL
TRAP DOOR
SPOUT
LEAD RESTRICTOR
FILL PIPE MODIFICATIONS
ROTARY SEAL
ROTAPY SEAL
TRAP DOOR
SPOUT
LEAD RESTRICTOR
C-83
-------
Figure A-6
NOZZLE / F1LLPIPE MODIFICATION
CONE SEAL
SPOUT
LEAD RESTRICTOR
TRAP DOOR
•LATCH COLLAR
CONICAL SEAL
NOZZLE / F1LLP1PE MODIFICATION
CONE SEAL
LEAD RESTRICTOR
TRAP DOOR
SPOUT
LATCH COLLAR
CONICAL SEAL
C-84
-------
s
C=3
H
19
CO
CO
CO
a
I
O
CO
c co
ffl
^j
en
3
S
Figure A-7
a &a
EH O
H M
63 05 3
> O W
i-3 CU
< 3 O
>Ms
63 O O
O CO =3
tf rH Q
5 • •*!
CM O >
63
Ce5
So
5^
S a
=3 >
^5=
> >
to cu
> cH
> 2
EH S
a o
3 O
r-l EH
i
CO =>
3 CO Crf
I-H CO
<*<*;&
EH CU ^
2H ( ^
8X EH
C3
CS
a • o
> 3 S*
i-J 1— 1
«a! 63
> ^° «
-S- 1-1 CO
CO O Sai &3 J
~ * z)
S < 3
CO 3
° s
-j pH
P3
^
0
s:
o
i— <
EH
O
a
3
3
O
O
EH
*£t
&a
>
i
.
C! >
C-85
-------
p.
S5S
J3 ^ _
"""s.£ 5
a
S12
z;
o
Figure A-8
C-86
-------
Procecurs £2 Set 2
• Mustang 5.CL (8218) 197$ 1*9 States
Date 7-17-78 Test 29
- 1*350 nl Canister, Tank & Garb. Eowl w/.lcG" Purge Orifice
|- I GrarLS of Vapor Purged (-)j Absorbed (+)
Figure A.-9
+55 I -98 I +27 1-50 I +35 I -38 I +39 I +8 I +95 ' -7^ 1 +18 !
si
2-4
a
•p
03
•H
c
o
Gas.
"Gss/ifi. CO
ON BOARD SIS. Bag 1,22:3 «J
Bag 1 CO C-=3.
Bag i CO- Gcs.
BASZLIIS SIS. Sag 1,2*3 CO Grs.
Gas/Mi CO
15-9
205
118
r-4 O
r-l n
CS
•H
-1-3
. o i
'-> '-> O
CU HO
CM '
r-l S^
- d
32 ^
o - o
— — s
— ! U
r-i ^S 6-1 "
** S
'2. "
"~"
ej !-i
a =:
S f-l JJ
i a
3i-l O
iH C/3
n
en
o
W S
ja
s- 3
§"
Test Seausnce
C-87
-------
to
03
O
X f->
O O
CO SI
•4^ T3
I CO ^ O
- o n to
d co c a
T3 J I c5 p
O O C—O,
O • &-I
O lf\ O r-j O
p -Pa
JLPk 3
CO t|
Cj ^" O
9-
s
ir\
^r
r-
RT
o
vo
°aSES '-Z
H T
CO
CO
o
o
TITJ
TCTc <0^r
001 I
^o trt -^ -3^37^0 -i= Q^
-------
o
o
00
co
a
co
tr
Cd
2
o
CO
fc.
o
M
t^*
r^
co
*
«
*
o
o
*
»
•»
h-
o
o
^
0 I
1C H>
u.
O
H-
z
O LU
T O
c
lit
0.
o
-------
-------
APPENDIX B
LDV and LDT Operation
and Usage Characteristics
C-91
-------
Table B-l
Survival Rates
Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Survival
LDV
0.999
0.990
0.980
0.953
0.922
0.882
0.833
0.765
0.685
0*589
0.507
0.408
0.324
0^259
0.20-8
0..168
O.,142
0..134
0.126
0.118
Rates
LOT
1.000
0.999
0.996
0.972
0.940
0.917
0. 886
0.852
0.818
0.780
0.735
0.688
0.632
0.575
0 . 512
0.450
0.390
0.325
0.270
0.210
Source: Letter, David Lax, Energy and Environmental
Analysis, Inc. to Robert Johnson, U.S. EPA,
November 10, 1983. Survival rates used in the Ninth
Quarterly Report of the Highway Fuel Consumption
Model.
C-92
-------
Table B-2
Annual
Year
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Vehicle Miles
LDV
13,925
13,125
12,372
11,661
10,992
10,361
9,766
9,205
8,677
8,179
7,716
7,274
6,857
6,464
6,094
5,745
5,415
5,105
' 4,813
4,537
Travelled
VMT
LDT
17,739
16,425
15,208
14,082
13,039
12,074
11,180
10,352
9,585
8,875
8,218
7,610
7,047
6,525
6,042
5,595
5,180
4r797
4,443
4,113
Source: Draft MOBILES Documentation, data provided by Lois
Platte, U.S. EPA, QMS, February 14, 1984.
C-93
-------
Table B-3
Registrations and Vehicle In-Use Fuel Economy
Total Registration [11
Fuel Economy [2]
Year
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
LDV
9.530
8.460
9.960
10.610
11.480
8.790
8.200
9.750
10.820
10.820
10.450
8.831
8.051
7.661
8.165
9.200
9.744
10.070
10.454
10.582
10.521
10.457
10.683
10.818
10.818
10.728
10.818
10.818
10.818
10.818
10.818
10.818
LDT
1.492
1.423
1.627
1.973
2.484
2.183
2.020
2.799
2.790
3.229
2.741
1.943
1.664
1.814
1.995
2.328
2.563
2.661
2.854
2.768
2.779
2.668
2.792
2.865
2.866
2.871
2.939
2.939
2.939
2.939
2.939
2.939
LDV
14.91
15.26
15.0"
15.17
14.71
15.53
13.46
14.87
15.59
16.95
17.25
20.05
21.44
22.23
22.20
22.80
23.50
24.30
25.20
26.30
27.50
28.80
30.30
31.60
33.00
33.63
34.22
34.22
34.22
34.22
34.22
34.22
LDT
10.88
10.98
10.99
11.02
11.02
10.91
10.49
11.16
10.91
10.92
11.73
14.08
15.68
15.96
16.37
16-. 82
17.31
17.98
18.61
19.28
19.93
20.60
20.66
20.73
20.79
20.85
20.91
20.91
20.91
20.91
20.91
20.91
?1? "The Highway Fuel Consumption Model-Ninth Quarterly
1 Report, Energy and Environmental Analysis, Inc., for U.S.
Department of Energy, February 1983 (millions of vehicles)
[2] Historical and projected in-use fuel economies (mpg).
C-94
-------
Table B-,4
LDV Age-Odometer Usage Pattern Factor
Age
1
2
3
4
5
6
7
8
9
10
11
12
13 . "...
14
15
16
17
18
19
20
Odometer LDV
1.013
1.014
1.005
0.997
0.994
0.992
0.967
0.97.6
0.954
0.939
0.935
0.918
0.917
0.910
0.898
0.896
0.862
0.829
0.805
0.800
Source: "The Highway Fuel Consumption Model-Ninth Quarterly
Report, Energy and Environmental Analysis, Inc., for
U.S. Department of Energy, February 1983.
C-95
-------
Rate of Misfueling By Means of
Riel FiUer Inlet Tampering
Versus Mileage
Legend
o J/MLDV
I/M Ragraaaion
• I/M LOT
• NorH/WLDT
01234 5 6789 10
Vehicle Mileage In 10,000 Mile Increments
C-96
-------
Rate of Evaporative Canister Disablement
Versus Mileage
25-
g.
C=3
co 15-
-------
APPENDIX C
Tables from
"Manufacturing Costs and Automotive Retail
Price Equivalent of On-Board Vapor
Recovery System for Gasoline-Filling Vapors"
C-98
-------
TaM* 1
COMPARATIVE COSTS
EVAPORATIVE CONTROLS
BASELINE VEHICLE SYSTEM
Division
Component Vendor
Assembly Costs
Canister .- 2.698
Purge Control .740
Vacuun Signal
Theraal Switch
Liauid-Vapor .20
Separator
Fill Pipe ' .300
Seal Assembly
Hose .700
Tubing 1.200
Cost to
Customer
RPE
6.293
1.727
EVAPORATIVE CONTROLS ON-BOARD
VEHICLE FILLING VAPOR CONTROLS
Division
Vendor
Costs
5.396
1.480
2.631
I .441
.460 ! .910
.700 ' 1.894 ''
1.600 1.400
2.76 2.400.
Cost to
Cus toner
RPE
12.386 .
3.454
6.138
.1.028
2.123
4.410
3.200
5.42
Costs
Delta
Vendor
Division
2.698
.740
2.631
.441
.710
1 . 594
.700
RPE
6.293
1.727
6.138
1.028
1.953
1.194
. 1.600
1.200 ' 2.760
Hardvare .40
C Modifications
Vehicle Assembly .98
rOTAL
.92
1.104
.80
2.200
-.720
1.84
5.06
1.656
.80
2.200
.240
12.954
1.84-0
5.060
.552
29.855
C-99
-------
TYUe 2
RETAIL PRICE EQUIVALENT AT THZ VEHICLE LEVEL
Part
Plant
or
Vendor
Selling
Price R&D
Invest
Tools &
Equip
Corp
Allocation
.20 VC
Corp
Profit
.20 VC
Dealer
Markup
.40 SP
Canister 2.698
Purge Control .740
Vacuun Solenoid 2.631
Thermal Switch .441
Vapor-Liquid .910
Valve
Fill Pipe Seal 1,.894
.539
.148
.526
.088
.182
,378
.539
.148
.526
.088
.182
.378
2.517
.691
2.455
.411
.849
1.76
Vehicle
Retail
Price
Equiva-
lent
6.293
1.727
6.138
1.028
2.123
Jt.41.
C-100
-------
*n
_*>
_o
&
1-
>
eg
o
C
2
O
O
CO
>
. Q
H
H
C
>••
ce
£3
b
<
u.
21
<
j
<
o
fr-
ee
vt e
Q • ^4 4)
*3 Q.i-4 O
SS S- «-* —
Si O 9J I*
S» O so a,
^ C^
. -i 3C
i- "o o
O fa «M
U Bo •
• BS 3C
C* J^
u ca C
O C «N
U U •
C
00 p"*
e
o
o •
J- e.
K
Ed
i_i jg
C • *•>
« eo a
— lio C
a. S U
w
Q
01
u J= C
C U «T
Q U '
O
.0
93
•
a
Z
•
b
S
a.
CO O 1-4 -i
^^ • *? ^^ *(P
vo r» ss «r
«S CM
*^ ^D ^3 C^
£ ^^
ft O en o
— C O C
m C cr>
CO «» CC - O 4»
0 3 «J JZ
O 0. > H
0
<_l
C
CN
O
••4
O
0
o
sO
2
o
o
f>
l^
o
a.
a
>
I
-4 V
3 >
CT i-*
«4 a
u >
1
—
PS)
g
O
'
in
O
vO
(S
*
«N
sC
-T
*
0
t~>
ri
^^
«
-------
BILL OF MATERIAL
MANU7ACTi.fi.iNG COSTS
Component
Description
Canister Body
Grid
Int. Filter
Ext. Filter
Activated Carbon
Canister Cap
Canister Connectors
Assembly
TOTAL
Purge Control Valve
Valve Body
Diaphragm
El. Connector
TOTAL
Vacuua Solenoid
Housing
Cap
Coil
Armature
Spring
Vacuum Springs
Diaphragm
Vacuum Hose
Air Nozzle
Solenoid Connector
Material
DB437
67
AY332
KZ1-4
5448
DB437
DB437
Assam.
DB437
ED38
1742
As sea
Alum.
Alum.
1752 Cu-Fe
Fe-HPM
820
820
ED38
Aluminum
Fe-HPM
Cu-1752
Weight
Founds
.30
.10
.10
.20
.50
40
. .05
—
—
.20
.01
.0.2
—
.20
.10
.05
.03
.01
.02
.01
.03
.02
.01
•Material
Costs
.24
.04
.10
.20
.50
.08
.04
M
1.200
—
.16 "
.01
.02
• 19
_
.300
.150
.075
.045
.010
.016
.010
.045
.020
.010
Labor
Overhead
Hrs. 1.40 '
.10
.01
.01
.001
.01
.001
.01
.001
.01
.001
. .01
.001
.01
.001
.10
• 95
.05
.01
.01
.100-
. 100
.100
.150
.020
.010
.010
.010
.050
.010
.005
.14
.014
.014
.014
.014
• 014 •
.014
.14
.07
.07
.014
.014
.140
.140
.140
.070
.028
.014
.014
.014
.070
.014
.007
Mfg.
Costs
.480
.054
.124
.224
.524
.104
.064
.240
1.820
.120
.280
.034
.044
.478
.240
.540
.390
.195
.093
.034
.044
.034
.165.
.044
.022
Reference
Current Siz
See Sketch
•»•
See Sketc
See Sketc
TOTAL
.681
1.801
C-102
-------
S -3.0 ( e S
BILL OF MATEXIAL
MANUFACTDSING
'oaponent
Ascription
"hermal Switch
•S - Body
pring
Tsytal
onnector
OTAL
apor-Liquid
eparator
ousing
tg. Flange
oao
-Ring
ronnet
sal
loac
Material
Assent*
Aluminum
820
Br-Si
Cu 1752
As sen.
1010 Steel
1010 Steel
Absorb
ED38
ED38
ED38
Steel
Weight
Founds
-
.05
.01
.01
.01
-
.20
.10
.05
.01
.02
.01
.05
Material
Costs
-
.075
.010
.030
.010
.125
.160
.080
.050
.005
.020
.005
.040
COSTS
Labor
Overhead
.03
.02
.005
.010
.005
.070
.01
.020
.010
.030
.005
.010
.005
.010
.042
.028
.007
.014
.007
.098
.014
.028
.014
.042
.007 .
.014
.007
.014
Mfg.
Costs Reference
.072 See Sketch
.123
.022
.054
.022
.293
.024 See Sketel
.208
.104
.122
.017
.044
.017
.064
OTAL
-.360
.10
.140
.600
C-103
-------
BILL OF MATERIAL
MANUFACTURING COSTS
Component
Description
Fill Pipe Rotary Seal
Lead Restrictor
Guide
Rotary Seal
Material
« Assent.
Steel
plastic
ED38
Weight
Pound s
.30
.20
.10
Material
Costs
.15
.20
.10
Labor
Overhead
.120
.220
.020,
.010
140
280
028
014
Mfg.
Costs
.240
.630
.248
.128
Refer en'c
Fig. 5A
TOTAL
.45
.330 .462 ' 1.246
C-104
-------
APPENDIX D
DETERMINATION OF IN-USE EMISSION REDUCTION
BENEFITS OF STAGE II PROGRAMS
-------
-------
I.. METHODOLOGY
Chapter 5.0 of this document sets forth the emission reduction
benefits which would be attained by each Stage II and onboard control
program option if the control systems performed optimally -i.e., at the
efficiencies achieved during the California Air Resources Board (ARE)
certification tests, in the case of Stage II systems, and in the American
Petroleum Institute (API) demonstration tests, in the case of onboard
control systems. The emission reductions which would actually be
achieved in the field by each program option are set forth as a function
of the level of effort employed in enforcing each option — the level of
effort being expressed in terms of frequency of in-use inspections, as
well as in terms of person-years of resources necessary to perform the
inspections and the associated .legal follow-up work. Appendix C deals
with in-use (with tampering) estimates for onboard controls. This
appendix deals strictly with Stage II controls. The input data for
this Stage II analysis were derived as follows:
1. Potential failure modes were identified for each type of vapor
recovery system (balance; hybrid; and vacuum assist). These failure
modes were grouped under the three general headings of misinstallation,
improper maintenance and tampering. See Table D-l for a list of failure
modes and Part II of this Appendix (p. D-17) for,an explanation of the
system defects covered by each failure mode.
2. The average percentage reductions in control system
efficiencies attributable to the various failure modes were estimated.
See Table D-2 for figures, and Part II for assumptions.1
1
The numbers appearing in Table D-2, as well as those appearing in
Table D-3 (see §3, immediately following in text) represent the best
estimates available at the present time. They are based on the data
presently available respecting Stage II technologies, or are estimates
or extrapolations based on existing knowledge. These numbers are
subject to future revision as refinements occur in the state-of-the-art
control systems and as further information becomes available regarding
the in-use performance of the various control systems.
D-3
-------
TABLE D-l
Potential Failure Modes*
Mi si retaliation
Improper Maintenance
Nozzles
Hoses
Processing Unit
Aspirator Units
Canister System
Fill pipe Seal
Tampering
Nozzles
(user-convehience,
economics or
fuel-switching
motivated)
Nozzles, hoses
(alleviate
misinstallatibn
problems)
Processing unit
Canister System
Fill pipe Seal
Stage II Systems
Balance Hybrid Vacuum-Assist
XXX
X
X
X
X
X
X
X
X
*Explanation of the specific defects covered by each failure mode appears in Part
II of this Appendix at pp. D-17 to D-44.
D-4
-------
TABLE D-2
Average Percentage Reduction In System
Efficiency Per Failure Mode
Misinstallation
Improper Maintenance
Nozzles
Hoses
Processing Unit
Aspirator Unit
Canister System
Fill pipe Seal
Tampering
Nozzles
(User-convenience,
economics or
fuel-switching
moti vated)
Nozzles, hoses
(Alleviate
misinstallation
problems)
Processing Unit
Canister System
Fill pipe Seal
Stage II Systems
Balance Hybri d Vacuum-Assist
5%
22%
10%
10%
15%
10%
7%
100%
100%
100%
100%
1%
5%
50%
75% (Weighted
Average)
100%
100%
100%
D-5
-------
3. The percentages of in-the-field systems2 which would exhibit
the various failure modes, assuming there was minimal in-use enforcement
(i.e., assuming compliance was essentially voluntary)3 were estimated.
See Table D-3 for estimates, Part II for assumptions.
4. Using the failure mode rates from Table D-3, and the recovery
efficiency decrements frpm Table D-2, the average in-use recovery
efficiency for nozzle-fill pipe losses for the various types of systems
(minimal enforcement case) were calculated. See Table D-4 for results.
The calculations were made according to the following formula:
Average
In-Use
System
Efficiency
n
E
1 = i
[100
N
E
j = i
n
E
i = 1
(A,)
where
N
E
j = i
(T.E.
n = total number of permutations and combinations of failure modes
occurring with each control system type
A-} = percentage occurrence of each combination of failure modes
Ei = in-use system efficiency associated with each combination of
failure modes
N = total number of failure modes for each recovery system type
Bj = that portion of the percentage occurrence of each failure
mode not occurring in combination with any other failure mode
A "system." as used in this context, refers to the combination of
nozzle, return hose, and return line components peculiar to each
individual nozzle plus, in the case of manifolded Stage II systems, the
common components at the installation site. The number of "systems" is
thus equal to the number of nozzles.
3
"Minimal Enforcement," or "Voluntary Compliance" means, in the case of
Stage II programs, the situation resulting if virtually no resources,
State or federal, were allocated to program enforcement.
D-6
-------
TABLE 0-3
PERCENTAGES OF IN-FIELD SYSTEMS
EXHIBITING VARIOUS FAILURE MODES
(Minimal Enforcement Case)
Misinstallation
Improper Maintenance
Nozzles
Hoses
Processing Unit
Aspirator Unit
Canister System
Fill pipe Seal
Tampering
Nozzles
(user-
convenience,
economics or
fuel-switch ing
motivated)
Nozzles,
Hoses
(alleviate
misinstallation
problems)
Processing Unit
Canister System
Fillpipe Seal
Stage II Systems
Balance Hybrid
? TO-1 TO 1 ^%-7 fW
L* • +J tQ j, *J fO .L • fJfO I • -J fO
45%
5%
15-20%
.85-5%
11%
5%
30%
10-15%
.4-2.5%
Vacuum-Assist
0-2%
7%
2%
2%
10-15%
negl.
15%
D-7
-------
TABLE D-4
Average In-Use System Recovery Efficiencies for Nozzle-Fill pipe Interface
Losses (Minimal Enforcement Case, Steady-State)
Balance
68%
Stage II Systems
Hybrid Vacuum Assist
78% 69%
TABLE D-5 . . .
Calculation of Stage II Weighted Average System
Recovery Efficiency for Nozzle Fillpipe Losses
(Minimal Enforcement Case/Non-Installation Rate = Zero)
WEIGHTED AVERAGE
SYSTEM RECOVERY
EFFICIENCY
68% x 80%
78% x 15%
69% x 5%
54.4
11.7
3.5
7t5%~~
TABLE P-6
Rates of Total Noncornpliance (Failure to
Install) for Stage II Programs as Function
of Time (Minimal Enforcement Case)
Time Elapsed from Final
Compliance Deadline (Years)
3 and thereafter
Rate of
Total Noncompliance
40%
30%
20%
Overall Stage II
Program Efficiency
42%
49%
56%
D-8
-------
Ej = in-use system efficiency associated with each failure mode
I.E. = optimum system efficiency (i.e., the efficiency achieved in
the ARB or API tests).
The quoted formula gives the weighted average of the efficiencies
of all in-use systems of a given type. The first part of the.
formula represents the weighted average efficiency of those systems
containing a combination of two or more failure modes. The second
represents the weighted average efficiency of systems with single
failure modes; the third, of systems with no failure modes. In
the case of combined failure modes, the net in-use efficiency is
determined by multiplying the efficiencies associated with the
single failure modes making up the combination. This procedure
involves the dual assumption that the effect of any failure mode
is independent of the effect of other failure modes and that the
effect of simultaneous failure modes is multiplicative. The
distribution of failure modes among systems was assumed to be
random, except in those instances where the types of failure mode
were mutually exclusive.
Sample Calculation - Assume that a group of balance systems has a 40%
rate of improperly maintained nozzles, a 20% rate of improperly
maintained hoses, and a 30% rate of tampered-with nozzles. The
average in-use system efficiency would be determined as follows:
Average
In-Use = 8% (.95 x .78 x .9) + 32% (.95 x .78) + 6% (.95 x .90)
System [Nozzle, Hose Defects] [Nozzle Defects] [Hose Defects]
Efficiency
+ 30% x 0 + 24% x .96
[Nozzle Tampering] [No Defects]
Average
In-Use
System
Efficiency
= 57%
,5. The aggregate in-use recovery efficiency (% of fill pipe losses
actually recovered) of Stage II systems, assuming there was no absolute
noncompliance (outright failure to install systems) was calculated,
D-9
-------
using the percentage efficiency for each type of system from Table D-4
and the following distribution of throughput coverage by system type:4
Balance - 80%
Hybrid - 15%
Vacuum - 5%
The results appear in Table D-5.
6. The percentage of gasoline-dispensing facilities absolutely
failing to comply with Stage II regulations (by not installing any
system whatsoever) in a minimal-enforcement situation, was estimated as
a function of time. The results appear in Table D-6. Bases for
projections are set forth in Part II of this Appendix (p. D-37).
7. Estimates were then made of the impact on the failure mode
rates set forth in Table D-3, and on the absolute noncompliance rates
set forth in Table D-6 of various levels of enforcement effort" ~
expressed in terms of frequency of in-use inspections of regulated
outlets. Three scenarios are depicted: bi-annual, annual, and quarterly
inspections. The results appear in Tables D-7, D-8, and D-9, respectively.
The impact of the various levels of enforcement effort on misinstallations
and on absolute noncompliance (total non-installation) was determined,
in each case, by assuming that the level of enforcement resources
utilized in the installation phase of the program would be the same as
that utilized in later years to perform in-use inspections and that,
in the installation phase, these resources would concentrate upon
assuring proper installation of systems. The results with regard to
absolute noncompliance rates appear in Table D-12.
8. Using the data set out in Tables D-7, D-8, and D-9, and the
computational method described in Paragraph 4 above, the average in-use
system recovery efficiencies for nozzle-fill pipe losses for Stage II
systems were calculated. The results are set out in Table D-ll. Also
set out in that table is the weighted average system recovery efficiency,
calculated according to the method set out in Paragraph 5 above.
The throughput distribution set forth is based on estimates of the
distribution of systems currently being installed in the San Diego and
South Coast areas.
D-10
-------
TABLE D-7
Percentages of In-Field Systems
Exhibiting Various Failure Modes
(Steady-state,; Bi-annual In-Use Inspections)
Mis installation
Improper Maintenance
Nozzles
Hoses
Processing Unit
Stage II Systems
Balance Hybrid
Dir.* N.O.V. Dir. N.O.V.
1.5% to 9% .8% to 4.5%
27% 29% 5% 6%
3% 3% 2% . 3%
Vac.
Dir.
0
3%
1%
2%
Asst.
N.O.V.
to 1 .2%
. ' 4%
1%
. .2%.
Aspirator Unit
Tamperi ng
Nozzles
(user-
convenience,
economics or
fuel-switch ing
motivated)
Nozzles,
Hoses
(alleviate
system
problems)
Processing
Unit
13%
30%
6%
30%
10%
negl
negl.
negl.
negl.
6%
10%
negl. negl.
10%
*"Dir." signifies direct enforcement; "N.O.V." signifies notice-of-violation
enforcement. For explanation of these terms, see Part II of this Appendix.
D-n
-------
TABLE D-8
Percentages of In-Field Systems
Exhibiting Various Failure Modes
(Steady-state; Annual In-Use Inspections)
Stage
Balance
Dir.* N.O.V.*
Misinstallation ' .5% to 3%
Improper Maintenance
Nozzles 15% 21%
Hoses 2% 2%
Processing Unit
Aspirator Unit
Tampering
II Systems
Hybrid Vac.
Dir. N.O.V. Dir.
.3% to 1.5% negl.
3% 4% 2%
1% ' 2% 1%
2%-
30% 30%
Asst.
N.O.V.
negl
3%
1%
2% .
Nozzles 2%
(user-convenience,
economics or
fuel-switching
motivated)
Nozzles, Hoses negl.
(alleviate sys-
tem problems)
Processing Unit
5%
negl.
2%
2%
4%
negl
negl. negl. negl
1%
5%
*See footnote, Table D-7.
D-12
-------
TABLE 0-9
Percentages of In-Field Systems
Exhibiting Various Failure Modes
(Steady-state; Quarterly In-Use Inspections)
Stage
Balance
Dir.* N.O.V.*
Misinstallation negl . neql .
Improper Maintenance
. Nozzles 4% 8%
• Hoses .5% 1% •
Processing Unit
II Systems
Hybrid Vac.
Dir. N.O.V. Dir.
negl. negl. negl.
1% . 2% negl.
negl. .5% negl.
2%
Asst.
N.O.V.
negl.
1%
negl .
2%
Aspirator Unit
Tampering
Nozzles
(user-
convenience,
economics or
fuel-switch ing
motivated)
Nozzles,
Hoses
(a! leviate system
problems)
Processing Unit
negl
negl
5%
30%
negl .
30%
negl.
negl. negl
negl.
negl.
negl.
negl. negl
negl. 2%
*See footnote, Table D-7.
D-13
-------
TABLE D-10
Average In-Use System Recovery Efficiencies
(Nozzle-Fillpipe Losses) as a Function of
Method of Enforcement and Frequency of
In-Use Inspections
Freq. of
Inspections
Quarterly
Annual ly
Bi -Annually
Balance Hybrid
N.O.V.* Dir. N.6.V. ' Dir.
95% 95%
86% 90% 88% 91%
77% 84% 82% 87%
Vac.
N.O.V.
-
84%
76%
Assist
Dir.
95%
90%
85%
*See footnote, Table D-7.
D-14
-------
i-H
,_^
1
CD
LU
I
CO
•a:
(^ ^—
O) O
0 XJ
e c
O) +J
LU O) O
S CU
>> Q-
i- 4*^ (/)
• ~""*
0 1
§ sE »-H
•M LU
Wl M-
>)<4- O
00 O
O) C O
woe
=> -r- O
1 +J 3
e= o o-
i— i c eu
3 S-
<
•^
l-H C
I — 1 1 I 1
• ?o co
o-o co r.
OO U_
LO
o
.£
0
•r"
<+_
LU
s_
0
o
O)
1
«^^
10
oo
0)
^
o> • cn
> <*-
eC C
_ LU
X? ^Q ^5
d> O) CO CM
!c la oo oo
CD -i->
•i- oo
CO
IO >^
< • LO
> CT>
O O ^- en
>> O 00 C\J
=c • oo oo
^
i_ &5 »5
•r- O «*
, |Z
o£ H ^ §
0 OJ r- C
• O) •*-> (8 C
CT Q. S_ = ef
-------
TABLE D-12
Absolute Non-Compliance as Function of Time
Time Elapsed from Final
Compliance Deadline (Years)
2
3
4 and thereafter
Rate of Total Non-Compliance
30%
15%
5%
0%
D-16
-------
II. EXPERIMENTAL DATA AND ANALYTICAL ASSUMPTIONS
Tables D-l, D-2, and D-3
MISINSTALLATION/TAMPERING: NOZZLES/HOSES
Stage II -- The nature of this type of defect is improperly laid piping,
resulting in lack of proper drainback of condensed vapors and, thereby,
excessive back pressure in the vapor return system. Estimates received
by EPA of the proportion of balance systems which were misinstalled
during the first round of Stage II vapor recovery in California range
from 5 percent to 30 percent (Latter estimate is contained in ARB
Staff Report Accompanying Proposed Revisions to-ARB Suggested Vapor
Recovery Rules (October 22, 1977, at p. 10)). As a result of this
uncertainty, it was decided to express the proportion of balance systems
which would be misinstalled in any future Stage II program as a range,
using the midpoint for calculation purposes, as expressing the most
probable estimate. -
It was assumed that even in a minimal-enforcement scenario, the
major oil companies who have been involved with Stage II in California
would learn from experience and be able to cut the number of misinstal-
lations in half. (It was not believed that a lower figure could be
achieved due to difficulties in supervising the large number of
contractors involved in an en masse installation of Stage II systems).
It was assumed that independents and smaller concerns would experience
the same installation problems as were experienced in California. As
61 percent of stations fly the brands of majors, a range of 3.3 percent
to 20 percent1 represents the weighted average of expectable misinstalla-
tions for balance systems.
The effect on system efficiency of a misinstallation varies with
the extent to which the piping has been mislaid and the type of system
involved. By requirement, the balance system nozzles certified by the
ARB contain a device for shutting off the flow of fuel when the back
pressure in the vapor recovery return system exceeds a certain limit.
Based on information from air pollution control officials in California
hereafter, "California officials", the point at which balance system
1
E.g., 3.3% = .61 x 2.5% + .39 x 5%.
D-17
-------
back-pressure devices at the State and local levels are shutting off
appears to be in the range of 6 inches of water pressure. Based on
available information, EPA estimates that no more than 25 percent of
all misinstallations have the potential for producing this level of
back pressure. Accordingly, for the minimal-enforcement scenario, the
range of estimates for severe misinstallations at balance system sites
was determined to be .85 percent to 5 percent, and the range for non-
severe misinstallation was determined to be 2.5 percent to 15 percent.
To the extent that the liquid trap in a mi sinstailed system was so
large that it from time-to-time created a back pressure exceeding the
cutoff point, it is believed that a service station operator would, at
least in a minimal-enforcement scenario, solve the problem by either
replacing the vapor recovery nozzle with a conventional nozzle or
disconnecting the vapor return system (e.g., disconnecting the-vapor
return hose). This would result in a 100 percent loss of recovery
efficiency. Because of the deliberate nature of this presumed "fix"
for this problem, severe misinstallations are listed in Tables D-l,
D-2, D-3, and analogous tables under the heading of "tampering."
The effect on system efficiency caused by a misinstallation
depends on the amount of liquid present in the liquid trap at a given
time. The impact of a non-severe misinstallation on the efficiency of
a balance system was estimated as follows: First, it was assumed that
the non-severe misinstallations -- i.e., those where the maximum liquid
column would be the equivalent of at most 6 inches of water pressure --
were distributed in such a way that the average maximum liquid trap
would be 3 inches. Based on discussions with California technical
personnel, it was estimated that the loss in balance system efficiency
which could be expected from this size liquid trap would thus be in the
range of 10 percent. The efficiency loss resulting from a balance
system with this average 3 inch liquid trap would thus range from zero
percent to 10 percent, based on the amount of liquid in the trap at any
given time. No estimates of the frequency of a given size build-up of
liquid, nor of corresponding dispersion rates, were available; accordingly,
for purposes of estimation, the average value of 5 percent efficiency
loss was used.
D-18
-------
Some hybrid system manufacturers have adopted the practice of
certifying installation contractors, with installation training a
prerequisite to certification. Accordingly, it was assumed that the
misinstallation rate for the hybrid system could be reduced by 50
percent over the balance system rate. Hybrid systems, like balance
systems, are required to have backpressure shut-off devices. Hence, a
range of 1.25 percent to 7.5 percent for moderate-effect hybrid mi sin-
stall ations and .4 percent to 2.5 percent for severe-effect hybrid
mi sinstallations was estimated. As in the case of balance systems, a
deliberate "fix" resulting in 100 percent loss in system efficiency was
assumed for severe mi sinstallations in the minimal-enforcement scenario.
Likewise, the average maximum liquid trap was estimated to be in the 3
inch water pressure range. As hybrid systems continuously recirculate
fuel through the vapor return piping, it was estimated that the effect
on hybrid efficiency of this size liquid trap would continually approxi-
mate the outer limit of balance system losses — i.e.^ would approximate
10 percent.
EPA understands that the Gulf Science and Technology Company --
producer of the Gulf Hasselman vacuum-assist system, has, like some
manufacturers of the hybrid systems, adopted the practice of screening
installation contractors. Accordingly, it is believed that the rate of
misinstalled systems would be about the same for vacuum-assist systems
as for hybrid systems. However, the Gulf Hasselman system vacuum pump
has a dead-head vacuum in excess of 20 inches of water column.2
Accordingly, the percentage of vacuum-assisted systems whose recovery
efficiencies would be actually affected by the liquid traps associated
with mi sinstallations should be small. It was estimated that the
percentage of systems moderately affected woul d not exceed 1 percent
and that the corresponding reduction in recovery efficiency would be
nominal -- also no more than 1 percent. It was estimated that the
percentage of systems which would be severely affected by a misinstalla-
tion would be negligible, especially as vacuum-assist systems are not
required to incorporate back-pressure shut-off devices.
_
'Certification Test Report, Gulf-Hasselman VCP-2 Vapor Recovery System
(ARB), at p.3.~
D-19
-------
IMPROPER MAINTENANCE
A. Nozzles
Stage II - The nature of this defect is torn bellows or faceplate,
interfering with the nozzle's in-use recovery efficiency. In the steady-
state condition, the probability that any individual nozzle will have such
a defect and, thus, the percentage of nozzles which can be expected to
be defective, can be expressed mathematically as follows:
b
[Formula D-l]
P * 1 -
1-a+b
where
"P" is the probability that any nozzle will be defective at any
point in time.
"b" is the probability that, in a given time period, a defective
nozzle will be repaired or replaced with a non-defective nozzle.
"1-a" is the probability that a non-defective nozzle will become
defective within the given time period — i.e., 1-a represents the rate
of deterioration.
"b," the probability that a defective nozzle will be repaired or
replaced can be re-expressed as:
b = bl + b2, where
bl » the probability that a nozzle will be repaired or
replaced as the result of program enforcement, and
b2 = the probability that a nozzle will be repaired or
replaced for some other reason.
One "other reason" why a defective Stage II vapor recovery nozzle would
be replaced is the failure of some, aspect of the nozzle's fuel-dispensing
apparatus. The fuel-dispensing portion of vapor recovery nozzles is no
different than that of conventional nozzles, the fuel-dispensing
apparatus of which eventually wears out or malfunctions. Based on
information obtained from nozzle manufacturers, EPA has concluded
that on average, a nozzle "wears out" or malfunctions at an age of
about 2 years. Thus, in the steady-state, b2 is equal to .0416. This
rate is assumed independent of the level of enforcement effort.
D-20
-------
The value of bl, of course, depends upon the level of in-use
enforcement effort. In a no-enforcement scenario, bl is equal to zero.
If in-use inspections were performed biannually, bl would be equal to .0416,
"1-a," the remaining factor in the equation, should, like "b2,"
be independent of the level of enforcement effort. A representative
value of this variable was determined using observational data obtained
in the South Coast area in October of 1979 and in the District of Columbia
area in October of 1978. In the South Coast, 150 certified balance
system nozzles were examined. These nozzles had been installed for
varying lengths of time, ranging from 1 to 8 months. The percentages
of nozzles with faceplate or bellows defects as a function of length of
time in-use were plotted as set forth in Figure D-l following. It was
assumed that a linear fit was appropriate for this curve on the hypothesis
that development of faceplate and bellows defects are, at least during
the periods of time involved, a random function of nozzle usage. The
slope of the best linear fit for the three data points yields a 4.2 percent
probability that any one nozzle will become defective in a given month.
In early October 1978, EPA personnel inspected 106 service
stations in the District of Columbia. The rates and sizes of nozzle
defects were recorded. The length of time the vapor recovery nozzles
had been in service at each station was not recorded. However, from
information obtained from District of Columbia Department of Environmental
Services officials, EPA was able to estimate that the systems had
been in-use between 20 and 25 months, and that, on average, the systems
had been in-use for 23 months.
Inspections were made on 198 first-generation, Emco-Wheaton Model 3003
nozzles. So far as bellows and faceplates construction is concerned,
these nozzles are similar to the Emco-Wheaton nozzle presently certified
by ARB for use with balance systems, the exception being that the ARB-
certified nozzle has a 20 percent greater bellows area on its unleaded
version. Of the 198 nozzles examined, four had catastrophic defects,
totally eliminating the vapor recovery function. Seventy-three had
defects .which, while not catastrophic, were deemed sufficiently substan-
tial to affect vapor recovery efficiency. Twenty-two had defective
bellows, forty seven had faceplate defects, and four had defects in
D-2.1
-------
FIGURE D-l
o
CD
O
-------
both areas. Thus, at first glance, the approximate figure for the rate
of defective nozzles would appear to be 33 percent (73 of 218).2a
This/figure is lower than the true rate, however. First, it does
not take into account the greater bellows area of the unleaded version
of the ARB-certifled nozzle, which is the nozzle that will actually be
used in Stage II programs. In addition, the District currently allows
the use of one conventional nozzle per product per service mode
{attendant/self-serve) at each station, and the 33 percent figure does
not take into account the fact that the vapor recovery nozzles inspected
were presumably underutilized due to the availability of the conventional
nozzles at the District's service stations. These two factors were
compensated for as follows: First, to account for defects which would
have been found had the conventional nozzles been Stage II nozzles, the
number of nozzles with bellows defects was increased by 20 percent —
from 22 to 27. Secondly, it was estimated that at least one vehicle in
fiye had deliberately chosen to use a conventional nozzle in preference
toii;vapor recovery nozzle^ This results in the need to multiply the
factual^throughput per nozzle; by a factor of 1.25 to equal the throughput
per'.nozzle vyhich woiild have p^ the absence ;pf the legal
: conyentional nozzTes. |y the same token, multiplying the total number
^ de^ec^ive noz2les by this same factor gives the number of nozzles
which would have been found defective if al 1 nozzles had been used
equally. The combined effect of these two compensations was to increase
the number of defective nozzles from, 73 to 98 [(73 +5) x 1.25].
Thus, the corrected rate of defective nozzles was determined to be 45
percent (98/218).
The rate of nozzle deterioration evidenced in the District was
determined from this one data point (45 percent of nozzTes defective,
at the 23-month point) by first developing a model for describing the '",.'-':
rate of nozzle defects as a function of time in a Stage II program. y
The model developed is depicted in Figure D-2 above. As can be seen,
the model has the number of defective nozzles rising more or less
linearly from zero to a certain maximum point, then falling to an
2a -- . •••..-.- • . ' . ••• .. •• -. ..• . , . : ••••
The total of 218 includes 20 conventional nozzles being used illegally.
(See discussion under Nozzles: Tampering, below at p. D-32).
D-2 3
-------
n 9 J 6flii'x
Srt,S,,,,t:S,2Gl LE £53 U 1,6 ; * §M[ : : ?.SO!f6 1^
sd :oJ teaqqs b fuow z
intermediate value*tnetbd steady38*atestt SaM^aM-Snsrsbowgtfoarrflns is
the"way the!cufye wtiuld; intact? njettavesio-'fa, eaiSe-iwbepe&tbentiozz^e ion
deteriorationffateTwas ab*ou.tr:eqdal ctoritRewsteadysstatetreplaGSfflfintn^ "io
rate to be:initheif4apeGqent»rangeysandisEI?.AIs6*nformafta'on3sbows ibater!*
nozz3e;rep1acementi'Rat§ iSJaftss app^^^Jraatieiy ^Bp^Kcentv^fssNi'nfibnsd'i'fi)
'•\s. p&Qlfigure;D^2 stiowscthe nQit)bersdf defiecti«§ ndzzilesopeaching a;ipeakon
.:-at about the ;20-month:pdjntf.and5Sbe£^steaaylsstatejyatoeybeJflga;Keaehedla*
about 32 months1. :This?i'sTthe?bena»torftwfitGhc'Oc6ars't,if rtheriwotn-oatsion
points of the fuel^'dispemsn-ngDapparatus^ol a;groupc'of£new5aokz3.esnat?eo3
distributed norma]lySaboot theiaverageonbzzTecli^
The; rate of nozzleadeteriorationvdemoitstfated?6y Stagsosl I'M
in the District wasi'determfhedsfromathesfenown 45recent5"23-m©nifi %
point "in a two-step pr6e§§IC3vFTrst,®theosteadyrf§
defective^noizles wa§ determinecl^by•-fitting ascurvi patterned"*'$foi
model'set but fn:Figare?S-2 %Q thi§ik^ownvdatasp51
steady-state^level bf^tntt flew ci5rv§iv6Tfiiqsteady^s*a'tefi^ye1&thusr;
determined Was:40 perellntl ^^Usiftl-tfiifj^valSei aftd
formula; Set out;at-pag^D2f giWdvev-slh^ietsHora^
tn h(=f-?'7R'- -s<: follbfti5'^. f fs "t! svf.tssleb bnuol' nssd avert b Fuow riarriw
UU DG t«/Oj do lUI lUWa-.•-"'- * "
if«OD ow? sesrlt to J:>9it1'9 bsnrdmoa 9rtT' .xrffiijps
Io isdroun sriJ
P
.4 = 1-
3t o3- t\ ,!
asw asfsson svrJosl-sb to s^s*1* bs^DsttoD sH*
.0416
l-a+.0416 -!81S\8e}
to
to 9te*t aril
Jnfoq'sJsb sno arr!,t. mo -ft
s'4? ATI sstimate
andJ):istricfco
of the- South Coast ?
(4.2SrX
cte da-ta ^iCliheR
feerl3f^a»|)ewtfent
6 o* oiss
rtrso as is son . fBnorJfl9l/fift»...QA :«;a0!.jf?,(H ,SIS .^p JaJpJ a
sS
-------
The South Coast and District of Columbia nozzle surveys were taken
at times when little or no in-use inspections of stations were occurring 3
From the average deterioration figure, then, the typical, steady-state
proportion of nozzle deficiencies in a minimal enforcement scenario can
be calculated using the probability formula. Specifically,
P = 1 - -0416 = 45
>.U33b + .0416
Accordingly, in a minimal enforcement scenario, EPA estimates that
the proportion of defective balance system nozzles would be approximately .
45 percent. J
The defects observed in the A-3003 nozzles, .as already noted were
of three types - catastrophic defects, faceplate rips, and bellows rios
Four of the 78 (73 + the 5 added to compensate for underutilization) ' '
nozzles exhibited defects so major that a 100 percent loss of recovery
efficiency was assumed. In the remaining 74 defective nozzles there
were 51 instances of faceplate defects deemed minor enough to have only
a marginal effect (assumed to average in the range of 10 percent) on
vapor recovery efficiency.
There were also 27 (22 + 5 compensating) instances of bellows
rips. The average bellows defect was determined to be a rip of 1.3 inches
m length. The average reduction in system efficiency resulting from
such a defect was determined by calculating the ratio of emission flow
through such a rip to. the uncontrolled emissions generated during a
refueling. This calculation was made as follows:
EFFICIENCY
REDUCTION
(as a fraction)
n r
ijo~ [Formula °-
where: Q = volumetric flow of vapor out of rip - (ft3/ sec)
Qo - total uncontrojled emissions.available at 6 gal/min
l .UU4 ft-Vsec) dispensing rate.
It was assumed that Q/Qo would be roughly the sa« at othe, dispensing rates
«••
D-25
-------
Restating,
Er = A x V
.0134 ft3/sec
where:
A s area of rip, and
V = velocity of emissions passing through the rip (ft/sec)
The standard rip was judged to be of rectangular shape with width equal
to 1/32 x L, where L is the length of the rip. (The proportion of width
to length is based on the engineering judgment that a 2-inch rip would
have a 1/16-inch cross section).
The velocity of flow out of the rip is expressed by the formula for
flow from a sharp-edged orifice at low pressure drops, namely:
V = cd
where:
Cd - coefficient of discharge of a sharp-edged orifice
= density of gasoline vapor
p = pressure drop across the orifice (i.e., rip)
Thus, restating:
Er = .0313 x L2 x C^ / 2 Ap
.0134 ft3/sec
L, as noted, was measured to be 1.3 inches.
CH = .60 Source: Marks, Mechanical Engineers Handbook
(5th ed.), p. 239.
* .0982 Ib /ft3. The vapors were assumed to be a mixture of 40 percent
air, and 60 percent propane. Sources of densities: 41 Federal Register
48053; Marks, op. cit., p. 1909.
p = '.1 inch of water. This is believed to represent a conservative
estimate of AD- According to data in ARB Exec. Order No. G-70-17,
certified balance systems operate roughly at .3"Ap at 6 gallons per
minute flow rates. It is known that a rip would reduce A_p_ and it is
D-26
-------
believed that a reduction to .1 inch water for a 1.3-inch rips if
anything, overstates the case.
Accordingly,
ER = (-0313) (1.3")2 (1. ft)2 (.60) / 2 (.1" HoQ) (5.20 lbf/ft2)
/ (1" H20)
(.0134 ft3/sec) (12 1n)2 / :—^
(.0982 1bm) (1bf sec2)
(ft3) (32.2 lbm ft)
0.3
Accordingly, the weighted average effect, on recovery efficiency
caused by defective nozzles was determined to be:
Weighted Average^
Efficiency Reduction = (51 x 10%) + (27 x 30%)' + (4 x 100%) = 22%
Per Defective Nozzle ^8 ~~
The expectable rate of defects in hybrid and vacuum-assist
nozzles (in a minimal enforcement scenario) was determined by examining
233 OPW-7V-A nozzles at District service stations. The bellows design
of the 7V-A nozzle is similar to that of Red Jacket and Hasselman system
nozzles. The differences are that the bellows on a Red Jacket nozzle
has a roughly 50 percent greater area than that of the OPW-7V-A and
the Red Jacket nozzle does not employ a faceplate; the Hasselman nozzle,
on the other hand, employs a different type of faceplate (concave metal
instead of flat rubber) than the 7V-A nozzle.
The number of 7V-A nozzles with defective bellows was determined
to be 15. As in the case of the Emco-Wheaton nozzles, this figure was
increased by 25 percent — to 19 — to compensate for the underutilization
attributable to the legal conventional nozzles in-use at D.C. service
stations. To estimate the number of defective Red Jacket nozzles which
would have been observed at District stations, the number was increased
by 50 percent again -- to 29 — to compensate for the greater bellows
areas of the Red Jacket nozzle. Accordingly, so far as the Red Jacket
system is concerned, the "measured" rate of nozzle defects at the
4
Does not sum to 74 since 4 nozzles exhibited both faceplate and bellows
defects.
D-27
-------
23-month point was 12 percent .(29/244).4a As was just done in the case
of the balance system, the steady-state number of deficient nozzles can
be determined .using the model curve set forth in Figure D-2 and the
12 percent/23-month data point. The extrapolated steady-state value
is 11 percent. The corresponding deterioration rate (monthly basis),
calculated using Formula D-l, is .5 percent.
In a similar manner, the proportion of Gulf-Hasselman system nozzles
which could be expected to be defective in the steady-state, in a no-
enforcement scenario, was determined to be 7 percent. The corresponding
Gulf-Hasselman deterioration rate (monthly basis) was determined to be
.31 percent.
The average size bellows rips in the OPW-7V-A nozzles was 1.6 inches
in length. Using Formula D-2, the effect of this size defect on the >
recovery efficiency of an OPW-7V-A used as a balance system nozzle would
calculate to 45 percent. A hybrid nozzle operates at a slightly negative
pressure, however, and a vacuum-assist nozzle at a substantially negative
pressure. Based on discussions with technical personnel in the San
Diego Air Pollution Control District (APCD) and the relative pressures
involved, it is EPA's judgment that the effect of a 1.6-inch rip on a
hybrid and a vacuum-assist nozzle would be in the range of 15 percent
and 5 percent, respectively.
B. Hoses
Stage II - The nature of this defect is kinking or flattening of the
vapor return hose with a resultant increase of back pressure in the
vapor return line. EPA's October 1978 survey at District of Columbia
service stations found the rate of such defects to be 18 percent
(77 hoses with defects, among 430 examined). This compares with a
29 percent defect rate measured by the California Air Resources Board
in a 1977 survey of service stations in the San Francisco Bay area.5
4a
The 244 figure includes 11 conventional nozzles being used illegally.
See "Nozzles: Tampering" discussion, below at p. D-32.
^Report: Harmon Wong-Woo to William Lewis, Subject: Field Survey of
Bay Area Air Pollution Control District's Phase II Vapor Recovery
Program; March 10, 1977.
D-28
-------
The causes of these hose defects appear to be twofold. First,
vehicles can run over that portion of a hose overhanging a service
station island. Second, flattening could occur at the hose/return pipe
interface when a hose was stretched to its full length at an angle to
the return pipe riser (portion of return pipe aboveground). At the
time of each survey, the stations in both the District of Columbia and
the Bay Area were using first generation Stage II technology. Among
other features of this technology is the use of the same length hoses
(12 feet) as are used with conventional nozzles. State-of-the-art
balance and hybrid systems, however, are required to use either 8-foot-
long vapor return hoses or standard length hoses attached to the dispenser
in an overhead retractor arrangement. Another feature which could be
required is the use of swivel connectors for attaching the vapor return
hose to the riser as well as for attaching the return hose to the
nozzle. Requiring 8-foot-long return hoses or an overhead retractor
arrangement, plus requiring the use of swivels, should have a significant
impact on the rate of hose defects. EPA estimates that requiring
such features would result in a 5 percent maximum rate of hose defects
in balance and hybrid systems in a no-enforcement scenario.
The impact of a hose defect on system recovery efficiency depends
on the degree of flattening or kinking. Defects in balance and hybrid
system hoses will run the gamut from ones causing only a slight impact
on efficiency up to defects of a size sufficient to trigger the balance
and hybrid back pressure shut-off mechanisms. It is recognized that
defects exceeding the upper bound of this range also will occur, but it
is assumed that hoses damaged to this extent will be repaired towing to
the effect of the damage on nozzle fuel-dispensing capability. The 5 per-
cent defect rate estimates refer only to defects within the specified range
It is assumed that the average defect will be midway in the
specified range — i.e., equivalent to the average misinstallation
defect. The impact of a given misinstallation defect, as already noted,
however, depends upon the amount of fuel caught in the liquid trap. A
crushed hose -- being a continual phenomenon -- is analogous to a
misinstallation where the trap is full. Accordingly, the effect of a
crushed hose is analogous to the effect of misinstalled piping in a hybrid
D-29
-------
system. This is true for both balance and hybrid systems. Accordingly,
the impact on balance and hybrid system efficiency of a hose defect is
estimated at 10 percent.
Vacuum-assisted system nozzles do not employ back pressure shut-off
mechanisms. Accordingly, there is no upper limit to the impact on
recovery efficiency which a hose defect might have. Impacts will range
from 100 percent, in the case of a totally collapsed hose, down to
negligible, in the case of a slight kink. It is assumed that the
average effect would occur at the midpoint of the range -- i.e., will
equal 50 percent.
Vacuum-assist systems use the same length hoses as conventional
nozzles. Accordingly, the gross number of crushed hoses would be
greater with such systems than with balance and hybrid systems. However,
because of the vacuum-assist system's negative operating pressure, it
is estimated that the proportion of hose defects capable of adversely
affecting vacuum-assist system performance will be low -2 percent at
most.
C. Processing Unit
The Hasselman vacuum-assist system processing unit consists of an
electrically powered blower, an electronically ignited incinerator, and
a control apparatus consisting of a number of solenoids and valves.
While the system is believed generally reliable, the electronic and
mechanical parts are, obviously, subject to wear and malfunction.
Accordingly, it was assumed that there would have to be some average
downtime associated with these units. A nominal 2 percent rate was
assumed. It was assumed that 50 percent of the downtime would be
downtime of the blower, with resulting 100 percent recovery efficiency
loss, and 50 percent would be downtime of the incinerator, with a
roughly 50 percent loss in recovery efficiency. The weighted average
recovery loss is thus 75 percent.
D. Hybrid Unit
The nature of this defect is miscalibration (by human error or
natural drift of equipment settings). In a study performed by the
South Coast Air Quality Management District, slightly over 30 percent
D-30
-------
of the units inspected were judged to be sufficiently out of calibration
to have an impact on system recovery efficiency. Fifty percent of the
units were miscalibrated so as to increase the vacuum at the nozzle-
fill pipe interface, and 50 percent miscalibrated so as to decrease the
vacuum. The engineer who conducted the South Coast test estimated that
a hybrid unit with a severe miscalibration of the second type would
collect about 75 percent of the vapors at the interface whereas a unit
with a severe miscalibration of the first type would suffer some
unspecifiable, but relatively small, incremental vent losses. On the
basis of this information, EPA estimates that the average efficiency
loss of a miscalibrated system of the second type would be 12 percent
(the midpoint between no loss and maximum loss) and the average loss
for a miscalibrated system of the first type would be perhaps 2.5 percent
(midpoint between zero loss and assumed nominal maximum loss of
5 percent). The weighted average efficiency loss of miscalibrated
systems in a minimal enforcement scenario calculates, therefore, to be
about 7 percent (.5 x 12%) + (.5 x 2.5%).
TAMPERING
A- Nozzles (Economics, Difficulties-of-Use, or Fuel Switching-Motivated)
This "deficiency" consists of the substitution of conventional
nozzles for required vapor recovery nozzles. In two important ways,
gasoline dealers will find vapor recovery nozzles substantially less
desirable than conventional nozzles. First, vapor recovery nozzles are
much more expensive than conventional nozzles. . A new conventional nozzle
runs about $45 to $50; a new balance system nozzle runs about $160.
A rebuilt conventional nozzle runs about $25 (net with trade-in). A
rebuilt balance system nozzle runs about $100 (net with trade-in).6
Secondly, in a number of different ways, vapor recovery nozzles
are less desirable to use than conventional nozzles. All vapor recovery
nozzles,for example, are to some extent heavier, bulkier and thus more
awkward to use with any vehicle than a conventional nozzle. Balance
Figures are all 1978 figures. EPA estimates that a nozzle must be
rebuilt (traded in) every 2 years on average.
D-31
-------
systems suffer from the added handicap of requiring exertion of pressure
(estimate of 5 pounds required) against the nozzle bellows spring tension
in order to activate and maintain the flow of fuel through the nozzle.
In addition to the general inconvenience problem, certain features
of balance system nozzles make refueling either vehicles in general or
some vehicles in particular more difficult. On some vehicles (e.g.,
Econoline vans), it is difficult to depress the fill pipe lead restrictor
no matter how much pressure is exerted against the bellows spring
tension. For those systems which have been installed with large liquid
traps, the back pressure shut-off features of balance system nozzles
will render refuel ings difficult or impossible whenever liquid has
built up in the trap.7 On vehicles with certain fill pipe/fill tank
configurations, the flow of fuel can be frequently interrupted (much to
the user's annoyance) because a system pressure build-up (created in
part by the tight-seal feature) either activates the nozzle back
pressure shut-off mechanism or causes fuel to activate the automatic
shut-off mechanism. Lastly, on vehicles with certain fillpipe
configurations (particularly side-fill vehicles), the weight of the
balance system nozzle precludes latching the nozzle securely in the
fillpipe and using it in the so-called "hands-off" mode of operation.
Given the sizable cost differences involved, and the substantial
differences in ease of use, it is not surprising that dealers would be
tempted to use conventional nozzles whenever they could. To the extent
enforcement was lax, therefore, one would expect to find a certain
number of the dealers using conventional nozzles on one or more of
their dispensers.
To estimate the rate at which this form of tampering might occur
in a minimal enforcement scenario, EPA personnel recorded occurrences
of the phenomenon during EPA's October 1978 survey of District of
Columbia service stations. At the stations which utilized Emco-Wheaton
Type A-3003 nozzles, there should have been a total of 218 vapor recovery
nozzles in place and functioning. The survey showed, however, that
three of the required nozzles had no bellows whatsoever, and that, in
The same phenomenon will occur to a lesser extent with hybrid systems.
D-32
-------
20 instances, a conventional nozzle had been substituted for the required
vapor recovery nozzle.8
At the stations utilizing OPW-7V-A nozzles, there should have been
a total of 244 vapor recovery nozzles in place and functioning, the
survey showed, however, that 12 nozzles had no bellows and that 11
conventional nozzles were being used illegally. Combining the figures
for the Emco-Wheaton A-3003 and OPW-7V-A stations, the overall observed
tampering rate was determined to be: '
Overall
Tampering
Rate
(D.C. Survey)
23 + 23 = 10%
218 + 244
The number at interest is of course the steady-state tampering
rate. To derive an estimate of this number from the 10 percent data-
point, it is necessary to make certain assumptions about what proportions
of that 10 percent are attributable to various motivations and how
much these proportions will change in the steady-state. In general, it
is assumed that most nozzle substitutions which would be attributable
to the inconvenience of the nozzles would occur soon after the nozzles
were put in place and that most of the nozzle substitution attributable
to economics would not occur until a nozzle had become defective due to
some failure of its fuel-dispensing apparatus. EPA estimates that
about three-fourths of the observed nozzle substitution is attributable
to inconvenience of use, and the balance to economics.
In the steady-state, given literally no enforcement whatsoever, it
is likely that tampering due to both rationales would increase very
substantially. For the minimal enforcement case, it was assumed that
the rate of nozzle substitution attributable to inconvenience of use
would not change materially in the steady-state.
8
The District currently allows the use of one conventional nozzle per
service mode provided there is at least one vapor recovery nozzle per
product per service mode (The two service modes referred to are attendant-
serve and self-serve). Any such "legal" conventional nozzles were
excluded from the count.
D-33
-------
So far as nozzle substitution attributable to economics is concerned,
a relevant consideration for the extrapola ion process is the length of
use of the nozzles in the District prior to the EPA tampering observations.
As noted above in the section discussing "Improper Maintenance: Nozzles,"
the nozzles at D.C. stations had been installed for only 23 months on
average and, further, were estimated to be 20 percent underutilized owing
to the presence of legal conventional nozzles. Under these circumstances
it is believed conservative to estimate a fourfold increase in the rate
of nozzle substitution attributable to economics in the steady-state,
minimal enforcement scenario. EPA thus estimates that the rate of
nozzle-switching in a minimal enforcement scenario would be 15 to 20 percent
at a minimum, with about 7 to 8 percent attributable to inconvenience
of use, and 8 to 12 percent based on economics. .
This rate of nozzle-switching pertains to balance systems. Hybrid
and vacuum-assist nozzles are much easier to use than balance-system
nozzles. It is believed that the differences in ease of use would be
sufficient to materially reduce the incentive for nozzle substitution
attributable to convenience considerations with these types of systems
— to perhaps one-third the rate prevailing in the case of balance
systems. On the other hand, it is believed that the incentives for
nozzle substitution attributable to economics would be about the same
for assisted systems as for balance systems. Accordingly, for assisted
systems, EPA estimates a steady-state nozzle substitution rate of
10 to 15 percent, with 2 to 3 percent attributable to inconvenience of
use and 8 to 12 percent attributable to economics.
B. Processing Unit
This "failure mode" consists of a user's deliberately turning off
the vacuum-assist secondary unit with resultant 100 percent loss of
vapor recovery efficiency. The incentive for shutting off a vacuum-assist
secondary unit would appear to be substantial. To begin with, there is the
fact that the unit consumes about $50 worth of electricity (annually) at a
typical station where such a system would be installed.9
Because of the relatively high capital cost involved, it is assumed
that vacuum-assist systems will be installed only at higher throughput
stations. The $50 figure is EPA's estimate for a nine-dispenser,
60,000 gal. per month station.
D-34
-------
More significantly, typical annual maintenance costs of the unit are
estimated at $330 per year.10 Particularly at outlets where maintenance
costs run higher than average, this level of expense will tend to induce
cost-saving system shutdowns. Processing units were found turned off
at 10 percent of the vacuum-assist equipped stations surveyed in the
South Coast California area. In a truly no-enforcement circumstance,
it is believed that in the long run the number of vacuum-assist processing
units which would typically be turned off would be quite high. Given
the rate observed in the South Coast, and the economic incentives, it
is believed that an estimate of a 15 percent minimum rate for this form
of tampering is reasonable for the minimal-enforcement case.
Table D-5
In order to determine the in-use efficiency of Stage II programs,
it is necessary to estimate the proportion of regulated throughout
which would be covered by each type of Stage II system. Officials in
the South Coast and San Diego areas — where systems are currently
being installed — appear to agree that balance systems are capturing
about 80 percent of the market (throughput basis), hybrid systems
are capturing about 15 percent, and vacuum-assisted systems about
5 percent. The balance systems, which require the lowest capital
outlays, appear to be favored by the major oil companies for self-serve,
as well as attendant-serve, applications. The assisted systems appear
to be favored by some independents, particularly at high-throughput
self-service operations. Because of their lower price, hybrid systems
appear to be capturing the lion's share of the assisted system market.
Table D-6
The estimates of total noncompliance rates were obtained by using
rates of noncompliance experienced with Stage I vapor recovery regulations
10
Source: EPA Stage II cost figures.
. . D-35
-------
as a baseline. In the summer of 1978, sources in EPA's Region II office
estimated that the total noncompliance rate at service stations was
12 percent, despite the fact that the Region had been active in enforcing
Stage I, and despite the fact that the regulation had then been in
effect for 2-1/2 years. The EPA Division of Stationary Source Enforcement
indicated at the same time that the then rate of total noncompliance
with Stage I regulations at small bulk plants was quite large — probably
well in excess of 40 percent. Given these two benchmarks, the 40 percent,
30 percent, and 20 percent estimates of total noncompliance appear
conservative, if anything:
1. Stage II regulations are substantially more onerous, both
economically and as a burden on service station operations,
than Stage I regulations (Stage I costs only about $900 per
station and requires Little, if any, operating and maintenance
costs; Stage II costs $7,000 at a typical station for.the
cheapest system, and costs several hundred dollars a year
to maintain). The economic impact of Stage I at bulk plants
(10-to 11-thousand-dollar investment) and, accordingly, the
rates of Stage II noncompliance, should be more akin to those
for Stage I at bulk plants.
2. The Stage I compliance figures reflect a situation where at least
one in ten stations per year were being checked for compliance,
whereas Table D-6 assumes a voluntary compliance scenario.
3. At the time Stage I noncompliance estimates were made, there
were not known to be any equipment-availability problems
associated with Stage I implementation. It must be assumed
that there will be some delays due to equipment shortages if
Stage II is implemented on a large scale.'n
Tables D-7. D-8, and D-9
Tables D-7, D-8, and D-9, set forth the rates of system defects which
could be expected to prevail in each of three scenarios where enforcement
is performed according to the strategy set forth in Section 9.0 of this
11
In 1976, for example, the Arthur Little Company estimated that an
18 to 24-month lead time would be needed to produce the requisite
number of nozzles if Stage II were required in only 11 AQCR's.
Arthur D. Little, Inc., Economic Impact of Stage II Vapor Recovery
Regulations, November 1976, at p. 204.~
D-36
-------
document, and where in-use enforcement inspections occur on a biannual,
annual, and quarterly basis, respectively. The defect rates reflect
EPA estimates of the impact on compliance of the enforcement strategy
pursued at the three different levels of intensity. A discussion of
the assumptions and judgments underlying these estimates follows.
Misinstallation
Stage II - The San Diego APCD has developed a strategy for insuring
that Stage II vapor recovery systems are installed properly. The
mechanism generally consists of making installation of a certified and
properly functioning system a prerequisite for a permit to operate a
station where installation of a vapor recovery system is required. The
vapor recovery aspect of the permitting process consists of the following
steps:
1. The service station owner applies for a permit to construct a
vapor recovery system at his location. The application must
set forth, among other things, an exact engineering specification
of the system to be installed.
2. The application is reviewed and a determination made whether
the proposed system is adequate. Upon such determination being
made, the permit to construct is issued.
3. Upon completion of construction, and prior to repaying the
service station, the APCD is notified. An on-site inspection
to determine the adequacy of the installation is then made.
In some cases, this inspection includes certain parameter
tests — particularly pressure/flow and liquid-blockage tests.
Alternatively, the installation contractor is required to
certify that these tests have been performed. If the instal-
lation meets requirements, the authority is given to repave the
station and to resume business operations.
4. After a certain operational period (somewhere between 30 and
90 days), the station is reinspected to ensure that the.
installed system is functioning properly. Upon satisfactory
completion of_this second inspection, a full permit to operate
is issued.
The misinstallation deficiency rates set forth in Tables D-7, D-8,
and D-9, were determined by assuming that a system modeled on the
San Diego mechanism would be used to monitor installation in the three
enforcement scenarios. It was further assumed that the control system
at any station subjected to such a procedure would be properly installed
D-37
-------
and totally free of defects. The only question to resolve, therefore,
was what proportion of stations could be subjected to the procedure in
each enforcement scenario.
To estimate this latter variable, it was assumed that the resources
available (measured in person-years) to monitor installation would be
the same as the resources available in the steady-state to conduct in-
use inspections of currently installed systems.12 The amount of time per
station available for installation monitoring was assumed to be .6 hours,
1.2 hours, and 4.8 hours for the biannual, annual, and quarterly
enforcement scenarios, respectively.
To assess the time available per station to conduct installation-
monitoring, it was necessary to estimate the amount of time required to
process a station through each stage of the procedure. The various
times were estimated as follows:13
1. Review application for permit to construct 3.0 hours
and prepare permit to construct
2. Conduct post-construction, pre-operational 1 hour,
inspection 50 minutes^
3. Conduct follow-up inspection after initial 1 hour,
operational period 10 minutes^
4. Prepare permit to operate 1 hour, 30 minutes
Total 7.5 hours
In actual practice, in order to avoid the situation where a dealer must
re-install vapor recovery upon a misintallation's being discovered after a
station has been repaved, it may prove desirable to temporarily utilize
additional resources, where necessary, to ensure that all installations can
be monitored fully in the first instance.
All figures include allotments of time for discussion of the program
• .I t _.__. • _i_i__. _A._J.^«...*I«*UB MiiAA-ic^AM** /% J« "ivtsi^ifi/iitoi c T* a T T ^ n
in the
San Diego area.
Consists of 30 minutes to examine visually the installed equipment,
30 minutes average per station (three nozzles randomly selected at every
other station at 20 minutes per nozzle) to conduct liquid blockage tests,
20 minutes to consult with the service station dealer and to fill out an
inspection form, and 30 minutes travel (Travel will have to be done on an
individual station basis owing to the need to conduct the inspection
before the station is repaved).
15
Consists of standard in-use inspection.
D-38
-------
It was assumed, in order to obtain optimum utilization of available
resources, that Stage II would be phased in equally over a 5-year period.16
Accordingly, the amount of time required per station for installation-
monitoring in each year of the 5 year-period would be 1.5 hours.
Comparing this 1.5-hour-per-station figure to the .6, 1.2, and 4.8 hours-
per-station time-available figures, the percentages of systems sure to
be properly installed (because fully installation-monitored) is estimated
to be 40 percent, 80 percent, and 100 percent, in the biannual, annual,
and quarterly enforcement scenarios, respectively. Accordingly, the
misinstallation figures in Tables D-7, D-8, and D-9, represent 40 percent,
80 percent and 100 percent reductions over the figures in Table D-3.]j
IMPROPER MAINTENANCE
Nozzles — Stage II - As set fort in the discussion of Tables D-l, D-2,
and D-3, under the subheading "IMPROPER MAINTENANCE: Nozzles, Stage-II,"
the steady-state percentage of defective nozzles can be expressed as:
P = 1 - bl + b2
1-a + bl + b2
where
"P" is the probability that any nozzle will be defective at
any point in time.
"1-a" is the probability that a non-defective nozzle will
become defective within the given time period — i.e.,
1-a represents the rate of deterioration.
"bl" = the probability that a nozzle will be repaired or
replaced as the result of program enforcement, and
"b2" = the probability that a nozzle will be repaired or
replaced for some other reason.
As discussed earlier, b2 turns out to be the rate at which nozzles are
refurbished or replaced as the result of failures of the fuel-dispensing
16
To the extent such a regime was not followed, supplementary installation-
monitoring resources would need to be utilized in the years where
installations would exceed a 5-year equal distribution.
Should an area provide adequate personnel to fully monitor installations,
the number of misinstallations should be negligible, regardless of the
frequency of in-use inspections.
D-39
-------
apparatus and is equal to .0416. On the assumption that every nozzle
found to be defective during an enforcement inspection will be quickly
repaired or replaced, bl can be simply stated as .0416, .0832, and
.33 for bi-annual, annual, and quarterly enforcement, respectively.
The only remaining factor needing to be re-examined in an in-use
enforcement context is 1-a -- the probability that a non-defective
nozzle will become defective in a given month. To what extent will the
rate prevailing in a no-enforcement context be modified in an in-use
inspection context? Given the financial disincentives to maintaining
nozzles properly, it is believed that little, if any, deterrence would
occur with an M.O.V. type of enforcement mechanism. With a "direct"
enforcement type of mechanism, on the other hand, the possibility of
obtaining a certain amount of deterrence presents itself. Discussions
with a number of local air pollution control officials, however,
support the conclusion that local officials may be reluctant to impose
a fine directly for each and every nozzle violation attributable solely
to a failure of maintenance. Rather, it appears that fines would be
directly imposed, mostly in cases where the service station operator is
perceived to be a willful "repeat offender." Since the capability to
detect willful offenders would be directly proportional to the frequency
of inuse inspections, one could expect the degree of deterrence to
vary with the frequency of inspections. It is estimated that deterrence
rates of 10 percent, 33 percent, and 50 percent, could be achieved with
biannual, annual, and quarterly inspections, respectively. Accordingly,
the factor 1-a is reduced by the appropriate proportion for these
scenarios compared to the value prevailing in the no-enforcement scenario.
Sample Calculation
Hybrid System/Direct Enforcement/Annual Inspections
= _bl + bZ
^a + bl + b2
P = 1 -
P = 1 -
.0416 + .0832
.0051 1.6/J + .0416 + .0832
= 1 - .97 = .03 = 3%
.
.1282
D-40
-------
Hoses -- It seems reasonable to assume that the impact of in-use
enforcement inspections upon maintenance of hoses will be proportional
to the impact of such inspections upon nozzle maintenance. Accordingly,
the "hose" figures in Tables D-7, D-8, and D-9, are in the same ratio
to the corresponding figures in Table D-3 as the nozzle figures in
Tables in D-7, D-8, and D-9, are to the nozzle figures in Table D-3
(The figures are rounded to the nearest half-percentage),
Processing Unit - As this figure refers only to system "downtime," it
would be unaffected by in-use enforcement.
Hybrid Unit -- According to the study performed by the engineering
staff of the South Coast Air Quality Management District, hybrid systems
appear susceptible to drifting substantially out of calibration even
when recently adjusted. See Phase II Vapor Recovery Evaluation Program
(SCAQMD), at pp. 28-29, 45-47. Indeed, the magnitude of the drift is
described as being comparable to that originally measured in the
District's study. _I_d_. ,-at 28. There does not appear to be any reason
to believe that the problems causing the drift are not correctable.
On the other hand, given the presently available data, it was decided
to treat this category of defect in terms of the worst-case analysis by
assuming that the impact of in-use enforcement inspections upon the
quality of hybrid calibrations would be relatively small and nonquanti-
fiable. Accordingly, the "hybrid unit" figures in Tables D-7, D-8, and
D-9, are the same as the figure presented in Table D-3.
Tampering
Nozzles, Stage II (User-convenience, economics or fuel-switching
motivated):
Stage II -- As in the case of improperly maintained nozzles, the dynamic
rate of nozzle-substitution in the steady-state context is expressed by
the formula {see p. D-19 supra):
P = 1 - b
1-a + b
Where nozzle-substitutuion is concerned, "b," the probability (in a
given month) that a substituted conventional nozzle will be replaced by
D-41
-------
a vapor recovery nozzle is equal to the probability in a given month
that a nozzle will be inspected in-use — i.e., is represented by the
inspection rate expressed on a monthly basis.
The factor "1-a," the probability in a given month that a vapor
recovery nozzle will be replaced with a conventional nozzle, is developed
as follows: First, it is assumed that, upon the inception of any in-
use enforcement program, nozzle-substitution, if it occurred at all,
would occur when the fuel-dispensing portion of the nozzle apparatus
wore out — i.e., when the nozzle was otherwise being replaced. With
that assumption, "1-a" can be expressed as .0416 x S, where "S" is the
percentage of service station operators inclined to indulge in the
substitution practice. It is assumed that the steady-state substitution
rates estimated in the subsection "Improper Maintenance Nozzles" (see
discussion of Table D-3 supra) is the same as the proportion of dealers
having a proclivity to substitute nozzles in a minimal-enforcement
scenario. EPA has made estimates of the impact which in-use inspections
would have upon the percentage of operators inclined to engage in nozzle-
substitution in both "N.O.V." and "direct" type enforcement scenarios
and these are set forth as follows:
Table D-13
S, Percentages of Dealers Will ing to
Engage in Nozzle Substitutionas a
Function of Enforcement Frequency
and type of Enforcement Mechanism
Balance
Inspection
Frequency
Bi-annual
Annual
Quarterly
Motivation for
Substitution
Inconvenience
Economics
Inconvenience
Economics
Inconvenience
Economics
N.O.V. Dir.
Assisted
N.O.V. Dir.
4-6
8-12
2-4
5-10
0-2
1-2
1-3
3-5
0-2
2-3
negl .
negl .
2-3
8-12
1-3
5-10
0-1
1-2
0-2
3-5
0-2
2-3
negl
negl
D-42
-------
From these figures for "S," the factor "1-a" was calculated. From the
determined values for "1-a" and "b," the various values of "P" reflected
in Tables D-7, D-8, and D-9, were calculated.
Sample Calculation -
Balance System/Annual Enforcement/N.O.V.-Type Enforcement Mechanism
P = 1 -
1-a + b
= 1 _ .0832
(.0416) x (.105) + .0832
= 1 -
.0832
.00436 + .0832
= 1 - '.95 = .05 = 5%
= 1 - -0832
.0876
Nozzles, Hoses (Alleviate Misinstallation Problems):
This form of tampering is.assumed to become negligible in the
steady-state condition under any frequency of in-use inspections. Once
the deficiency (misinstallation) is remedied, the incentive for the
nozzle-substitution would disappear. Reinspection of stations where
nozzles had initially been substituted would assure correction of the
deficiency.
Processing Unit -- Tables D-7, D-8, and D-9, set forth OMSNE's estimates
of the rates of noncompliance to which the 15 percent figure set forth
in Tables D-3 would be reduced as the result of in-use enforcement.
Both the N.O.V. and, particularly, the direct mechanism figures, show
substantial reductions. The reductions should be especially significant
with a direct type of enforcement mechanism, as switching off the
processing unit would undoubtedly be treated as a serious violation.
Tables D-10 and D-ll
The figures in Tables D-10 and D-ll are analogous to the figures in
Tables D-4 and D-5 and have been calculated in the same manner. In
Table D-ll and subsequent tables, the Stage II weighted average
efficiencies are broken down into Federal and State enforcement
categories, rather than into N.O.V. and direct enforcement categories.
•Federal enforcement is 100 percent of the N.O.V. type, while state
enforcement presupposes a 50 percent N.O.V. and 50 percent direct mix.
D-43
-------
Table D-12
Table D-6 sets forth estimates of absolute noncompliance (failure
to install a system) in a no-enforcement scenario, as a function of
time. The noncompliance rates were assumed to result from delays
attributable to lack of station owner diligence and/or to equipment or
installation supply problems, as well as from out-and-out lack of
cooperation. The steady-state noncompliance rate in the no-enforcement
scenario, and therefore the rate attributable to lack of cooperation,
was estimated at 20 percent.
As has been noted earlier, the first phase of enforcement would
consist of an information gathering visit to all facilities, one result
of which would be knowledge of when each facility could be expected to
meet the various regulatory compliance dates. It would add little time
to the installation-monitoring effort described above ("Misinstallation")
to revisit, for enforcement purposes, the uncooperative 20 percent of
the subgoup expected to install Stage II during each year of a five-
year phase-in program. As a result, with any reasonable enforcement
effort, it is believed that the portion of total noncompliance attribu-
table to lack of cooperation could be eliminated. That would leave the
portion attributable to lack of due diligence and/or logistics problems
which, by implication, is estimated to be 20 percent, 10 percent, and
zero percent in the first, second, and third years, respectively, following
the final compliance deadline. These would be the respective yearly
rates -- if the lack of cooperation could be eliminated immediately.
An assumption that a certain amount of foot-dragging, including legal
foot-dragging, would continue to occur, even in the presence of an
enforcement effort is reflected in the noncompliance rates set forth in
Table D-12.
D-44
-------
APPENDIX E
CUMULATIVE VALUES OF EMISSION REDUCTIONS
(1986-2020)
E-l
-------
-------
Table E-i. Cumulative VOC Emission Reductions (Mg/yr)
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
13%
1997
1998
1999
sum
2001
£082
2883
2004
2936
2087
2008
2009
2013
2011
2012
2013
2014
2015
201S
2017
2018
2019
2820
Cum. Total
NPV of Total
Bulk
Terminals
VOC
(Mg/yr)
0
17,222
50,272
65,088
63,906
62,471
61,627
60,867
59, 178
58,334
57,574
56,730
55,886
55,042
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
1,864,120
518,730
Bulk Plants
no ex.
VOC
(«g/yr)
0
26,754
78,097
101,114
99,278
97,048
95,737
94,556
91,933
90,622
89,442
88, 130
86,819
85,507
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
2,395,300
805,345
Bulk Plants
ex.
VOC
(Mg/yr)
0
23,998
70,052
90,597
89,050
87,051
85,874
84,816
82,463
81,286
80,228
79,051
77,875
75,699
75,640
75,540
75,640
75,648
75,648
. 75,640
75,640
75,640
75,640
75,640
75,640
75,640
75,540
75,640
75,640
75,640
75,640
75,640
75,540
75,640
75,640
2,597,579
722,831
Storage
Tanks
VOC
(Mg/yr)
0
6,381
19, 143
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
842,292
222,573
St 1-Nation
no ex.
VQC
(Mg/yr)
0
33,025
96,405
124,816
122,550
119,798.
118,179
116,722
113,484
111,865
110,408
108,789
107, 171
105,552
104,095
104,095
104,095
104,095
104,095
'104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104.095
104,095
104,095
3,574,752
994,749
St I-Nation
ex.
vac
(Mg/yr)
0
21,120
61,650
79,819
78,370
76,610
75,575
74,643
72,572
71,537
70,685
69,570
68.535
67,500
66,568
66,568
66,568
66,568
66,563
66,568
66.568
66,568
66,568
DO; 5ofl
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
2,286,031
636,136
E-3
-------
Table E-l. Cuauiative VOC Emission Reductions (Mg/yr)
Year
1985
1987
1983
1989
1990
1991
1992
1993
1994
1995
19%
1997
1993
1999
2000
2001
2002
2003
2804
2005
2088
2007
2008
2009
2018
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Cua. Total
PV of Total
St I-flll Nfl
no ex.
VOC
(Mg/yr)
2,819
5,491
5.343
5,188
5.894
4J979
4,912
4,852
4,717
4,550
4,589
4,522
4,455
4,387
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
156,860
49,814
St I-flll Nfl St
ex
VOC
(Hg/yr)
2,419
4,712
4.585
4,452
4,371
4,273
4,215
4,163
4,048
3,990
3,938
3,880
3,822
3,765
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3.713
3,713
3,713
134,598
42,058
II-Nation St
no ex.
VOC
(Mg/yr)
0
72,613
211,370
281,673
284,232
277,849
274,095
270,715
263,206
259,451
256,072
252,317
248,562
244,808
241,428
241,428
241,428
241,428
241,428
. 241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
8,266,963
2,287,545
II-Nation St
ex
VOC
(Mg/yr)
0
49,801
142,631
191,555
195, 100
190,719
188, 141
185,822
180,667
178,090
175,770
173, 193
170,616
168,039
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165.719
165,719
165,719
5,669,446
1,566,057
II-flll Nfl St
no ex.
VOC
(Mg/yr)
45,391
90,975
93,825
93,636
91,936
89,872
88,657
87,564
85,135
' 83,921
82,828
81,613
80,399
79, 184
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78.091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,891
, 2,814,849
869,598
II-flll Nft
ex.
VOC
(Mg/yr)
31.029
62,895
66,065
66,547
65,338
63,871
63,008
62,231
60,505
59,642
58,865
58,902
57,139
56,275
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55.499
55,499
55,499
55,499
55.499
55,499
55,499
55,499
55,499
55,499
55,499
1,9%, 880
614,676
E-4
-------
Table £-1. Cumulative VQC Emission Reductions (Mg/yr)
Year
1986
1987
1988
1989
1993
1991
1992
1993
1994
1995
1996
1997
1998
1999
2008
2081
2882
2003
2004
2005
• ' 2036
2087
2088
2009
2010
201 1-
2012
; 2813
2814
' ' 2015
281S
2017
2018
2019
2828
Cum. Total
NPV of Total
St II-Sel Nfl
no ex.
VOC
(Mg/yr)
15,046
32,151
33, 158
33,101
32,500
31,778
31,341
38,955
38,895
29,657
29,288
28,851
28,422
27,992
27,686
27,606
27,685
27,505
27,686
27,S06
27,586
27,686
27,586
27,686
27,686
27,686
27,686
27,685
27.686
27,606'
27,506
27,686
27,686
27,686
27,686
995, 875
387,411
St II-Sel Nfl
ex.
VGC
(Mg/yr)
11,225
22,752
23,899
24,873
23,636
23,105
22,793
22,512
21,887
21,575
21,294
213,982
20,678
28,357
28,876
28,076
28,876
20,876
20,876
28,876
28,876
20,876
28,876
20,076
20,876
20,076
28,875
20,875
28,875
28,076
20,876
28,875
20,876
20,376
28,075
722,363
222,356
St II-Nation
no ex.
Combo VOC
(Mg/yr)
0
72,513
188, 120
221,676
197,541
168,377
143,626
123,446
138,281
83,543
67,859
54,753
43,881
32,884
24,867
19,797
13,846
5,019
574
13
0
8
8
0
0
0
0
0
8
8
0
8
8 "
0
0 :
1,561,758
938,893
St I I -tot ion
ex
CoBibo VOC
(Mg/yr)
8
49,801
126,942
158,754
135,595
115,576
98,586
84,735
68,834
57,345
45,579
37,583
29,517
22,517
17,869
13,589
9,553
3,566
471
18
0
0
8
' • . 8
0
0
8
0
0
0
8
0
0
8
8
1,867,828
548,338
St II-flll Nfl
no ex.
Coabo VOC
(Mg/yr)
45,391
98,^975
83,585
73,692
63,895
54,462
46,456
39,929
32,436
27,022
21,949
17,718
13,909
10,611
8,843
3,695
613
155
5
0
0
0
0
8
0
0
0
0
0
8
8
0 '
0'
8
8
634,454
428.588
St II-flll NA
ex.
Combo VOC
(Hn/yr)
31,829
62,895
58,797
52,372
45,418
38,786
33,016
28,377
23,852
19,285
15,599
12,586
9,885
7,541
5,716
2,594
534
143
6
0
8
8
0
0
0
8
8
8
0
8
0
0
fl
0
8
447,555
381.349
E-5
-------
Table E-l. Cuaulative VOC Emission Reductions (Mg/yr)
Year
1965
1987
1388
1989
1338
1991
1992
1993
1934
1935
19%
1997
1998
1993
2000
2891
2022
2003
2004
2035
2006
2007
2808
2893
2810
2011
2812
2013
2814
2815
2816
2817
2818
2619
2820
Cum Total
NPV of Total
St II-Sel Nfl
no ex.
Co«bo VOC
(Mg/yr)
16,046
32,161
25,528
26,051
22,588
19,253
16,423
14,115
11,467
9,553
7,759
6,251
4,917
3,751
2,843
1,386
217
55
2
8
8
8
8
8
8
8
8
0
8
8
8
0
8
8
0
224,285
151,587
St II-Sel Nfl
ex.
Combo VOC
(Mg/yr)
11,225
22,752
21,270
18,345
16,427
14,002
11,343
18,255
8,333
6,347
5,543
4,553
3,576
2,728
2,068
375
133
52
2
0
0
0
0
0
8
0
0
0
8
0
0
8
8
0
0
161,385
103,012
Refueling Refueling Evao. Evap. Ref.+Evap Ref.+Evap
Era. Red." Em. Red. EM. Red. Era. Red. Em. Red. Em. Red.
Cura. VOC Cua. VOC Cum. VOC Cua. VOC CUB. VOC CUR. VOC
(No Tamo.) (W. Tarap.) (No Tamp.) (W. Tamp.) (No Tamo.) (M. Taap.)
(Xg/yr) (Mg/yr) (Mg/yr) (Mg/yr) (Mg/yr) (Mg/yr)
3
9
34495
S5385
92149
116538
138219
157854
173488
187533
199523
209940
£18353
£25555
£31174
235581
23838G
240332
242392
243134
243995
244395
244395
244395
£44395
£44395
£44395
244395
244395
£44395
244395
244335
244395
244395
244395
8
33694
63303
88544
111380
131812
147839
161862
1742B1
184698
193112
199923
205933
210340
215148
218753
221959
223562
223952
2£4363
£24353
224363
224353
224363
224363
£24363
224353
224363
224353
224363
224353
224363
224363
224353
0
0
26918
51878
75018
96388
117788
136630
153538
168878
182918
193838
284230
211900
219180
219180
219180
213188
213180
219188
219188
219180
219188
219188
213188
219188
219180
219188
219180
219188
213188
219188
219180
219180
219180
8
£6398
58440
72540
93210
112320
129610
144690
158340
178560
179538
188110
195268
200720
288720
200728
£08720
280728
208720
200728
£00720
200728
208728
288720
280728
200720
2887£0
£007£0
200728
200720
200728
200720
£00728
208720
8
8
61405
117175
167159
213568
£55999
293684
327818
356373
382433
403778 '
422583
437465
458354
454761
457556
459572
451572
452374
453175
463575
453575
463575
463575.
463575
463575
463575
463575
463575
453575
463575
463575
463575
463575
68384
113743
161084
284590
243332
£77449
386552
332621
355258
372642
388833
481193
411850
41586B
419473
422679
424282
424682
425883
425883
425883
425883
425883
425083
425083
425083
425083
425033
425083
425083
425883
425883
425883
5914319 6374748 5223238 5736120 13138843 12118868
1421343 1320882 1265531 1173718 2637844 2500528
E-6
-------
Table E-2. Cuaulative Benzene Reductions (Ma/vr)
Year
1988
1987
1988
1989
1993
1991
1992
1993
1994
1995
19%
1997
1998
1999
2683
2804
2885
2087
2009
2018
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Bulk
Terminals
Bz
<«g/yr)
8
103
332
391
383
375
370
365
355
358
345
340
335
330
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
325
325
326
11,185
3,112
Bulk Plants
no ex.
Bz
(Mg/yr)
8
161
459
607
596
582
574
567
552
544
537
529
521
513
506
506
5%
586
506
505
506
536
506
506
5S5
506
506
506
506
506
506
506
506
506
506
17,375
4,835
Bulk Plants
ex.
Bz
(Hg/yr)
0
144
420
544
534
522
515
589
495
488
481
474
467
46®
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
15,585
4,337
38
115
153
153
153
153
153
153
153
153
153
153
153
153-
153
153
153
153
153
153
153
153
153
153
153
153
153
153
153
153
153
153
153
153
5,354
1,335
St !-Nation
no ex.
Bz
(Mg/yr)
0
198
578
749
735
719
709
700
681
671
662
653
643
633
625
625
625.
625
525
625
625
525
625
625
625
625
625
625
625 ,
625
625
625
525
625
625
21,449
5,958
St I-Nation
ex.
Bz
(Mg/yr)
a
127
378
479
470
450
453
448
435
429
424
417
411
485
399
399
• 399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
13,716
3,817
E-7
-------
Table E-2. Cumulative Benzene Reductions (Mg/yr)
Year
ises
19B7
1SS8
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2901
2882
2003
2084
2005
2006
2097
2003
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
St I-flll Nfl
no ex.
Bz
(Mg/yr)
17
33
32
31
31
30
29
29
2B
28
28
27
27
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
941
294
St I-flll Nfl St II-Nation St II-Nation St II-flll Nfl St II-flll Nfl
ex
Bz
(Mi/yr)
15
28
28
27
26
26
25
25
24
24
24
23
23
23
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
808
252
no ex.
Bz
(Mg/yr)
0
477
1,388
1,850
1,867
1,825
1,800
1,778
1,728
1,704
1,682
1,657
1,632
1,608
1,585
1,585
1,585
1,585
1,585
. 1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
54,288
15,022
ex
Bz
(Mg/yr)
0
322
937
1,258
1,281
1,252
1,235
1,220
1,186
1,169
1,154
1,137
1,120
1,103
1,088
1,088
1,088
1,068
1,088
1,088
1,088
1,088
1,088
1,088
1,088
1,088
1,088
1,388
1,088
1,088
1,088
1,088
1,088
1,088
1,088
37,230
10,284
no ex.
Bz
(Mg/yr)
£98
597
616
615
604
590
582
575
559
551
544
536
528
520
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
18,485
5,711
ex.
Bz
(Mg/yr)
204
413
434
437
429
419
414
409
397
392
387
381
375
370
364
364
364
364
364
364
364
364
364
364
364
364
364
364
364
354
364
364
3£4
364
364
13,113
4,036
E-3
-------
Table E-2. Cumulative Benzene Reductions (Mg/yr)
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2094
2005
2006
2007
2008
2009
2910
£011
2012
2013
2014
2015
2016
2017
2018
2019
2020
St Il-Sel NR
no ex.
Bz
(Hg/yr)
105
211
218
217
213
209
206
203
198
195
192
189
187
184
181
181
181
181
181
181
- 181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
6,535
2,019
St Il-Sel Nfl
ex.
Bz
(Mg/yr)
74
149
157
158
155
152
150
148
144
142
140
. 138
136
134
132
132
132
132
132
132
132
132
132
132
132
132
132
. 132
132
132
132
132
132
132
132
4,744
1,460
St Il-ifation
no ex.
Combo Bz
(Mg/yr)
0
477
1,235
1,456
1,297
1,106
943
811
659
549
446
360
282
215
163
130
91
33
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10,256
6,160
St II-Nation
ex
'Coabo Bz
(Mg/yr)
0
322
834
990
890
759
647
555
452
377
306
. 247
194
148
112
89
63
23
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7,012
4,205
St 11-911 Nfl
no ex.
Combo Bz
(Mg/yr)
298
597
548
484
420
358
305
262
213
177
144
116
91
70
53
24
4
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4,166
2,814
St II-flll Nfl
ex.
Combo Bz
(Mg/yr)
204
413
386
344
298
254
217
186
151
126
83
65
50
38
18
4
1
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2,939
1,979
E-9
-------
Table E-2. Cuaulative Benzene Reductions (Mg/yr)
Year
1986
1987
1988
1989
1999
1991
1992
1993
1994
1995
1995
1997
1998
1999
2389
2001
2002
2003
2004
2005
2086
2007
2803
2809
2818
2811
2012
2813
2814
2815
2816
2817
2818
2819
8828
St II-Sel Nfl
no ex.
Cofflbo Bz
(«g/yr)
105
211
194
171
148
126
108
93
75
63
51
41
32
25
19
9
1
0
0
0
8
0
0
0
9
0
0
0
8
0
8
0
0
0
9
1,473
995
St II-Sel Nfl
ex.
Conbo Bz
Wg/yr)
74
149
140
124
108
92
78
67
55
46
37
38
23
18
14
6
1
0
0
8
0
0
0
0
0
0
0
0
0
0
8
0
0
0
8
1,053
716
Refueling
Bz Em Red
Cua.
(No Tamp.)
(Mg/yr)
0
0
228
431
608
769
912
1937
1145
1238
1317
1388
1441
1489
152S
1555
1573
1587
1608
1685
1619
1613
1613
1613
1613
1613
1613
1613
1613
1613
1513
1513
1613
1613
1513
45638
9381
Refueling
Bz En Red
Cua.
(U. Tamp.)
(Mg/yr)
Hi
%
222
418
5(54
735
865
976
11968
1150
1219
1275
1319
1359
1388
1420
1444
1465
1475
1478
1481
1481
1481
1481
1481
1481
1481
1481
1481
1481
1481
1481
- 1481
1431
1481
42873
8717
Evao. tvap.
Bz EM Red Bz En Red
Cua. CUB.
(No TaraD.) (W. Tamp.)
(Mg/yr) (Mg/yr)
0
178
342
495
. 549
777
902
1013
1115
1207
1279
1348
1399
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
1447
41873
8359
9
0
174
333
479
515
741
855
955
1045
1126
1185
1242
1289
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
1325
37358
7785
Onboard
Ref.+Evao.
Bz EM Red
(No Tatap.)
(Mg/yr)
0
405
773
1103
1410
1630
1938
2158
2352
2524
2655
2789
2887
2972
3001
3020
3033
3046
3052
3057
3050
3050
3050
3050
3060
3050
3060
3060
3060
3060
Onboard
Ref. +Evap.
Bz Ea Red
(W. Tamp.)
(Mg/yr)
0
0
397
751
1052
1350
1686
1831
2023
2195
2345
2459
2561
2548
2713
2745
2769
2790
2803
2803
2S8&
2806
2806
2805
2886
2806
28%
2806
3050
3060
3060
86711
17740
2806
2886
2306
28%
2806
2386
79932
16584
E-10
-------
Table E-3. Cuaulative EDB Emission Reductions (Mn/vr)
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2800
2001
2082
2003
2004
2005
2006
2007
2808
2009
2010
2011
2012
2013
2014
2015
2016
2017
2818
2019
2020
St I-Nation
no ex.
EDB
(Mg/yr)
8.880
8.401
1.868
1.234
1.865
8.942
0.799
0.555
3.543
8.437
0.361
0.298
8.217
0.146
0.072
8.072
0.072
0.872
0.87S
8.072
0.072
0.872
0.072
0.872
8.872
0.872
8.872
0.872
0.072
8.872
8.872
0.872
0.072
0.072
0.072
9.659
5.239
St I-Nation
ex.
EDB
(Mg/yr)
0.380
0.256
3. 678
8. 789
8.681
8.683
0.511
0.419
8.347
8.288
0.231
8.186
0.139
0.093
0.046
0.846
0.046
0.846
0.046
0.846
0.346
8.846
0.846
8.046
3.846
0.846
0.046
8.046
8.846
8.846
8.846
8.846
0.846
8.046
8.846
6.177
3.350
St I-flll Nfl
no ex.
EDB
(Mg/yr)
0.838
8.867
0.859
8.851
0.844
8.039
8.833
0.827
3.023
0.818
9.H15
0.812
3.009
8.806
8.003
0.803
8.803
0.883
0.803
0.803
8.803
8.803
0.883
0.803
0.803
8.003
8.883
0.083
0.803
0.883
0.003
0.803
0.383
8.803
3.383
8.584
8.313
St I-flll Nfl
ex
EDB
(Mg/yr)
8.033
8.857
3.350
8.844
8.336
0.834
3.329
8.023
0.819
8.016
0.813
8.018
0.808
0.885
8.003
8.883
0.803
0.033
0.803
8.803
0.803
0.803
0.803
0.003
0.003
0.803
0.803
0.803
0.003
8.803
0.303
8.803
8.803
8.883
3.383
8.433
8.269
St II-Nation
no ex.
EDB
(Mg/yr)
0.003
8.998
2.610
3.129
2.776
2.455
2.082
1.706
1.414
1.139
0.939
0.756
8.565
8.379
3.187
0.187
8.187
0.187
8.187
0.187
0.187
0. 187
8.187
8.187
8.187
0.187
8.187
8.187
8.187
0.187
3.187
8.187
8.187
8.187
3.187
24.872
13.488
St II-Nation
ex
EDB
(Wg/yr)
8.088
8.668
1.761
2.128
1.905
1.685
1.429
1.171
8.971
8.782
0.645
0.519
8.388
0.260
0.128
8.128
8.128
0.128
8.128
8.128
3.128
0.128
0.128
0.128
8.128
8.128
3.128
8.128
8.128
0.128
8.128
0.128
8.128
0.128
8.128
17.011
9.153
St II-flll Nfl
no ex.
EDB
(Mg/yr)
8.690
1.241
1.159
1.048
0.898
8.794
3.673
8.552
0.457
8.369
0.304
8.245
0.183
8.123
8.061
0.061
0.061
8.061
0.061
8.061
8.061
8.061
0.061
0.861
8.061
8.061
8.061
8.861
0.061
0.861
8.061
8.861
0.861
0.061
0.061
9.997
6.144
-------
Table E-3. Cuaulative EDB Emission Reductions (Mg/yr)
Year
St II-flll Nfl St II-Sel Nfl St II-Sel Nfl St II-Nation St II-Nation St II-flll Nfl
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
199S
1997
1998
1999
2809
mi
2102
20S3
2004
£865
£006
£807
£008
2009
8010
£011
2012
£013
2014
£015
£016
2017
2018
£019
2020
ex.
EDB
(Mg/yr)
0.471
8.658
0.816
ft.739
0.638
0.564
0.479
0.392
0.325
0.262
0.216
0.174
0.130
0.087
0.043
8.043
8.043
8.043
0.043
0.043
8.043
0.043
0.043
0.043
8.043
0.043
0.043
8.043
8.043
8.043
0.043
0.843
0.043
0.043
0.043
no ex.
EDB
(Mg/yr)
0.244
0.439
0.410
0.368
0.317
0.281
0.238
0.195
0.162
0.130
8.107
0.886
0.865
0.043
0.821
0.821
0.021
8.821
0.821
8.821
8.021
0.821
0.021
8.021
8.021
0.821
8.821
3.321
0.821
0.021
0.021
0.821
0.021
0.821
0.821
ex.
EDB
(Mg/yr)
0.171
0.318
8.295
8.257
8.231
0.284
8.173
0.142
8.118
0.895
0.878
3.363
8.847
0.332
8.815
0.816
8.816
0.016
0.816
. 0.816
0.016
0.816
0.016
0.316
0.816
0.816
0.016
8.316
0.016
8.315
0.816
8.016
0.816
8.815
0.816
no ex.
Cosibo EDB
(Mg/yr)
8.888
8.990
2.323
2.463
1.929
1.488
1.891
8.778
8.539
8.357
8.249
8.164
8.898
3.351
8.019
3.815
8.011
8.884
.088
.888
0.808
8.088
0.888
8.388
8.888
8.388
3.888
8. '388
0.888
3.383
8.888
8.388
3.800
8.888
3.888
ex
Combo EDB
(Mg/yr)
3.888
8.653
1.558
1.675
1.324
1.821
3.749
8.534
8.370
' 3.252
8.171
0.113
8.367
3.335
3.313
0.811
8.387
3.333
.388
.388
8.338
8.880
8.838
3.388
3.083
3.388
8.823
3.383
0.038
3.388
3.833
3.333
8.388
3.333
8.333
no ex.
Combo EDB
(Mg/yr)
8.698
1.241
1.031
0.319
0.624
3.481
8.353
3.252
0.174
3.119
8.881
3.853
0.832
0.816
3.385
3,383
.383
.380
.838
3.388
3.338
3.833
3.838
3.388
3.333
3.030 •
£.338
3. 338
3.3S3
2. -333
?.v38
3.338
' 3.380
3. 333
8. 333
St II-flll Nft
ex.
Combo EDB
(Mg/yr)
8.471
8.858
8.725
8.532
3.443
3.342
3.251
0.179
3.124
3.884
3.857
3.838
8.32E
3.312
•3. 884
3.382
.m
.338
.388
3.388
0.383
3.308
2.380
3.388
J.388
3.388
3.32-3
3, 383
3.32-3
3.333
3.238
3. 338
'J.M0'
3. 333
3, eft
7.855
4.320
3.534
2.172
2.552
1.563
12.58
8.55
8.58
5.S2
5,97
4.54
3.18
E-12
-------
Table E-3. Cumulative EDB Snission Reductions (Hg/yr)
St II-Sel Nfl St II-Sel Nfl
no ex. ex.
Year Cosbo EDB Combo EDB
(Hg/yr) (Mg/yr)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2933
2001
0.244
0. 439
0.365
0.289
0.221
0.170
0.125
0.089
0.062
0.042
0.028
0.919
0.011
0.006
0.082
0.001
0.171
8.31S
(3. £63
0.218
8.168
0.124
0.891
0.355
0.345
0.031
0.021
0.014
0.008
0.284
0.302
0.001
2085
2006
2007
2008
2010
2011
2012
2013
2014
2015
2016
2017
2018
20-19
2020
0.000
0.000
8,00?!
0.900
8. £08
0.000
1.52
1.15
E-13
-------
Table E-4. Cuauiative EBC Emission Reductions (."ij/yr)
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2808
2001
2002
2003
2085
2066
2097
2008
2009
2010
2611
2012
2013
2014
2015
2316
2017
2018
2019
2020
Bulk
Terainals
EDC
(Mg/yr)
0.00
2.14
5.65
6.58
5.68
5.82
4.26
3.49
2.89
2.33
1.92
1.55
1.16
0.78
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
51.47
27.91
Bulk Plants
no ex.
EDC
(Mg/yr)
0.00
3.32
8.77
10.22
8.82
7.80
6.61
5.42
4.49
3.62
2.98
2.40
1.80
1.21
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
79.95
43.36
Bulk Plants
ex.
EDC
(Mg/yr)
0.00
2.38
7.87
9.16
7.91
7.00
5.93
4.86
4.03
3.25
2.68
2.15
1.61
1.08
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
71.72
38.90
Storace
Tanks
EDC
(Mg/yr)
3.20
8.79
£.15
2.58
2.27
2.05
1.76
1.46
1.25
1.82
0.85
0.70
0.53
0.36
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.1.8
0.18
21.55
11.31
St I -Mat ion
no ex.
EBC
•Mg/yr)
3.30
4.10
10.83
12.61
13.89
3.63
8.16
6.69
5.55
' 4.47
3.68
2.97
2.22
1.49
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
3.73
0.73
8.73
98.69
53.53
St I-Nation
ex.
EBC
(Mg/yr)
3.38
2.62
E.33
3.07
5.36
6.16
5.22
4.28
3.55
2.66
2.36
1.30
1.42
0.95
0.47
0.47
0.47
0.47
0.47
0.47
0.47
3.47
0.47
8.47
0.47
0.47
0.47
0.47
3.47
8.47
0.47
0.47
0.47
0.47
0.47
63.11
34.23
St I-fill Mfl
no ex.
EDC
(Mg/yr)
8.39
0,53
3.63
3.52
3.45
3. 40
3.34
2.28
3.23
0.19
a. is
0.12
0,89
0.06
8.33
0.33
OS
0.03
0.03
0.03
8.03
0.03
0.03
0.03
0.93
0.03
8.83
0.93
0.93
0.03
0.93
S8.33
3.33
0.93
2.03
5.15
3.28
E-14
-------
Table E-4. Cuaulative EDC Emission Reductions (Mg/yr)
Year
1986
1987
1588
1989
1998
1991
1992
1993
1994
1995'
19%
1997
1998
1999
2882
2883
21904
2085
,2886
2007
2888
2889
2810
2811
2012
2813
2014
2815
2816
2817
2818
2019
2020
St I-flll Nfl
ex
EDC
(Sg/yr)
8.33
8.58
0.51
8.45
8.39
8.34
8.29
8.24
0.20
0.16
8.13
8.11
8.08
0.05
0.03
8.83
8.83
8.83
0.03
0.03
0.83
0.03
0.03
0.83
0.03
0.83
8.83
8.03
0.83
0.03
0.03
0.03
0.03
0.83
0.83
St II-Nation
no ex.
EDC
(«g/yr>
0.08
10.10
26.61
31.98
28.29
25.32
21.22
17.39
14.42
11.62
9.58
7.71
5.76
3.87
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
St II-Nation
ex
EDC
(Wg/yr)
8.88
6.81
17.95
21.59
19.42
17.18
14.57
11.94
9.98
7.97
6.57
5.29
3.95
2.66
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31.
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
St II-flll Nfl
no ex.
EDC
(Mg/yr)
.7.03
12.65
11.81
18.68
9.15
8.89
6.86
5.63
4.66
3.76
3.10
2.49
1.86
1.25
0.62
0.62
0.62
0.62
9.62
8.62
0.62
0.62
0.62
0.62
0.62
0.62
0.62
8.62
8.62
0.62
8.62
0.62
8.62
0.62
0.62
St II-flll Nfi
ex.
EDC
(Hn/yr)
4.80
8.75
8.32
7.54
6.58
5.75
4.88
4.88
3.31
' 2.57
2.20
1.77
1.32
8.89
0.44
8.44
0.44
0.44
0.44
8.44
0.44'
0.44
0.44
0.44
8.44
0.44
8.44
0.44
0.44
8.44
3.44
8.44
8.44
3.44
3.44
St II-Sel Nfl
no ex.
EDC
!!>!g/yr)
2.48
4.47
4.18
3.75
3.24
2.86
2.43
1.99
1.65
1.33 .
1.09
8.88
8.66
3.44
8.22
8.22
8.22
0.22
8. £2
8.22
8.22
0.22
8.22
0.22
0.22
8.22
8.22
0.22
3.22
8.22
3.22
0.22
8.22
0.22
3.22
St II-Sei Nfl
ex.
EDC
(Mg/yr)
1.74
3.16
3.81
2.73
2.35
2. '28
1.76
1.45
1.20
3.97
a.ae
8.64
8.48
0,32
3.16
0.16
3.16
3.15
3. IS
3.15
3. 16
8.15
3.16
9.15
0.16
8.15
3.15
8.16
3.15
0.16
3.16
0.15
3.15
2.15
(2.15
4.42
2.75
253.53
136.68
173.40
93.38
181.91
52.63
71.91
44.83
36.83
22.14
26.01
15. S3
E-15
-------
Table E-4. Cuaulative EDC Emission Reductions (Mg/yr)
Year
St II-Nation St II-Nation St II-flll NR St Il-flll Nfl St II-Sel Nfl St II-Sel NR
1986
1987
19B8
1989
1990
1991
1932
1993
1994
1995
19%
1997
199B
1999
em
2881
2882
2883
2004
2885
2886
mi
2888
2009
2818
2811
2912
2013
2814
2815
2816
2817
2818
2819
2028
no ex.
Coabo EDC
(Mg/yr)
8.88
18.10
23.68
25.18
19.66
15.16
11.12
7.93
5.49
3.74
2.54
1.67
1.88
8.52
8.28
8.16
8.11
0.84
.88
.08
8.88
0.88
8.88
0.00
8.88
8.88
8.08
0.08
8.00
0.80
0.08
8.88
8.80
0.88
8.08
128.22
87.11
ex
Coabo EDC
(Mg/yr)
8.08
6.81
15.98
17.07
13.50
18.41
7.63
5.44
3.77
2.57
1.74
1.15
8.68
0.36
0.13
0.11
0.88
0.83
.88
.08
8.88
8.88
0.88
8.08
8.00
0.08
0.08
0.00
0.80
0.00
0.08
0.08
0.80
8.00
0.88
87.46
59.34
no ex.
Combo EDC
(Mg/yr)
7.83
12.65
10.51
8.34
6.36
4.91
3.69
2.57
1.78
1.21
8.82
8.54
0.32
0.17
0.86
0.83
.80
.80
.00
0.90
0.00
0.08
8.38
8.08
8.38
0.08
0.00
8.08
8.80
0.80
0.00
0.88
8.88
8.88
8.08
60.98
46.25
ex.
Combo EDC
(Mg/yr)
4.88
8.75
7.48
5.93
4.52
3.49
2.56
1.82
1.25
8.86
0.58
8.38
8.23
0.12
0.05
8.82
.88
.88
.08
8.88
. 8.08
' 0.88
0.00
0.88
8.88
0.88
8.80
8.%
8.08
0.88
8.00
0.88
8.00
8.88
0.00
42.78
32.40
no ex.
Combo EDC
(Mg/yr)
2.48
4.47
3.72
2.95
2.25
1.73
1.27
8.91
8.63
0.43
8.29
8.19
0.11
0.86
8.82
0.01
.08-
.08
.00
8.00
8.80
0.®
0.00
8.89
0.88
8.00
0.08
0.00
0.00
0.00
0.98
0.88
8.88
0.00
0.88
21.53
16.35
ex.
Combo EDC
(Mg/yr)
1.74
3.16
2.58
2.15
1.64
1.26
8.92
3.66
2.46
8.31
2.21
0.14
a. 0s
0.84
8.82
0. 01
.m
.00
.88
8.08
8.28
%.W
8.80
0.1213
*,m
8.*B
8.N
0.80
8.88
0.83
9. 83
0.80
3.89
8.08
8.120
15.47
11.72
E-16
-------
Table E-5. Total Baseline Emissions Given No Additional Controls
Irm — — T-r-Fij
REFUELING
<«g/yr)
1982 407,308
1983 388,952
1984 365,893
1985 353,683
1986 341,866
1987 332,112
1988 323,158
1989 313,797
1998 388,899
1991 381,188
1992 297,116
1993 293,447
1994 285,387
1995 281,237
19% 277,574
1997 273,584
1998 269,434
1999 265,364
2888 261,781
2001 261,781
2882 261,781
2083 261,781
2804 251,781
2805 261,701
2886 261,781
2887 261.781
2088 261,701
2809 261.701
2818 261,781
2011 261,781
2812 261,781
2013 261,701
2014 261,781
2815 261,781
2016 261,701
2017 261,781
2018 251,701
2019 251.701
2828 261,781
VOC -
EMISSIONS
STflGE I
(Mg/yr)
621,088
581.256
558,279
539,649
520,398
506,736
493,074
478,791
478,897
459,540
453,338
447,741
435,321
429,111
423,522
417,312
411,182
484,892
399,383
399,383
399,383
399,383
399,383
399,303
399,383
399,383
399,383
399.303
399,383
399.383
399,383
399,303
399,383
399.383
399,383
399,383
399,303
399,383
399,383
TOTflL
(Mg/yr!
1,823,888
962,288
924, 172
893,332
361,464
838,848
816,232
792,588
778, 195
768,728
758,448
741,188
720,528
710,348
781,896
690,316
688,536
678,256
661,884
661,084
661,884
661.004
661,884
661,884
661,884
661.884
661 j 804
661,884
661,884
661,884
661,884
561.884
661,884
661,004
661,804
661,884
661,884
651,884
661,884
———"-—
REFUELING
(Mg/yr)
2,686
2,514
2,415
2.334
2,251
2,192
2,133
2,071
2,833
1.988
1,961
1,937
1,883
1,856
1,832
1,885
1,778
1,751
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1.727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
BENZENE -
EMISSIONS
STflGE I
(Mg/yr)
3,726
3,488
3,358
3,233
3,122
3,840
2,958
2,873
2,821
2,757
2,720
2,686
2,512
2,575
2,541
2.584
2,467
2,429
2,396
2,3%
2,396
2,3%
2,396
2,3%
2,395
2,3%
2,396
2.3%
2,396
2,3%
2,396
2,3%
2,396
2.3%
2,396
2,3%
2,396
2.396
2,396
TOTflL
(Hg/yr)
6,412
6.882
5,765
5,572
5,373
5,232
5,391
4,944
4,854
4,745
4,681
4,623
4,495
4,431
4,373
4,389
4,245
4,181
4,123
4,123
4,123
4,123
4, 123
4.123
4,123
4.123
4,123
4.123
4,123
4.123
4,123
4,123
4,123
4,123
4,123
4.123
4,123
4,123
4, i£3
REFUELING
(Mg/yrO
18
9
7
6
5
5
4
4
3
3
2
3
£
1
1
1
1
3
8
8
8
8
8
3
3
0
3
0
3
3
8
3
0
3
0
•8
8
••*
V
8
EDB
EMISSIONS
STflGE I
(Mg/yr)
14
12
18
a
• • . 7
5
5
5
4
4
3
3
2
2
1
1
1
i
t.
0
0
8
8
.0
0
0
0
0
0
8
0
0
0
8
3
3
8
fl
a
8
TOTflL
(Mg/yy)
£4
23
17
14
12
11
9
a
^
i
6
5
4
4
3
2
2
1
i
8
e
3
3
8
8
8
3
8
$
0
3
0
3
0
8
3
3
0
US
0
11155638 17335514 28282152
73693
182219
175912
59
94
153
E-17
-------
Taole £-5. Total Baseline Emissions Giver; No additional Controls
inn
1982
1983
1984
1985
1966
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998"
1999
2000
2001
2082
2003
2084
2005
2006
2007
2008
2009
2010
2011
2012
2013
2814
2015
2016
2017
2018
2019
2020
REFUELING
(Hg/yr)
104
88
74
62
53
46
41
36
31
27
23
19
16
13
10
8
6
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
euu
EMISSIONS
STflSE I
(Mg/yr)
140
119
101
84
72
63
55
48
42
37
31
26
21
17
14
11
9
6
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
TOTflL
(Mg/yr)
244
266
175
147
125
189
%
84
73
64
54
45
37
30
25
20
15
18
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
706
955
1651
E-18
-------
MINLMflL
VOC
NESHflP
EX.
(Mg/yr)
257,728
9
28,827
84.047
112,957
115,806
112,424
110,904
109,537
106,499
104,979
103,612
102,393
188.574
99,354
97,687
97,687
97,687
97,587
97.687
97,587
97,687
97,587
97,687
97,587
97,687
97,687
97,687
97,537
97,687
97,537
97,687
97,687
97,687
97,687
97.687
3,341,952
923.110
FflBLE E-6.
MINIMflL
VOC
flLL Nfl flREflS
NO-EX.
(Mg/yr)
121,448
26,730
53,628
55,286
55,196
54,194
52,977
52,251
51,617
•50,185
49,469
48,825
48,109
47,393
46,677
46,033
46,633
45,033
46,033
45,833
45,033
46,033
46,033
46,033
46,033
46,033
46,033
45,033
45,033
45,033
46,033
45,033
45,033
46,033
45,033
46,033
1,559,238
512,560
:RVICE STATION IN--USE
MINIMflL
VOC
flLL Nfl flREflS
EX.
(«g/yr)
86,312
18,303
37,895
38,961
39,227
38,515
37,650
37, 141
36,684
35,666
35,157
34,699
34,199
33,682
33, 173
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
1, 177, 159
362,381
MINIMftL
VOC
SELECTED Nfl flREflS
EX.
-------
Bl-ftNN.
VOC
SLIM AREAS
NO-EX.
(Hg/yr)
39,140
78,526
88,954
38,323
79,355
77,573
76,525
75,582
73,485
72,437
71,493
78,445
69,397
68,348
67,485
67,405
67,495
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
2,429,587
750,535
BI-fflN.
VOC
fiLL Nfl fiREflS
EX.
(«g/yr)
26,881
54,318
57,858
57,448
55,397
55,131
54,386
53,715
52,225
51,488
58,818
58,865
49,320
48,575
47,984
47,984
47,984
47,904
47,984
47,904
47,904
47,904
47,984
47,904
47,984
47,904
47,984
47,984
47,904
47,984
47,984
47,904
47,984
47,904
47,904
1,723,697
530,629
BI-fWN.
VOC
SELECTED Nfl flREflS
EX.
(Mg/yr)
9,695
19,649
20,637
20,779
28,401
19,943
19,674
19,431
18,892
18,623
18,388
18,111
17,841
17,572
17,329
• 17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
523,540
191,953
ANNUftL
VOC
NESHflP
NO-EX.
(Mg/yr)
9
65618
191346
254886'
257305
251527
248128
245869
238271
234872
231812
228413
225814
221615
218556
218556
218556
218556
218556
218556
2185%
218556
218556
218556
218555
218556
218555
218556
218556
218556
218555
218556
218555
218556
218555
.7483551
2078641
fiNNUfll
VOC
NESHflP
EX.
(Mg/yr)
0
44270
129073
' 173485
176617
172651
178318
168218
163551
161218
159119
156785
154452
152119
158019
158819
158819
150819
150019
158819
158819
158819
150019
150019
150019
158819
158819
158019
158819
158019
150019
150819
150019
158019
150019
5132283
1417633
ftNNUfiL
VOC
fill Nfl flREflS
NO-EX.
(Mg/yr)
42004
84272
36878
' 86737
85162
83249
82124
81112
78852
77737
76724
75599
74474
73349
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
2607362
805452
E-20
-------
ANNUAL
voc
LL NA AREAS
EX.
(Mg/yr)
28762
58293
61224
61643
68524
59165
58365
57646
56846
55S47
54527
53728
J2928
£129
(1489
J1409
51409
51489
51489
51489
51409
51409
51409
51489
51409
51409
51409 '
51409
51409
51409
51409
51409
51409
51409
51409
1349821
569455
ANNUAL
VOC
SELECTED NA AREAS
EX.
(Mg/yr)
10405
21087
22147
22299
21894
21403
21113
20853
28275
19985
19725
19436
19147
18857
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
669165
205998
QUflRTERLY
VOC
NESHflP
NQ-EX.
(Mg/yr)
0
68661
200246
266741
269273
263226
259669
256467
249353
245796
242594
239037
235480
231923
228722
228722
228722
228722
228722
£28733
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
7831623
2166950
QUARTERLY
VOC
NESHAP
EX.
(Mg/yr)
0
46329
135076
181554
184832
180681
178239
176842
171159
168717
166519
164078
161636
159195
155997
156997
156997
156997
155997
156997
156997
156997
156997
156997
156997
155997
156997
155997
156997
156997
155997
155997
155997
156997
156997
5370994
1483570
QUARTERLY
VOC
ALL NA AREAS
NO-EX.
(Mg/yr)
43913
38102
90827
90579
89033
87033
85857
84799
82447
81278
80212
79036
77850
76584
75525
75625
75525
75525
75625
75625
75525
75625
75625
75625
75525
75625
75525
75625
75625
75625
75525
75525
75625
75525
75525
2725878
842863
QUARTERLY
VOC
ALL NA AREAS
EX.
(Mg/yr)
30059
60942
64007
64445
63275
61854
61018
60266
58594
57758
57805
56170
55334
54498
53746
53746
53746
53745
53746
53746
53746
53745
53745
53746
53745
53746
53746
53746
53746
53746
53746
53746
53746
53746
53745
1933904
595339
QUARTERLY
VOC
SEL. NA AREAS
EX.
(Mg/yr)
10877
22046
23154
23313
22889
22375
22073
21801
211%
20894
20522
20319
20017
19715
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
19442
699581
215362
E-21
-------
KINIMftL
3z.
NESHftP
NO-EX.
(Mg/yr)
375,472
0
281
818
1,1993
1,1%
1,076
1,061
1,048
1,819
1,284
991
977
958
948
935
935
935
935
935
935
935
935
935
935
935
935
935
935
935
935*
935
935
935
935
935
32, Ml
8.855
MINIHflL
Bz.
NESHflP
EX.
(Mg/yr)
257,728
189
552
742
755
738
728
719
699
689
680
670
660
650
642
642
642
642
642
642
642
642
642
642
642
642
642
542
642
642
642
642
642
642
642
21,947
6,062
MINIMflL MINIMflL MINIMflL BI-flNN.
Bz. Bz. Bz. Bz.
flLL Nfl flREflS flLL Nfl flREflS SELECTED Nfl flREP.S NESHflP
NO-EX. EX. EX. MO-EX.
(Mg/yr) (Mg/yr) (Mg/yr) (Mg/yr)
121,448 86,312 31,223
175 120 43 0
352 244 88 391
363 256 93 1,148
362 258 93 1,518
356 253 91 1,533
348 247 89 1,498
343 244 88 1,478
339 241 87 1,460
330 234 85 1,419
325 231 84 1,399
321 228 82 1,381
316 225 81 1,360
311 221 80 1,340
307 218 79 1,320
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302 '
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 78 1,302
302 215 . 78 1,302
10,896 7,730 2,796 44,573
3,366 2,380 861 12.333
E-22
-------
Bi-fim,
Bz.
NESHfiP
EX.
(Mg/yr)
8
264
769
1,033
1,852
1,823
1,814
1,002
974
953
348
934
920
985
894
894
894
894
394
894
894
894
894
894
894
894
894
894
894
894
894
894
894
894
894
38,568
8,444
BI-flNN.
Bz.
flLL N8 flREflS
NO-EX.
(Hg/yr)
257
516
532
531
521
589
583
4%
483
476
469
463
456
449
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
15,955
4,929
BI-flW.
Bz.
fill Nfl flREflS
EX.
(Mg/yr)
176
357
375
377
370
362
357
353
343
338
334
329
324
319
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
11,320
3,485
64
129
136
136,
134
131
129
128
124
122
121
119
117
115
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
114
4,095
1,261
431
1257
1674
1690
1652
1629
1689
1565
1542
1522
1500
1478
1455
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
1435
49144
13598
0
291
348
1139
1160
1134
1118
1105
1074
1059
1045
1039
1014
999
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
985
33704
9310
E-23
-------
imxL
Bz.
ALLNRflRQS
HQ-EX.
(Mg/yr)
276
553
571
578
559
547
539
533
518
518
584
495
489
482
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
fi}«UflL
Bz.'
flLL Nft flREAS
EX.
(Mg/yr)
189
383
482
405 '
397
389
383
379
368
363
358
353
348
342
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
mm.
Bz.
SELECTED Nfi RREflS
EX.
(Hg/yr)
68
138
145
146
144
141
139
137
133
131
138
128
126
124
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
QUftRTERLY
Bz.
NESKflP
NO-EX.
(Mg/yr)
0
451
1315
' 1752
1768
1729
1785
1684
1638
1614
1593
1570
1546
1523
1.502
1502
H502
1582
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502 .
QUfiRTERLY
Bz.
NESHflP
EX.
(Hg/yr)
9
304
887
1192 '
1214
1187
1170
1156
1124
1108
1094
1077
1061
1045
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
8USRTERLY
Bz.
flLL Nfl PRESS
NO-EX.
(Mg/yr)
288
579
596
595
585
572
564
557
541
534
527
519
511
504
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
QUfiRTERLY
Bz.
fill Nfl flREflS
EX.
(Mg/yr)
197
400
420
• 423
416
406
401
' 396
385
379
374
369
363
358
353
353
353
353
353
353
353
353
353
353
353
353
. 353
353
353
353
353
353
353
353
353
17123
5290
12148
3740
4394
1353
51430
14230
35271
9743
17931
5530
12700
3910
E-24
-------
QUflRTERLY
Bz.
SEL Nfl flREflS
EX.
(Mg/yr)
71
145
IK
153
150
147
145
243
139
137
135
133
131
129
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
128
4594
1414
E-25
-------
-------
APPENDIX F
EXPOSURE AND HEALTH RISK
ANALYSIS
F-l
-------
-------
Table F-l. CONVERSION FACTOR CALCULATIONS
Pollutant
Gasoline Vapors
- Plausible Upper Limit
rat studies (kidney)
mice studies (liver)
Seventy year unit risk Molecular
factor (per ppro)
3.53 x ID"3
2.14 x 10-3
Ratio to YOC emissions
From storage tanks
1
From refueling
1
- Maximum Likelihood Estimate
rat studies (kidney) 2.01 x 10-3
mice studies (liver) 1.44 x 10-3
2.2 x ID'2
4.2 x 10-1
2.8 x 10-2
Benzene
EDB
EDO
78.11
187.87
98.97
0.0060
0.000046
0.00047
0.0066
0.000052
0.00053
For Sulk Plants, Bulk Terminals and Service Station Inloading
Risk from gasoline vapors
Risk from Benzene
= 3.53 x ig~3 gasoline vapors incidence/ppm gasoline vapor x
2.2 x 10-2 benzene incidence/ ppm benzene
x 78.11 g/g-mole Bz x 1 g gasoline vapor/gYOC = 31.6 (Plausible Upper Limit, rat)
66 g/g-mole gasoline 0.0060 g benzene/gVOC
vapor
= 2.14 x 10-3 x (78.11) x (1)
272X 10-2 (66) X (0.0060)
19.2 (Plausible Upper Limit, mice)
• 2.01 x 10-3 x (78.11) x (1) = 18.0 (Maximum Liklihood Estimate, rat)
2.2 X 1Q-2 x (66) X (0.0060)
= 1.44 x 10-7 x (78.111 x (1)
2.2 x 10-2 (66) X (0.0060)
Risk from EDC
Risk from EDB
Risk from EDB
Risk from benzene
Risk from EDC
= 12.9 (Maximum Likelihood Est
= 1.29
= 4.2 x 10-1 x (78.11) x (0.000046) = 6.09 x 10-2
= 2.8 x 10-2 x (187.87) x (0.00047)
4.2 X 10-1 x (98.97) X (0.000046)
Z7Z X 10-^ X (187.87) X (0.0060)
= 2.8 x 10-2 x (78.11) x (0-00047)
Risk from benzene" 2.2 x 10-2 x (98.97) x (0.0060)
= 7.87 x 10-2
For Service Station Refueling (Outloading)
Risk from gasoline vapors
Risk from benzene
Risk from EDC
Risk from EDB
Risk from EDB
Risk from Benzene
Risk from EDC
Risk from benzene
= 3.53 x 10-3 x (78.11) x (1)
2.2 X 10-2 x (66) X (0.0066)
= 2.14 x 10-3 x (78.11) x (1)
2.2 X ID'2 X (66) X (0.0066)
= 2.01 x 10-3 x (78.11) x (1)
2.2 x 10-2 x (66) x (0.0066)
= 2.gi x IP"3 x (78.11) x (1)
2.2 X 10-2 x (66) X (0.0066)
= 2.8 x 10-2 x (187.87) x (0.00053) =
4.2 x 10-1 x (98.97)X (0.000052)
= 4.2 x 10-1 x (78.11) x (0.0000521 =
2.2 x 10-^ x (187.87) x (0.0066)
= 2.8 x 10-2 x (78.il) x (g.gggss)
'i.'i X 10-2 x (98.9/i X (0.0066)
28.8 (Plausible Upper Limit, rat)
17.4 (Plausible Upper Limit, nice)
16.4 (Maximum Likelihood Estimate, rat)
11.7 (Maximum Likelihood Estimate, mice)
1.29
6.25 x 10-2
8.07 x 10-2
F-3
-------
Table F-2. Bulk Terminal Incidence - Theoretical
Ytar
1985
1937
1238
1583
1538
1391
1S9£
1993
1934
1S95
1395
1997
1998
1999
2001
£§02
£603
2804
2S05
SMI
3)38
2011
i«i£
£013
£614
£015
£.315
£017
£918
£319
£923
Baseline
Bulk
Terainal
Incidence
Due To
Benzene
8.3883
0.0788
0.3773
0.0753
3.3749
3.0738
0.3731
0.0725
0.0712
0.07%
0.3699
0.0693
0.8686
0.0680
0.3574
0.0674
0.3674
0.0674
0.2674
0.0674
8.0674
0.0674
0.8674
0.0674
0.0574
0.0674
0.2574
0.2574
•3.3674
0.0674
3.8674
0.C674
3.3674
0.0674
8.2674
Controlled
Bulk
Terminal
Incidence
Due To
Benzene
0.0803
0.0723
0.0582
0.3508
0.0501
0.0493
•3.0483
0.0484
0.0474
0.0459
0.0465
0.0460
3.0455
0.0451
0.0446
0.0446
0.0446
. 0.0446
0.0445
0.0445
0.0445
0.0446
0.0446
0.0446
0.3445
0.3445
3.0445
0.3445
3.0445
, 0.9445
0.0445
0.0446
0.0445
0.0446
3.0445
Baseline
Bulk
Terainal
Incidence
Due To
Sas Vauors
(PUL,rat)
2.540
2.493
2.447
2.408
2.359
2.335
2.315
2.294
2.254
2.233
2.213
2.193
2.172
2.152
2.132
2.132
£.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
.2.132.
2.132
2.132
2. 132
2.132
2.132
Controlled
Bulk
Terminal
Incidence
Due To
Sas Vaoors
(PlL,rat)
£.540
2.289
1.843
1.608
' 1.585
1.560
1.545
1.530
1.501
1.486
1.471
1.456
1.441
1.426
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
. 1.412
1.412
1.412
1.412
1.412
1.412
Baseline
Bulk
Terminal
Incidence
Due To
Sas Vapors
(PULaice)
1.540
1.512
1.483
1.455
1.436
1.415
1.403
1.391
1.366
1.354
1.342
1.329
1.317
1.305
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
. 1.292
1.292
1.292
1.292
1.292
1.292
Controlled
Bulk
Terminal
Incidence
Due To
Sas Vauors
(PULBiice)
1.540
1.388
1.117
3.975
0.961
3.946
3.937
0.928
0.913
0.901
0.892
0.883
0.874
3.855
3.856
8.856
0.856
3.856
3.356
3.855
3.856
0.855
3.855
3.856
3.856
0.855
0.855
3.855
3.856
0.855
3.856
9.855
3.856
3.856
0.855
Baseline
Bulk
Terminal
Incidence
Due To
Gas Vapors
(MLLrat)
1.446
1.420
1.393
1.357
1.349
1.338
1.318
1.305
1.283
1.272
1.263
1.249
1.237
1.225
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
. 1.214
1.214
1.214
1.214
1.214
1.214
2.44
0.773
1.67
3.565
77.18
24.45
52.92
17.88
45.79
14.33
32.38
13.84
43.95
13.93
F-4
-------
Table F-2. Bulk Terainal Incidence - Theoretical
fear
1986
1387
1388
1389
1390
1331
1992
1933
1394
1935
19%
1937
1393
1339
2330
£881
£002
2883
£384
2005
2306
2007
2888
2909
2818
2311
2012
2013
2014
2015
2916
2017
2018
2319
2320
SUM =
NPV =
Controlled
Bulk
Terminal
Incidence
Due To
Gas Vaoors
OLE, rat)
1.446
1.333
1.349
0.915
3,903
0.388
• 8.389
8.871
3.354
8.846
3.338
3.329
8.821
0.812
8.384
3.884
9.884
3.384
8.384
0.304
3.884
8.834
8.384
8. 304
0.804
0.804
•3.884
8.834
3.384,
3. 384
3.884
0.884
0.884
3.304
8.804
38.13
10.18
Baseline
Bulk
Tersinal
Incidence
Due To
Gas Vapors
(ME, Mice)
1.336
1..017
8.338
3.379
8.367
0.352
3.344
3.336
3.919
0.311
8.983
3.894
8.886
8.878
8.878
8.878
8.878
8.873
8.878
3.873
0.878
8.878
8.878
8.378
8.878
3.873
3.870
0.870
3.878
3.878
8.870
3.870
0.870
3.878
8.873
31.43
9.38
Control led '
Bulk
Terainal
Incidence
Due To
Gas Vapors
(ME, Bice)
1.835
8.334
8.752
3.655
8.647
8.636
0.638
8.624
8.512
3.636
8.688
8.534
8.588
8.582
0.576
8.576
8.576
3.576
8.576
8.576
8.576
8.576
8.576
3.576
3.576
3.576
8.575
0.575
8.576
8.576
8.576
8.576
8.575
8.576
8.575
21.53
7.23
Baseline
Bulk
Terminal
Incidence
Due To
EDB
8.88144
8.88131
8.88121
8.88183
3.30094
3.83385
8.38872
8.88364
3.88858
8.30344
8.88841
3.80837
3.83833
8.83329
8. 08825
8.08825
3.00825
8.03825
8.63325
3.38825
8.38825
0.88325
8.08825
3.83825
3.88025
8.08825
8.08825
8.38825
8.88325
3.80825
8.08825
8.88325
0.83825
8.83825
8.S8825
3.8157
8.8878
Control led
Bulk
Terminal
Incidence
Due To
EDB
3.88144
8.88128
3.38898
8.38063
8.33862
3.08856
3.88847
3.88342
3.33833
8.38823
8.88826
3.08823
3.83823
8.38817
8.88814
8.38814
3.88814
3.88814
8.08814
3.88314
3.88814
8.38814
8.33814
0.38814
8.68814
8.33314
3.83814
3.33314
3.38814
8.88814
8.38814
8.08814
8.83814
0.38814
3.38814
8.8188
3.08597
Baseline
Bulk
Terrainai
Incidence
Due To
EDC
3.38186
0.88178
3.38157
8.88133
8.88122
3.88189
8.38833
8.83883
8. 83865
0.38857
3.38852
8.38347
3.38842
8.88937
8.38832
8.88832
8.38832
3.88832
0.33832
3.83332
8.88832
8.33832
3.88832
8.83832
8.08332
8.83832
8.08332
3.33332
8.88832
0.83332
3.S3832
3.88832
8.38832
3.00032
3.88832 .
0.8283
0.8101
Controlled
Bulk
Terminal
Incidence
Due To
EDC
3.38186
3.88156
8.03117
3.88389
8.86383
0.88872
8.33361
8.83854
3.88043
8.08837
8.38833
8.83838
3.88826
3.88822
8.33819
8.33819
0.33019
3.08819
8.88819
0.88813
0.88819
8.88819
8.88319
3.83819
3.88819
0.38819
8.80813
3.38319
8.33319
3.83319
0.88819
8.88819
3.S8819
8.33813
3.33319
8.8139
8.0877
F-5
-------
Table F-3. Bulk Plant Incidence - Theoretical
ar
1986
1937
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
sees
£00i
2002
m$
sew
2B05
sm
£097
20ea
£899
2813
1311
2012
£913
2814
£015
2016
£617
2918
£019
2028
Baseline
Bulk Plant
Incidence
Due To
Benzene
3. 8439
0.0427
0.0415
0.0484
0.93%
0.0387
0. 8382
3.0377
0.0367
0.0362
0.0357
3.0352
8.0346
0.8341
0.8336
8.0335
0.0336
0.8336
0.8336
8.8336
0.0336
0.0336
0.8336
0.8336
0.8336
3.3335
8.0336
8.8336
0.8336
8.3336
0.0336
0.0336
8.3336
St. 8335
0.8336
Controlled
(EX)
Bulk Plant
Incidence
Due To
Benzene
0.0439
0.8365
0.0238
0.0174
8.0178
8.8167
0.0164
0.8162
0.0158
0.0156
0.8153
0.0151
0.0149
0.0147
0.0145
0.8145
8.0145
0.0145
0.8145
0.0145
0.0145
0.0145
0.0145
0.0145
0.0145
0.0145
8.8145
3.0145
0.8145
0.8145
0.8145
0.8145
0.0145
8.8145
0.3145
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Benzene
8.3439
8.8358
8.8215
0.0144
0.0141
8.8138
8.8136
0.0134
0.0131
0.8129
0.0127
8.8125
0.0124
0.0122
0.8120
0.0120
0.0120
8.0120
0.0128
0.0120
0.8120
0.0120
0.0120
0.8120
8.8128
8.8128
8.8128
8.8120
8.8128
8.0128
0.8120
0.0128
3.8120
0.8120
0.8128
Baseline
Bulk Plant
Incidence
Due To
Bas Vapors
(Pll,rat)
1.389
1.351
1.314
1.277
1.253
1.226
1.218
1.193
1.161
1.145
1.129
1.113
1.097
1.881
1.864
1.064
1.864
1.064
1.864
1.864
1.064
i.064
1.864
1.964
1.864
1.064
1.864
1.864
1.864
1.864
1.864
1.854
1.064
1.854
1.864
Controlled
(EX)
Bulk Plant
Incidence
Due To
Gas Vapors
(PUL,rat)
1.389
1.159
8.752
8.549
0.539
8.527
8.520
0.513
0.499
0.492
8.485
0.479
3.472
8.465
0.458
8.458
8.458
0.458
8.458
0.458
0.458
0.458
0.458
8.453
0.458
8.458
0.458
0.458
8.458
3.458
3.458
8.458
0.458
8.458
0.458
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Gas Vapors
(PUL,rat)
1.389
1.134
8.6B8
8.455
8.447
0.437
8.431
8.426
0.414
8.488
8.403
8.397
0.391
8.385
8.388
0.388
0.380
8.388
0.388
8.388
8.338
8.380
8.380
0.388
8.388
8.388
8.380
0.380
8.380
. 8.388
8.388
8.388
8.388
8.388
0.388
Baseline
Bulk Plant
Incidence
Due To
Gas Vapors
(Pit, mice)
8.842
8.319
0.797
8.774
0.760
8.743
0.733
8.723
0.704
0.694
8.684
0.675
8.665
8.655
8.645
8.645
8.645
8.645
0.645
3.545
0.645
8.645
8.645
8.645
0.645
8.645
8.645
3.645
0.645
8.645
0.645
8.645
3.645
8.545
0.645
1.24
0.403
0.583
8.220
0.498
8.196
39.29
12.75
18.45
S.96
15.77
6.21
£3.82
7.73
F-6
-------
Table F-3. Bulk Plant Incidence - Theoretical
Year
1985
1987
i'388
1989
1958
1991
1992
1993
1994
1995
19%
1997
1998
1999
£088
20(91
£00£
2003
£884
£835
2036
£007
£088
£339
£010
£911
2012
2813
£014
£015
£216
£017
£018
£019
20£0
SUM =
NPV =
Controlled
(EX)
Bulk Plant
Incidence
Due To
Gas Vaoors
(PUL, rnics)
3.842
0.703
0.455
0.333
3.327
0.328
0.315
3.311
0.383
3.298
0.294
0.290
0.286
0.282
0.277
3.277
9.277
0.377
8.277
3.277
8.277
3.277
9.277
8.277
9.277
3.277
8.277
0.277
9.277
3.277
0.277
3.277
0.277
0.277
0.277
11.19
4.22
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Sas Vaoors
(PUL,fflice)
0.842
0.686
0.412
0.276
3.271
0.265
0.261
0.258
0.251
0.248
0.244
0.241
8.237
0.234
0.230
0.230
d.230
0.230
0.233
0.230
0.220
0.230
0.230
8.238
0.230
0.230
0.238
0.230
0.230
0.238
0.230
0.230
0.233
0.230
0.£33
9.56
3.76
Baseline
Bulk Plant
Incidence
Due To
Sas Vapors
(BLE.rat)
3.791
0.778
0.748
8.727
0.713
0.698
0.689
0.688
0.661
0.652
8.643
0.634
0.624
0.615
0.686
8.606
„. ,0.685
0.606
8.686
0.606
0.606
0.686
0.686
0.606
0.636
0.686
0.606
0.606
0.606
8.686
0.606
0.686
8.606
0.686
0.606
22.37
7. £6
Controlled
(EX)
Bulk Plant
Incidence
Due To
Sas Vapors
-------
Table F-3. Bulk Plant Incidence - Theoretical
ar
1986
1S87
1958
1989
1930
1991
1932
1993
1934
1995
19%
1997
1958
1999
2001
2022
2e03
2C34
2885
2807
2828
2639
2913
2011
2012
2013
2014
2015
2016
2017
2818
2919
202®
8UN-
»V«
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Gas Vaoors
(ME, aice)
0.566
0.463
0.277
9.186
0.182
3.178
0.176
0.174
0.169
0.157
0.164
0.162
0.160
0.157
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
6.43
2.53
Baseline
Bulk Plant
Incidence
Due To
EDB
0.S00784
0.000685
0.000605
0.000527
0.000456
0.000403
0.000241
0.000279
0.000233
0.000186
0.000155
0.000124
0.000893
0.000062
0.000031
0.000031
0.000031
0.000031
0.000031
0.000031
0.000031
3.000031
0.000031
0.000031
0.600031
0.000031
0.000031
0.000031
0.008031
0.300031
0.800031
3.000031
0.000031
0.000031
0.000031
0.00558
0.00362
Controlled
(EX)
Bulk Plant
Incidence
Due To
EDB
0. (500784
0.000588
0.000346
0.800227
0.000196
0.800173
0.300147
0.000120
0.000130
0.030080
0.000867
0.000053
0.003040
0.000027
0.000013
0.000013
0.000013
0.000013
0.800013
0.300013
0.000013
0.000013
0.006013
0.030013
0.800013
0.i30013
0.000013
0.000013
0.000013
0.000013
0.003813
0.030013
0.000013
0.238313
0.000013
0.80323
3.00234
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
EDB
3.300784
0.003575
0.003313
0.000188
0.'800163
0.000144
8.300122
0.000100
' 0.000083
0.303066
0.S00055
0. '380044
0.000033
0.003022
3.303011
0.030011
0.000011
0.000011
3.033011
0.380011
0.008811
0.030011
0.000011
0.000011
0.300011
8.800311
9.323011
0.303011
0. 000611
0.333011
8.000011
3.000011
3.003911
0.000011
3.390011
0.89292
3.550218
Baseline
Bulk Plant
Incidence
Due To
EDC
0.001314
0.000886
0.803782
3.300682
3.300583
0.300521
3.003441
0.000361
0.000301
0.000241
0.303200
0.030160
0.303120
0.030080
0.000040
0.030040
0.000040
0.003040
0.303040
0.030040
3.800340
0.000040
3.303840
0.300040
9. 338940
0.000040
0.333040
3.030040
0.308040
0.033040
0.308040
3.330040
3.003040
3.030040
3.303040
8.00722
0.30468
Controlled
(EX)
Bulk Plant
Incidence'
Due To
EDC
3.301014
0.300760
0.000448
3.000293
0.000253
0.003224
0.380190
0.300155
0.390129
0.000103
0.300086
3.000069
0.000052
0.000034
0.300017
3.300017
0.330017
0.303017
0. 000017
0.300017
0.038017
0.000017
0.000017
3.300017
3.030017
3.300017
0.303017
3.300017
0.300017
0.000317
' 0.033017
3.303017
3.300017
3.300017
3.303017
0.00417
0.30383
Controlled
(NO. EX)
Bulk Plant
Incidence
Due To
EDC
3.001014
0.300744
0.300405
0.033243
3.300210
3.300186
3.300157
3.000129
0.300107
' 0.1300086
0.000071
0.000057
0.000043
0.300029
3.000014
0.300014
3.303014
3.030014
3.303014
0.333014
3.303014
0.303014
0.303014
3. 303014
0.308014
3.300014
3.030314
3.303014
3.030314
3.333014
' 3.800014
0.300014
8. 338814
8.330014
8. 803014
3.30378
3.33281
F-8
-------
Table F-4. Service Station Incidence Due To Benzene - Theoretical
year
1985
1987
1388
1939
1993
1591
1992
1993
1994
1995
1996
1997
1998
1999
ma®
2881
2882
2383
£8**
£005
SM&
2887
£3ea
•2239
£312
£8ii
23:2
£013
£814
2315
i316
2017
£818
£819
£828
SUN =
•iPV =
Baseline
Total
Incidence Due
To Benzene
9. ££91
3. £233
3.2159
3.2138
3.2368
S.£322
3.19%
.3. 1959
3.1916
«. 1889
3.1863
3.1835
3. 1818
3. 1783
3.1756
3.1756
3.1756
3. 1756
3. 1756
3. 1756
3.1756
3.1756
8. 1755
3.1756
8. 1755
3. 1755
8. 1756
3.1756
3. 1755
3.1756
3. 1756
3. 1756
3. 1755
3. 1756
3. 1755
6.43
£.13
St.II-Nfl*
(EX)
Total
Incidence Due
To Benzene
3. £117
3.1877
3. 1799
3. 1734
3. 1782
3. 1564
•3. 1642
3. 1621
3.1577
3.1555
3. 1533
3.1511
8. 1489
3. 1467
8. 1445
3. 1445
3. 1445
3.1445
•3. 1445
3. 1445
3. 1445
3.1445
0. 1445
3. 1445
3. 1445
3. 14*5
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
•2. 1445
5.35
1.76
St. II-Nfl*
(NO EX) .
Total
Incidence Due
To Benzene
3. 1938
•3. 1623
3. 1543
3.1483
3.1455
3.1423
3. 1434
8.1385
•3. 1348
3. 1329
3. 1311
0.1292
8. 1273
0.1255
8.1236
3. 1235
0. 1236
3. 1236
8. 1236
3. 1235
3. 1236
3. 1236
3. 1236
3. 1235
8. 1235
3. 1235
3. 1235
3. 1235
3. 1236
3. 1235
3. 1235
3. 1236
3. 1235
3.1235
8. 1235
4. Si
1.53
St.II-NA
(EX)
Total
Incidence Due
To Benzene
3. 1933
8. 1584
3, 1436
3.1339
8. 1313
8.1285
8. 1258
3. 1251
3. 1217
3. 1288
•3. 1183
8.1156
8. 1149
8.1133
8. 1115
8.1116
3.1116
3. iliS
•2.1116
8.1115
3.1 US
8.1116
3. 1116
8.1116
3.1115
3.1116
3.1115
8. 1115
3.1116
8.1115
3.1116
3.1116
'2.1115
3.1115
3.1115
4.18
1.48
.St. II-Nfl
!NO EX)
Total
Incidence Due
To Senzene
3. 1712
8. 1378
3. -3974
3.8914
3.3897
8.3377
3. 3365
8.3854
8.^331
3.3819
•3. 3888
3.87%
3. 2785
8.8773
3.3762
3.8762
§. 8762
8.8762
3.3752
3.8762
,8.3762
3.8762
3.8762
3.3762
3.3762
8.3752
3.3752
3.8762
3.0752
8.3752
3.8762
8.3762
3.3762
8.3752
3.3752
2.98
1.08
St.I-Naticn
(EX)
Total
Incidence Due
To Benzene
3.2291
3.2174
3.2084
3. 1894
•3. 1858
3. 1817
3. 1794
3. 1778
3.1722
3. 1598
3. 1574
3. 1650
3.1625
3. 1682
•3, 1578
3. 1578
8. 1578
8. 1578
•3. 1578
3.1578
3. 1578
8. 1578
8.1578
3. 1578
3. 1578
8. 1578
•3. 1573
8. 1578
3. 1578
3. 1578
3. 1578
3.1573
3. 1578
3. 1578
3. 1578
5.87 '
1.93
St. I-Nation
(NO EX)
Total
Incidence Due
To Benzene
3.2291
3.2136
3. 1895
3. 1754
3.1723
3. 1683
3. 1651
3. 1639
8. i594
8. 1572
3. 1558
3. 1528
3. 1586
3.1484
3. 1461
8.1451
3. 1451
8. 1461
8. 1461
8.1461
3. 1461
8. 1461
3. 1451
3. 1461
8. 1461
3. 1461
8. 1451
3.1461
8. 1461
3. 1461
3. 1461
8.1461
3.1461
3. 1451
8. 1461
5.47
1.82
F-9
-------
Table F-4. Service Station Incidence Due To Benzene - Theoretical
Year
St.II-Nation
'.EX)
St.II-N'ation
API (EX)
St.II-Xation St. I ill-Nat ion St.Iill-Nation Onboard-Nation
(NO EX) (EX) (NO EX)
St.II-Nfl*
(EX)
& Onboard
•"oral Total Total Total Total Total Total
Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due
"o Benzene To Benzene To Benzene To Benzene To Benzene To Benzene To Benzene
1936
1987
1988
1989
1990
1991
1952
1993
1394
19S5
19%
1997
1938
1999
283d
ae0i
23K
c«03
•a»
2385
2835
cw?
zm
£289
•914
a3h
e'312
£313
1-314
£$15
2916
2017
318
S919
£?££
£12 -
vpy =
3.2291
9. 1991
0.1473
0. 1172
9.1116
3. 1892
8. 1377
0.1@63
8. 1834
0.1820
0. 1005
0.0991
0.3977
0.0962
0.0948
8.8948
8.3948
0.0948
0.0948
0.8948
0.3948
8.8948
8. 8948
0.0948
0.8948
0.0948
3.0948
3.3948
3.3948
0.0948
8.8948
0.0948
8.3S43
3.3948
0.2948
3.72
1.34
0.2291
0.2210
0. 1959
3.1555
3. 1259
0. 1134
3.1141
3.1381
3.1034
0. 1020
0. 1005
0.0991
0.0977
0.3962
0.0948
0.0948
0.8948
0.0948
0.0948
0.0948
0.0948
0.8948
3.0948
• 0.0948
3.3948
8.0948
'3.3948
8.0948
3.8948
0.0948
3.8948
0.8948
•3.3943
0.0S48
0.3948
3. 85
1.45
0.2291
0.1S63
3. 1102
3.2685
3.S633
0.0619
3.3611
0.0603
3.0537 '
0.3579
0.0570
0.0562
0.0554
0.8546
0.0538
0.0538
•3.3538
S.0538
•3.3538
3.8538
3.3538
0.0538
0.8538
•3.3538
0.0538
8.8538
3.8538
3.2538
•3.8538
0.0538
3.3538
3.2538
0.8338
8.8538
0.8538
2.31
3.35
8.2291
3. 1334
3.1308
8.3959
3.8986
8.0887
0.5575
8.8863
3.8843
3.0828
0.8817
8.8885
3.8793
3.8782
8.8770
0.8778
0.8770
3.0778
8.3773
•3.9778
8.1773
3.0773
0.3773
0.3778
•3.8773
8.3773
8.8773
3.3773
3.8773
0.3778
8.3770
8.3778
3.8778
8.3778
8.3773
3..11
1.17
8.2291
3. 1785
3.13374
0.3391
3.8347
8.3339
3.3334
8.8333
3.3321
3.8317
3.3312
3.8388
3.8333
8.8299
3.8294
8.8294
8.3294
8.8294
•3.8294
0.8294
8.8294
3.8294
3.3234
3.8294
3.3294
3.3294
3.8234
3.8294
3.8294
3.0294
•3.3294
8.8294
8.0294
8.8294
3.8294
1.47
3.72
2.2291
8.2233
3.2014
3. 1815
3.1655
3. 1588
•3. 1377
3.1266
3. 1139
3. 1849 "
3.8969
3.38%
3.3831
3.8772
3.0721
8.3721
3.3721
3.3721
3.3721
3.8721
8. 3721
3.3721
3.3721
3.3721
3.3721
8.3721
3.3721
3.8721
3.3721
0.8721
8.3721
3.8721
8.8721
3.0721
8.8721
3.49
1.48
3.2117
3. 1877
<3. 1583
3. 1515
3. 1393
3. 1273
3. 1179
8. 1894
8.8995
0.3926
8.8864
3.8887
3. 3757
3.8711
3.8670
0.2691
3.3714
3.8719
3.8721
3.3721
8.8721
3.8721
3.3721
3.8721
3.8721
3.8721
3.3721
3.3721
3.3721
3.3721
8.3721
8.8721
. 3.3721
3.8721
3.3721
3.22
1.33
F-10
-------
Table F-4. Service Station Incidence DUB To Benzers -Theoretical
'aar
1586
1587
1938
1589
1998
1931
1952
1533
1994
1395
1996
1557
1998
1999
3580
£381
2822
£303
£084
£635
3826
£237
£328
£039
2018
£811
£212
£813
£814
£315
£316
£817
£018
£219
£028
m =
*V =
St.lI-Nfl*
(NO EX)
i Cr.-board
"otai
Inci de?scs Due
To Benzene
3. 1988
8. 1623
3.1448
3.1332
2.1199
8.1299
•2. 1321
8.0953
3. -2367
8.363-3
3.0757
3.3739
3.2667
8.3623
8. 3594
8.0646
3. 3737
8.8717
0.3721
3.3721
3.3721
0.8721
3.3721
3. S721
•2.3721
3.5721
3, 3721
3.3721
8. 07£i
3.3721
8. 3721
3.3721
8.3721
0..3721
3.0721
3.83
-..15
3t.II-.Nfl
(EX)
4 Onboard
Total
Incidence Due
To Benzene
8.1933
8. 1584
0. 1327
3.1189
0.1181
8.1817
•2.0958
3.8898
8.8318
8.3769
8.3724
8.8633
8.8547
8.8614
8.0584
8.8640
a. 8703
8.8715
8. 3722
8.3721
2.8721
8.8721
8.8721
8.8721
8.8721
3.8721
8.8721
8.8721
8.0721
8.8721
6.8721
3.8721
8.3721
3.8721
«. 3721
2.51
1.10
St.II-Nfl
(NO EX)
4 Onboard
Total
Incidence Due
To Benzene
" 0. 1712
3.1870
8.3929
8.8829
8. -2775
8.8725
0.0685
8.8649
8.8684
8.8574
0.8547
0.8522
0.2508
3.0479
8,0460
3.3570
8.3S93
0.8713
8.0720
3.8721
3.3721
8.8721
• 3.3721
0.0721
3.3721
8.3721
3.3721
3.3721
0.8721
8.8721
3.3721
3.3721
3. 3721
8.0721
3. 3721
2.53
3.37
St. I (EX)
& Onboard-
Nation
Total
Incidence Due
To Benzene
3.2291
3.2174
3. 1849
8. 1681
0. 1445
8.1255
8.1174
8.1866
8.0945
0.8858
3.3780
,8.0710
0.0648
8.3592
0.0543
3.3543
3. 3543
8.3543
0.0543
3.3543
0.8543
0.3543
3.8543
3.3543
3. 8543
3.3543
3.2543
3. 3543
3. 3543
3. 3543
8.3543
3. 3543
3.3543
0.3543
3. 0543
£.83
1.31
St. I (NO EX)
J Onboard-
Nation
Total
Incidence Due
To Benzene
8.2291
8.2136
8. 1741
8. 1461
0.1388
3.1151
3.1041
8.8935
3.8817
0.3732
8.3655
8.3587
3.8527
0.8473
8.8426
0.0426
3. 3426
8.0426
8.0425
8.8425
3.3426
0.3426
3.3425
3.3426
8.8425
3.0426
3.8426
3.3426
3.8425
3.0426
3.8425
3. 3425
•3. 34£5
8.3425
3.8425
£.48
i.£9
& St. I (EX) i St.I (NO EX)
it Gnbd-Nation & Onbd-Nation
;al
rice Due
izene
•3. 1933
3.1464
3. 1218
8. 1037
3.0953
8.8871
8.2807
3.0748
3.0631
0.8533
3.3591
0.0552
3.0517
0.3486
8.0453
8.0493
3. J532
8.8539
3.3543
3. 3543
8.8543
3.3543
0.0543
3. 3543
3.3543
3.3543
3. 3543
3. 3543
3.8543
3.3543
-3.3543
3.8543
3. 3543
3.3543
3. 0543
Total
Incidence Due
To Benzene
. 0. 1712
8.1828
3.3782
3.8639
3.8590
3.8543
0.8505
3.3472
3.8432
3. 3484
3.0388
3.8357
3.8337
3.8318
3.3332
8.0354
8. 8413
3.8422
3.8425
3.8425
3.8425
3.0425
3. 0425
3.3425
3.3425
3. 3426
3.8426
3. 34£6
8. 3425
3. 0426
3.3426
3. 34£b
3.3425
3.8425
3.3425
£.37
3.97
1.72
8.69
F-ll
-------
"able F-4. Service Station
Due To Benzene - Theoretical
ar
1336
13S7
.338
1*83
.353
1=31
*:3£
13S3
1H&
1595
13%
1997
.393
1S99
22TO
£231
£232
2233
2234
£225
£?5?5
2237
£w38
£.235
I!lO
2S11
•II 12
£?13
£2114
£315
£316
2217
2! 13
£215
£323
5t.II JEX)
i Qnboard-
\a*ion
Total
incidence Due
~o Senzane
£ flJW
3. .991
3. 1422
3. 1375
2. -2979
S.391S
2. 2372
3.0829
0.3775
0.8741
3.3739
8.8679
8.3652
8.0627
3.3634
8.3684
3.3538
0.8683
3.3715
3.8721
3.3721
0.3721
3.3721
3.3721
•3.9721
0.3721
3.3721
3.2721
•3.8721
3.3721
3.3721
3.3721
3.8721
8.8721
3.3721
St. II (NO EX)
i Ondoarc-
Nasion
Total
Incidence Due
~; Banzsne
3.2291
8. 1363
3. 1397
3.8674
3.0618
3.3681
8.8589
8.8578
3.0559
8.8549
3.8539
0.3529
3.0519
0.3510
3.0501
0.8501
3.8550
3.0651
3.8712
0.3721
8.8721
0.8721
3.3721
0.0721
•3.8721
3.3721
3.0721
3.3721
3.3721
3.8721
3.0721
3.3721
3.8721
3.3721
•3.8721
St. I 411 (EX)
& Onboard-
Mat ion
"otai
Incidence Due
To Senzene
. 3.2291
3. 1934
3.1257
8.3862
8.8769
0.8713
3.8578
8.0638
8.8582
3.8558
0.8520
•3,8493
3.8469
8.0446
0.8425
8.0426
0.0452
0.0505
3.8537
8.0542
3.3543
8.8543
3.8543
8.0543
8.8543
8.8543
3.8543
3.8543
3.8543
8.3543
3.3543
8.3543
3.8543
8.8543
3.8543
St.Iill {MO EX)
i Onboard-
Nation
Total
Incidence Due
To Benzene
3.2291
3.1785
8.8868
8.3381
3.3332
3.0328
0.8312
0.3385
0.8293
•3. 3287
8.8280
8. 8274
8.8268
0.8263.
8.0257
8.8257
8.8295
0.8373
3.8419
8.8426
8.8426
3.8425
8.8426
3.8426
3.8425
3.3426
8.8425
0.8426
3.8425
3.8426
3.8426
3.8426
•3.3425
3.3425
•3.8426
Slil =
2.93
2.53
8.97
2.32
•3.98
1.67
3.72
F-12
-------
Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Upper Liait) - Theoretical
iar
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2001
2882
2083
2084
2085
2036
2007
2088
2089
2010
2011
2012
2313
2014
2015
2016
2017
2018
2019
2020
m =
>v =
Basel ire
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.759
S.588
6.488
6.227
6.189
5.975
' 5.897
5.818
5.661
5.582
5.583
5.425
5.346
5.268
5.189
5.189
5.189
5.189
5. 189
5.189
5.189
5.189
5.189
5.189
5.189
5.189
5.189
5.189
5. 189
5.189
5.189
5.189
5.189
5.189
5.189
191.54
62.19
St.II-Nfl*
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.265
5.567
5.335
5.145
5.348
4.937
4.872
4.807
4.678
4.613
4.548
4.483
4.418
4.353
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.283
4.288
4.288
4.288
159. 11
52.20
St.II-HW*
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.884
4.815
4.579
4.481
4.317
4.223
4.167
4.112
4.081
3.945
3.889
3.334
3.778
3.723
3.657
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.657
3.667
3.667
3.667
3.667
3.667
3.567
3.667
3.667
135.58
45.24
St.II-Nfl
(EX)
Total Incidence
Due To Gasoline
Vaoors(PUL,rat)
5.724
4.472
4.185
3.985
3.911
3.325
3.775
3.725
3.624
3.574
3.523
3.473
3.423
3.372
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
124.35
41.58
St.II-Nfl
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.874
3.192
2.386
2.738
2.578
2.619
2.585
2.552
2.481
2.447
2.413
2.373
2.344
2.309
2.275
2.275
2.S75
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2. 275
86.47
29.72
St.I-Nation
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.769
5.410
5.386
5.551
5.445
5.326
5.255
5.185
5.846
4.975
4.985
4.335
4.765
4.596
4.626
4.626
4.626
4.625
4.S26
4.626
4.625
4.625
4.525
4.525
4.626
4.526
4.5£6
4.625
4.6££
4.625
4.S2S
4.626
4.S2S
4.526
4.5£S
172. 19
56.38
F-13
-------
Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Upper Liait) - Theoretical
ar
198S
1987
1988
1989
1998
1991
1933
1993
1994
1995
19%
1997
1998
1999
2000
2801
2082
2083
2084
2083
28@6
2037
28€8
2089
2910
2811
2312
2013
2014
2015
2016
2017
2018
2019
2020
UN*
PV =
St.I-Nation
(fffl EX)
Total Incidence
Due To Gasoline
Vapors«PUL,rat)
6.769
6.232
5.543
5.106
5.310
4.900
4.336
4.771
4.642
4.578
4.513
4.449
4.384
4.320
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
159.48
53. 2S
St.II-Nation
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.769
5.897
• 4.395
3.522
3.357
3.283
3.240
3.197
3.110
3.067
3.024
2.981
2.938
2.894
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
111.55
40.09
St.II-Nation
ftPI (EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.769
6.529
5.801
4.628
3.769
3.579
3.425
3.249
3.110
3.1357
3.024
2.981
2.938
2.S94
2.851
2.851
2. 851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
115.64
43.25
St.II-Nation
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL. rat)
6.769
5.524
3.318
2.896
1.344
1.932
1.877
1.852
1.801
1.776
1.751
1.726
1.781
1.676
1.651
1.651
1.651
1.651
1.651
1.551
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.551
1.651
1.651
1.651
1.651
78.39
28.66
St. nil-Nation
(EX)
Total Incidence
Due To Gasoline
Vapor s (PIL, rat)
6.759
5.718
3.874
2.846
2.593
2.634
2.600
2.555
2.4%
2.461
2.425
2.392
2.357
2.322
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
92.28
34.70
St. Hill-Nation
(NC EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat!
S.763
S.27&
2.588
1.166
1.337
1.314
1.881
3.987
8.951
3.947
0.934
3.921
8.907
3.894
0.881
8.881
0.881
3.381
8.881
3.881
3.881
3.881
0.881
3.381
2.881
3.881
0.381
8.881
0.881
9.381
0.881
3.881
9. 881
a. 381
0.881
43.89
21.28
F-14
-------
Table F-5. Service Station Incidence Due To Gasoline Vanors (Plausible Uocer Liait) - Theoretical
jar
198&
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2080
2801
£082
2083
2084
200S
2006
2007
2008
2009
2010
2011
2012
2012
2014
2815
2315
2017
2018
£019
2020
Onboard-Nation
Total Incidence
Due To Baseline
Vapors (PUL. rat)
6.769
6.588
5.963
5.385
4.321
4.472
4.115
3.794
3.425
3.165
2.932
2.719
2.532
£.360
£.209
£.289
2.209
2.209
2.209
£.£09
2.209
2.209
2.289
2.209
2.209
2.209
2.289
2.209
2.209
2.289
£.£89
2.209
2.289
2.209
£.209
St.IHW*
(EX)
J Onboard
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.265
5.567
5.002
4.515
4.159
3.813
3.539
3.293
3.004
2.804
2.523
2.458
2.31£
2.177
£.358
2.120
2. 198
2.203
2.209
2.289
2.289
2.289
2.209
2.209
2.209
2.209
2.289
2.209
2.209
2.209
2.209
2.209
2.239
2.209
2.289
St.IHift*
(NO EX)
J Onboard
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.884
4.815
4.334
3.879
3.581
3.292
3.053
2.857
2.615.
2.447
2.296
2.157
2.1234
1.921
1.821
1.985
2.168
2.198
2.209
£.£09
2.209
2.209
2.209
2.209
2.239
2.209
2.239
2.209
2.289
2.209
2.289
2.209
2.209
2.209
2, £'39
St.II-Nfl
(EX)
1 Onboard
Total Incidence
Due To Sasoline
Vapors (PUL, rat)
5.724
4.472
3.357
3.554
3.301
3.054
2.860
2.585
2.475
2.333
2.203
2.384
1.978
1.880
1.792
1.962
2. 155
2.193
2.208
2.209
2. £39
2.209
2. £89
2.209
2. £39
2.209
2.209
2.209
£.£39
2.209
£.£39
2.209
2. £09
£.£09
£.239
St.IHffl
(NO EX)
& Onboard
Total Incidence
Due To Sasoline
Vapors (PUL, rat)
5.374
3.192
2.776
2.434
£.332
2. 131
2.256
1.951
1.833
1.743
1.663
1.538
1.524
1.462
1.406
1.746
£. 124
2.185
2.208
£.£09
2.209
2.239
£.£39
2.239
2.239
2. £&3
2. £39
£.239
£.£39
£.£99
2.239
2. £09
2. £39
2. £05
£.£39
St. I (EX)
t Onboard-
Mat ion
Total Incidence
Due To Basoline
Vapors (PUL, rat)
6.759
6.418
5.441
4.739
4.258
3.3£4
3.475
2. 162
£.313
£.559
£.334
2.133
1.951
1.788
1.546
1.546
1.546
1.S46
1.646
1,546
1,646
1,846
1.546
1.S46
1.S46
i.G4S
I. £45
i.546
1.346
L.34&
i.546
1.346
1. 546
i.S4b
t •' ; C
it wTQ
SUM =
NPV =
135.53
44.19
97.66
39.65
93.87
34,89
88.22
33.12
-26,14
85.13
38. Z3
F-15
-------
Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Upper Limit) - Tnsoretical
ar
1986
1987
1968
1989
1993
1991
1993
1993
1994
1995
19%
1997
1996
1999
2888
2081
2882
2803
2684
2805
28%
2887
2608
2839
2818
2811
2812
2313
2814
2815
2816
£817
2018
2819
2828
IH*
pys
St. I (NO EX)
i Onboard-
Nation
Total Incidence
Due To Gasoline
VaoorstPUL.rat)
6.769
6.292
5. 898
4.265
3.822
3.397
3.854
2.747
2.406
2.161
1.942
1.743
1.578
1.412
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.275
73.47
35.26
St.II-Nfl (EX)
& St. I (EX)
& Onbd-Nation
Total Incidence
Due To Gasoline
Vapors(PULrat)
5.724
4.345
3.588
3.876
2.832
2.595
2.487
2.238
2.841
1.984
1.788
1.667
1.557
1.475
1.393
1.495
1.613
1.636
1.645
1.646
1.646
1.646
1.646
!• D*Tu
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
71.35
28.97
St.II-Nfl (NO EX)
£ St. I (NO EX)
4 Onbd-Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.374
3.333
2.313
1.884
1.743
1.685
1.497
1.488
1.284
1.285
1.132
1.866
1.888
8.954
8.986
1.862
1.236
1.265
1.275
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.275
1.276
51.35
20.46
St. 1 1 (EX)
& Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.769
5.897
4.248
3.243
2.953
2.785
2.649
2.525
2.369
2.266
2.171
2.883
2.204
1.938
1.863
1.863
1.939
2.8%
2.192
2.289
2.209
2.289
2.289
2.289
2.289
2.209
2.289
2.289
2.209
'2.209
2.209
2.289
2.209
2.289
2.289
89.20
34.73
St. II (NO EX)
i Onboard-
Nation
Total Incidence
Due To Sasolir,e
Vapors (PUL. rat)
5.759
5.524
3.294
£.£56
1.982
1.848
1.813
1.779
1.721
1.698
1.659
1.630
1.681
1.572
1.545
1.545
1.694
2.000
2.182
2.209
2.209
2.289
2.209
2.289
2.289
2.209
2.289
2.289
2.209
2.239
2.209
2.209
2.209
2.209
2.209
79.18
29.19
St.IJII (EX)
& Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.759
5.718
3.726
2.557
2.259
£.136
2.339
1.594
1.754
1.650
1.574
1.494
1.424
1.358
1.308
1.300
1.375
1.533
1.629
1.645
1.546
1.646
1.645
1.546
1.546
1.546
1.646
1.646
1.646
1.S46
1.546
1.546
1.645
1.546
1.646
69.35
£9.34
F-16
-------
Table F-5. Service Station Incidence Due To Baseline Vapors (Plausible Upper Limit) - Theoretical
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Oaon
COCJ
2010
2011
2012
£013
2014
2015
2016
£017
2018
2019
2020
JM =
>V =
St.Itll (NO EX)
& Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.769
5.276
2.572
1.136
0.994
0.950
• 0.937
0.915
0.881
0.861
0.842
0.824
0.806
0.790
0.774
0.774
0.887
1.118
1.255
1.275
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.275
1.276
1.276
1.275
1.275
1.276
1.275
1.276
49.78
21.52
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
£004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
sua =
NPV =
Baseline
Total Incidence
Due To Gasoline
Vapors (PUL, lice)
4.10
3.99
3.88
3.77
3.70
3.52
3.57
3.53
3.43
3.38
3.34
3.29
3.24
3.19
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
. 3.15
116. 12
37.7-3
' St.II-Nfl*
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.80
3.37
3.23
3.12
3.06
2.99
2.95
2.91
2.84
2.80
2.76
2.72
2.68
2.54
2.60
2.50
2.60
2.50
2.60
2.50
2.60
2.50
2.60
2.50
2.60
2.60
2.60
2.50
2.50
2.60
2.50
2.50
2.60
2.50
2.50
95.46
31 . fiR
, St.II-Nfl*
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.57
2.92
2.78
2.57
2.62
2.56
2.53
2.49
2.43
2.39
2.36
2.32
2.29
2.26
2.22
£.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22'
2.22
2.22
2.22
2.22
32.85
•37 A-5
St. II-Nfl
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.47
2.71
2.54
2.42
2.37
2.32
•p oa
2.2S
-O <7
Ut i l
2.14
2.11
2.87
o 84
i~* VT
2.81
2&1
» Vl
2.81
2.31
2.31
2.31
2.91
2.01
2.31
2.31
? 31
u. -01
2. 01
2.81
2.01
2.81
2.01
2.31
3 31
L.« -Si.
2.91
. 2.01
75.39
S*^ J f
25.16
F-17
-------
Table F-5. Service Station Incidence Due To Baseline Vapors (Plausible Upper Limit) - Theoretical
ar
19BS
1987
1388
1369
1990
1991
1992
1993
1994
1995
1995
1997
1993
1999
2000
2061
2062
2803
2804
£005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
£819
2020
St.IHW
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, dice)
3.08
1.93
1.76
1.65
1.62
1.59
1.57
1.55
1.59
1.48
1.46
1.44
1.42
1.40
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
St.I-Nation
(EX)
Total Incidence
Due To Baseline
Vapors (PUL, aice)
4.18
3.89
3.57
3.36
3.30
3.23
3.19
3.14
3.06
3.02
2.97
2.93
2.89
2.85
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
St.I-Nation
(NO EX)
Total Incidence
Due To Baseline
Vapors (PUL, sice)
4.10
3.81
3.36
3.10
3.94
2.97
2.93
2.89
•2.81
2.78
2.74
2.70
2.66
2.62
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
St. II-Nation
(EX)
Total Incidence
Due To Baseline
Vapors (PUL, aice)
4,10
3.57
2.55
2.14
2.33
1.99
1.95
1.94
1.89
1.86
1.83
1.81
1.78
1.75
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
. 1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
St. II-Nation
flPI (EX)-
Total Incidence
Dus To Gasoline
Vapors (PUL, aice)
4.10
3.96
3.52
£.81
2. 25
2.17
2.88
1.97
1.89
1.86
1.83
1.81
1.78
1.75
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
St. II-Nation
(NO EX)
Total Incidence
Due To Sasoline
Vapors (PUL, mice)
4.10
3.35
2.31
1.27
1.18
1.15
1.14
1. 1£
1.89
1. 88
1.86
1.05
1.03
1.82
1.88
1.88
1.80
1.88
1.38
:.8iB
1.00
1.38
1.38
1.03
1.80
1.80
1.06
1.08
1.38
1.88
1.00
1.39
1.63
1.83
1.00
9UM»
52.42
18.02
104.39
34.44
36.68
32.29
67.63
24.30
78.11
26.22
42.67
17.37
F-18
-------
Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Upper Limit) - Theoretical
fear
19%
1987
1988
1989
1998
1991
1992
1993
1994
1995
19%
1997
1998
1999
2880
2081
2082
2883
2884
2885
2886
2887
2888
2889
2818
2811
2012
2813
2814
2815
2816
2017
2013
2019
2828
St.IHI-Nation
(EX)
Total Incidence
Due To Gasoline
Vapors
-------
Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Upper Limit) - Theoretical
ar *
1985
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2880
2001
2302
2003
2004
2005
2005
2007
2088
2009
2010
2011
2012
2313
2914
2015
2016
2017
2018
2019
2029
St.II-NA
(NO EX)
& Onboard
Total Incidence
Due To Gasoline
Vapors (PUL,*iee)
3.08
1.93
1.68
1.51
1.41
1.32
1.25
1.19
1.11
1.86
1.01
0.95
0.92
0.89
0.85
1.06
1.29
1.32
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
St. I (EX)
J Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL.sice)
4.10
3.89
3.30
2.85
2.58
2.32
2.11
1.92
1.70
1.55
1.42
1.29
1.18
1.08
1.00
1.00
1.00
1.83
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.38
1.00
1.00
St. I (NO EX)
& Onboard-
Nation
Total Incidence
Due To Sasoline
Vapors (PUL,i raice)
4.19
3.81
3.39
2.59
2.32
2.85
1.85
1.67
•1.45
1.31
1.18
1.06
0.95
0.86
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
8.77
0.77
0.77
8.77
0.77
0.77
0.77
8.77
0.77
8.77
3.77
St.II-Nfl (EX)
J St. I (EX)
J Onbd-Nation
Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.47
2.63
2.17
1.86
1.72
1.57
1.46
1.36
1.24
1.15
1.08
1.01
3.95
0.89
0.34
0.91
0.98
0.99
1.00
1.00
1.00
1.30
1.38)
1.00
1.00
1.80
1.190
1.00
. 1.89
1.00
1.38
1.00
1.80
1.00
1.83
St.II-Nfl (NO EX)
i St. I (NO EX)
4 Onbd-Nation
Total Incidence
Due To Sasolins
Vapors (PUL,;sice)
3.38
1.84
1.40
L14
1.26
0.97
0.91
0.35
0.78
0.73
0.69
8.55
0.61
0.58
0.55
0.64
8.75
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
8.77
0.77
0.77
0.77
0.77
0.77
0.77
8.77
0.77
0.77
St. II (EX)
J Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL,raice)
4.13
3.57
2.53
1.97
1.39
1.69
1.51
1.53
1.44
1.37
1.32
1.26
1.21
1.17
1.13
1.13
1.18
1.27
1.33
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
M>V*
46.62
15.85
52.25
23.52
44.54
21.38
43.26
17.56
31.13
12.41
54.08
21.05
F-20
-------
Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Uoper Liait) - Theoretical
Year St.II (NO EX)
J Onboard-
Nation
Total Incidence
Due To Sasoiine
St.IJII (EX)
S Qnboard-
Nation
Total Incidence
Due To Gasoline
St.ISII '(NO EX)
4 Onboard-
Nation
Total Incidence
Due To Baseline
Vapors (Pll, mice) Vapors (PUL, nice) Vapors (PIL, mice)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2801
£00£
2803
2004
2005
2906
2007
2008
2009
2910
2011
2012
2013
.2014
2015
2016
2017
2018
2019
2020
SUM =
M>V =
4.18
3.35
2.00
1.25
1.15
1.12
1.10
1.08
1.04
1.02
1.01
0.99
0.97
0.95
0.94
0.94
1.03
1.21
1.32
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
, 1.34
1.34
48.@0
17.69
4.10
3.47
2.25
1.56
1.39
1.29
1.22
1.15
1.06
1.01
0.95
0.91
0.86
0.82
0.79
0.79
0.83
0.93
0.99
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
42.35
17.79
4.13
3.23
1.5S
0.69
0.60
8.58
0.57
3.55
•0.33
9.52
0.51
3.50
0.49
0.48
0.47
8.47
0.54
'0.68
0.75
0.77
0.77
3.77
0.77
3.77
3.77
3.77
0.77
3.77
0.77
3.77
3.77
3.77
3.77
3.77
8.77
33.18
13.04
F-21
-------
Table F-6. Service Station Incidence Due To Gasoline Vapors (Maxiaua Likelihood Estimate) - Theoretical
iar
1986
1987
1968
1989
1999
1991
1992
1993
1994
1995
19%
1997
1999
1999
2000
2881
2082
2383
2004
2@65
2006
2007
2008
-------
Table F-6. Service Station Incidence Due To Sasoline Vapors (Maxiaum Likelihood Estiaate) - Theoretical
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
ma
2081
2002
2003
£004
2005
20%
2007
2008
2009
2010
2011
2012
2013
£014
2015
2016
2017
2018
2019
2020
SUS =
NPV =
St.I-Nation
(NO EX)
Total Incidence
Due To Baseline
Vapors (MLE, rat)
3.354
3.583
3.156
2.908
2.853
2.790
2.753
2.717
2.643
2.607
2.570
2.533
2.496
2.460
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423'
90.81
33.33
St.II-Nation
(EX)
Total Incidence
Due To Sasoline
Vapors (MLE, rat)
3.854
3.358
2.503
2.006
1.911
1.859
1.845
1.820
1.771
1.746
1.722
1.697
1.673
1.648
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.523
1.623
1.623
1.623
1.623
63.52
£2.83
St.II-Nation
fiPI (EX)
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
3.854
3.718
3.303
2.635
2.146
2.038
1.951
1.850
1.771
1.746
1.722
1.697
1.673
1.643
1.623
1.623
1.523
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.523
1.623
1.523
1.623-
1.523
55.35
24.53
St.II-Nation
(NO EX)
Total Incidence
Due To Sasoline
Vapors (MLE, rat)
3.854
3.145
1.885
1.194
1.107
1.083
1.869
1.054
1.326
1.012
8.997
8.983
0.959
0.955
3.940
0.940
0.940
0.940
0.940
0.940
0.940
0.940
0.940
0.940
9.948
0.940
3.940
0.940
0.940
0.940
8.940
0.940
8.940
3.940
3.940
40.08
15.32
St.Iill-Nation
(EX)
Total Incidence
Due To Sasoline
Vapors (MLE, rat)
3.354
3.256
£.236
1.621
1.534
1.500
1.480
1.461
1.421
1.401
1.382
1.362
1.342
1.322
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.383
52.53
19.76
St. IJII-Nation
(NO EX)
Total Incidence
Due To Sasoline
Vapors (SLE, rat)
3. 854
3.304
1.474
3.564
0.590
0.577
0.570
3.562
•3.547
3.539
0.532
3.524
3.517
3.509
3.501
0.501
3.501
3.581
3.581
3.501
8.501
3.581
0.501.
0. 581
3. 531
e.501
0.501
3.581
3. 501
3.531
3.501
3.581
3. 531
3.581
0.501-
24.99
12.11
F-23
-------
Table F-6. Service Station Incidence Due To Gasoline Vapor?; (Maxiaua Likelihood Estimate) - Theoretical
ar Onboard-Nation
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2091
2002
2£03
2004
2005
2006
2897
sm
2039
2010
2011
2312
2013
2014
2615
2315
2317
2018
2019
2020
Total Incidence
Dug To Gasoline
Vaoors(HE,rat)
3.854
3.751
3.395
3.066
2.682
2.547
2.343
2.150
1.950
1.802
1.669
1.548
1.442
1.344
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.253
1.258
1.258
1.258
St.II-Nfl*
(EX)
& Onboard
Total Incidence
Due To Gasoline
Vapors (MLE. rat)
3.567
3.170
2.848
2.571
2.358
2.171
2.015
1.875
1.711
1.597
1.494
1.399
1.316
1.240
1.172
1.207
1.247
1.255
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St.II-Nfl*
(NO EX)
4 Onboard
Total Incidence
Due To Gasoline
Vapors (ICE, rat)
3.350
2.741
2.451
2.209
2.039'
1.874
1.744
1.627
1.489
1.394
1.307
1.228
1.158
1.894
1.037
1.130
1.235
1.251
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St.II-Nfl
(EX)
4 Onboard
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
3.259
2.545
2.253
2.024
1.880
1.739
1.629
1.529
1.410
1.328
1.254
1.185
1.126
1.070
1.020
1.117
1.227
1.249
1.257
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St.II-Nfl
(NO EX)
4 Onboard
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
2.889
1.817
1.581
1.415
1.328
1.242
1. 176
1.115
1.842
0.992
0.947
0.905
0.867
0.832
0.801
0.994
1.209
1.244
1.257
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St. I (EX)
4 Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
3.854
3. 850
3.298
2.681
2.424
2.177
1.978
1.803
1.690
1.457
1.329
1.213
1.111
1.018
0.937
0.937
0.937
3.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
8.337
0.937
0.937
0.937
0.937
0.937
3.937
0.337
3.937
0.937
NPV*
60.09
25.16
55.51
22.23
51.74
19.87
50.23
18.85
43.78
14.89
49.87
22.03
F-24
-------
Table F-6. Service Station Incidence Due To Gasoline Vapors (Maxiraua Likelihood Estimate) - Theoretical
Year St. I (NO EX) St.lI-Nfl (EX) St.II-Ntt (NO EX) St. II (EX) St. II (NO EX) St.IJII (EX)
1 Onboard- 4 St. I (EX) | St. I (NO EX) i Onboard- & Onboard- J Onboard-
Nation ( Onbd-Nation ft Onbd-Nation Nation Nation Nation
Total Incidence Total Incidence Total Incidence Total Incidence Total Incidence Total Incidence
« i?f I! SUB tSaS°lira °Ue T° SaS°Hne DueT° Sa5oiine D» T° Gasoline Sue To Gasoline
Vaoors(8LE,rat) Vapors (!t£, rat) Vapors (MLE, rat) Vapors (MLE, rat) Vapors (ME, rat) Vapors OLE, rat)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1 fwi
1999
2080
mi
2002
2203
2084
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
5m =
!*V =
3.854
3.583
2. 583
2.428
2.176
1.934
• 1.739
1.564
1.378
1.230
1.106
0.992
0.894
0.804
0.726
0.725
0.726
0.726
0.726
0.726
0.726
0.726
0.726
0.725
0.726
0.726
0.726
0.726
0.725
0.726
0.726
0.726
0.725
0.726
0.726
41.83
20.08
3.259
2.474
2.043
1.751
1.612
1.477
1.371
1.275
1.162
1.084
1.014
0.949
0.892
0.840
0.793
0.852
0.918
0.932
0.937
0.937
0.937
0.937
0.937
0.937
0.937
8.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
3.937
0.937
40.53
15.50
2.889
1.727
1.317
1.073
0.992
0.914
0.852
0.797
8.731
0.686
0.545
0.607
0.574
0.543
0.516
0.505
0.784
0.720
0.726
0.725
0.725
0.725
0.726
0.726
0.726
0.726
0.726
0.725
0.725
0.726
3.726
0.726
3.725
0.726
0.726
29.24
11.65
3.854
3.358
2.419
' 1.847
1.587
1.586
1.508 •
1.438
1.349
1.290
1.236
1.186
1.141
1.099
1.061
1.861
1.134
1.194
1.248
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
50.79
19.77
3.354
3.145
1.876
1.176
1.083
1.8S?
1.332
1.013
0.980
0.962 -
0.945
0.928
0.911
0.895
0.880
0.880
0.965
1.139
1.242
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
45.09
15.62
3.354
7 1*C
ujt i_UD
2 l?o
*-• l&to
1.452
1.309
1.216
1.144
1.878
0.999
8.945
0.395
0.851
0.811
0.773
3.740
37&ffl
* i *rtf
0.784
3.873
0.927
3.937
a 537
v« Jw (
3.937
0.937
3.937
Q Q77
u. j&t
3.937
0.937
3.937
a 077
V. 3wl
a
-------
Table F-6. Service Station Incidence Due To Gasoline Vacors (Maxiaura Likelihood Estimate) - Theoretical
ar
1386
13S7
19BB
1989
1939
1991
1333
1993
1934
1995
19%
1997
1998
1399
203®
2001
2882
2883
2394
2035
28%
2887
2803
2809
£010
2811
2912
2013
2814
2315
2016
2817
2818
2813
2828
St.UII (NO EX)
4 Onboard-
Nation
Total Incidence
Due To Gasoline
Vaoors(i4£,rat)
3.854
3.884
1.465
8.647
8.566
8.547
8.533
8.521
8.581
8.438
8.479
8.463
8.459
8.458
8.441
8.441
8.585
8.636
8.715
8.726
8.726
8.726
8.725
8.726
8.725
8.726
8.726
8.726
8.726
8.726
8.725
8.726
8.725
8.726
8.726
Year
1986
1387
1388
1383
1998
1991
1992
1993
1934
1995
1936
1337
1998
1993
2888
2881
2882
2883
2884
2085
2886
2887
2888
2883
£818
2811
2812
2813
2814
2815
2816
2817
2818
2819
2828
Baseline
St.II-Nfi*
(EX)
St.II-Nfl*
(NO EX)
St.II-Nfl
(EX)
Total Incidence Total Incidence Total Incidence Total Incidence
Due To Gasoline Due To Gasoline Due To Gasoline Due To Gasoline
Vapors (lfl-E, sice) Vapors (MLE, sice) Vapors (!t£, nice) Vapors («LE,aice)
28.34 SUM =
12.25 NPV =
2.75
2.59
2.61
2.54
2.43
2.44
2.41
2.37
2.31
2.28
2.25
2.21
2.18
2.15
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
78.14
25.37
2.58
2.27
2.18
2.18
2.86
2.01
1.93
1.%
1.31
1.88
1.86
1.83
1.88
1.78
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
64.91
21.39
2.48
1.%
1.87
1.88
1.76
1.72
1.78
1.68
1.63
1.61
1.59
1.56
1.54
1.52
1.58
1.53
1.58
1.58
1.58
1.58
1.50
1.53
1.50
1.58
1.50
1.58
1.58
1.58
1.58
1.53
1.58
1.53
1.50
1.50
1.58
55.75
18.45
2.34
1.32
1.71
1.63
1.60
1.55
1.54
1.52
1.48
-1.46
1.44
1.42
1.48
1.38
1.36
1.36
1.3S
1.36
1.35
1.36
1.36
1.36
1.36
1.36
1.36
1.36
1.36
1.36
1.36
1.36
1.36
1.3S
1.36
1.36
1.36
50.73
16.93
F-26
-------
Table F-6. Service Station Incidence Due To Gasoline Vapors (Maximum Likelihood Estimate) - Theoretical
Year
St. II-Nfl St
(NO EX)
:.I-Nation St.I-Nation St.II-Nation St.IHtotion St.II-Nation
ltJU (m Ex> (EX) flPi (EX) (NO EX)
Total Incidence Total Incidence Total Incidence Total Incidence Total Incidence Total
Due To Sasohne Due To Gasoline Due To Basoline Due To Gasoline Due To Saso'iw Due To
Vapors OLE, nee) Vapors (HLE, rice) Vapors OLE,riw) Vapors OLE, rice). Vapors «€£, rice) Vapors
1986
1387
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
•'MVVk
2000
2001
2802
2003
2004
3l>iV?
CB05
2006
2007
•3AAn
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2.87
1.30
1.19
1.11
1.09
1.07
1.35
1.04
1.01
1.00
0.98
0.97
0.%
0.94
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
2.76
2.61
2.40
2.26
2.22
2.17
2.14
2.12
2.0S
2.03
2.00
1.97
1.94
1.32
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
,1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
2.76
2.57
2.25
2.08
2.84
2.00
1.97
1.95
'1.39
1.87
1.84
1.81
1.79
1.76
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
2.75
2.41
1.79
1.44
1.37
1.34
1.32 •
1.30
1.27
1.25
1.23
1.22
1.20
1.18
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.15
1.16
1.15
1.16
1.16
1.16
1.15
1.16
1.16
2.76
2.66
2.37
1.89
1.54
1.45
1.40
1.33
1.27
1.25 '
1.23
1.22
1.20
1.18
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1
-------
Table F-6. Service Station Incidence. Due To Gasoline Vapors (Maxiau* Likelihood Estimate) - Theoretical
Year
St.ISlI-NaUon
(EX)
Total Incidence
Due To Gasoline
Vapors (HE, aice)
St. Ifcll-Nation
(NO EX)
Total Incidence
Due To Gasoline
Vapors (MLE, mice)
Onboard-Nation
Total Incidence
Due To Gasoline i
Vaoors(MLE,niice) '
St.II-Nfl*
(EX)
& Onboard
fatal Incidence
St.II-Nfl*
(NO EX)
& Onboard
' Total Incidence
i Due To Gasoline
>) Vapors (MLE.ciice)
St.II-Nfl
(EX)
& Onboard
Total Incidence
Due To Sasoline
Vapors (MLEjfflics)
19B6
1987
1988
1989
1999
1991
1992
1993
1994
1995
19%
1997
1998
1999
2031
2302
2803
£034
2805
2036
2087
2889
2310
2311
2312
2313
2314
2815
2816
2817
2818
2019
2329
SUH*
WV =
2.76
2.33
1.58
1.16
1.18
1.37
1.36
1.35
1.32
1.00
3.99
8.98
8.%
3.95
8.93
3.93
8.93
8.93
• 8.93
8.93
8.93
8.93
3.93
8.93
8.93
8.93
8.93
0.93
8.93
0.93
8.93
3.93
3.93
3.93
0.93
37.61
14.16
2.76
2.15
1.36
8.48
8.42
8.41
8.41
8.43
8.39
8.39
3.38
8.38
8.37
8.36
8.36
8.36
8.36
8.36
8.36
8.36
8.36
3.36
8.36
8.36
3.36
8.36
8.36
8.36
8.36
8.36
3.36
8.36
8.35
8.36
8.36
17.98
8.68
2.76
2.69
2.43
2.28
2.81
1.82
1.63
1.55
•1.48
1.29
1.28
1.11
1.83
3.%
0.98
3.93
8.98
0.98
3.98
8.98
8.98
8.98
8.98
8.93
3.98
0.98
8.98
8.93
8.38
0.98
0.90
8.90
0.98
8.30
3.30
43.05
18.33
2.56
2.27
2.84
1.84
1.78
1.56
1.44 •
1.34
1.23
1.14
1.37
1.08
8.94
3.89
8.84
8.86
8.89
8.90
0.90
3.93
0.98
8.93
3.38
3.90
8.90
3.98
8.93
8.90
8.98
8.30
8.33
0.33
3.93
8.93
39. 84
15.33
2.40
1.96
1.76
1.58
1.46
1.34
1.25
1.17
1.37
1.00 "
8.94
8.88
0.83
8.78
3.74
0ni
.ol
8.88
3.30
0.30
3.30
8.30
3.30
3.30
3.90
3.90
8.38
a oa
V* J*f
3.98
0.38
8.90
3.38
8.93
0.98
3.93
3.98
37.37
14.23
2.24
1.82
1.61
1.45
1.35
1.25
1.17
1.10
1.31
3.35
8.98
0.85
8.81
8.77
3.73
a aa
v> UV
3.38
9.89
8.38
8.30
3.90
3. 30
0.30
3.30
3.90
3.90
3.98
3.90
8.98
' 3.30
3.38
a. 30
8.33
0.30
8.90
35.39
13.51
F-2S
-------
Table F-6. Service Station Incidence Due To Gasoline Vaoors (Maxim™ Likelihood Estimate) - Theoretical
St.II-Nfl(NOEX) St. I! (EX)
Year St.II-NA St. I (EX) St. I (NO EX) St.II-Nfl (EX)
4 Onboard Nation Nation $ Onbd-Nation 8° Qnbd-Nation' Nation
,otal Incadence Total Incidence Total Incidence Total Incidence Total Incxdence Total I,
Due To Gasoline Due To Sasohne Due To Gasoline Due To Gasoline Due To Sasoli^ Due To Gasolin-
Mice) VaporsfKLMice) Vapors Manors (MLE, .ice) Vapors (HLE, lice) Vapors (MLE,««)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
<*WlflU&
2000
2001
2062
2003
2004
2005
2006
2007
2008
2889
2018
2011
2012
2313
2014
2015
2015
2017
2018
2019
2020
SUM =
KPV-
2.37
1.30
1.13
1.01
0.95
0.89
3.84
0.80
3.75
0.71
0.68
0.65
0.62
0.60
0.57
0.71
8.87
0.89
0.90
0.90
0.90
0.90
0.90
0.90
3.90
0.93
3.93
0.90
0.90
0.98
3.98
0.90
3.90
/ 8.98
0.90
31.37
10.67
2.76
2.61
2.22
1.92
1.74
1.56
1.42
1.29
1.15
1.04
3.95
0.87
3.30
0.73
0.57
0.67
8.67
0.67
3.57
0.67
0.57
0.67
3.57
0.67
3.57
0.57
0.57
0.57
0.57
0.57
3.67
0.67
3.67
3.67
3.67
35.16
15.33
2.75
2.57
2.38
1.74
1.55
1.39
1.25
1.12
•0.98
0.38
3.79
0.71
8.64
0.58
8.52
8.52
0.52
3.52
8.52
0.52
0.52
8.52
0.52
0.52
0.52
0.52
8.52
8.52
0.52
0.52
3.52
3.52
3.52
0.52
0.52
29.97
14.39
2.34
1.77
1.45
1.25
1.16
1.86
3.98 •
8.91
3.33
0.78
0.73
3.68
3.54
8.60
3.57
8.61
8.66
0.67
3.67
8.67
3.67
8.67
3.67
8.57
3.67
0.67
8.67
3.67
3.57
0.67
3.67
3.67
8.67
8.57
3.67
29.11
11.82
2.37
1.24
3.94
8.77
3.71
3.65
3.61
3.57
8.52
3.49 -
0.46
8.44
3.41
0.39
3.37
0.43
0.58
0.52
3.52
0.52
0.52
0.52
8.52
3.52
8.52
0.52
8.52 .
9.52
8.52
0.52
8.52
8.52
8.52
3.52 .
28.95
8.35
2.75
2.41
1.73
1..32
1.21
1.14
( fflQ
i. BO
1.03
3.97
3,92
0.89
8.85
8.82
a 70
VI 1 J
8.76
0.76
0.79
2.86
3.89
3.90
8.98
8.90
3QA
• yo
3.90
8.98
3.30
3.90
3.90
8.98
8.30
3.98
a. 90
3.93
3.30
3.98
35.39
14.17
F-29
-------
Table F-6. Service Station Incidence Due To Baseline Vapors (Maxin.ua Likelihood Estimate) - Theoretical
Year
St. II (NO EX)
1 Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors
-------
Table F-7. Service Station Incidence Due To EDB find EDC - Theoretical
Year
1985
;3S7
1388
1389
1393
1331
1532
1933
1334
1955
i err
1535
; QC"?
i957
1333
193?
2888
2831
2282
2834
28:35
2235
2887
2288
28S9
2310
2311
2312
2313
2314
2315
--•il •< r
CtiO
£317
2318
2313
2222
Baseline
Total
Incidence Due
To EDS
3.38418
8.30365
3.30322
3.88281
3.33243
0.08215
3.C81S2
8. 80143
3.80124
8.88939
3. 30083
3.23858
3.80333
3; 80817
8.88817
3.S3317
3. 88317
8.88317
3. 30317
3. 88817
8. 08317
8.38817
S. 30317
3.30317
3. 83817
3.30817
3.S8817
3.88317
3. 28017
3. 33017
8. 02317
3.88317
3. 33017
3.80817
Si.II-Nfl*
/r*v\
(EX)
Total
Incidence Due
To EBB
3.38386
0.33337
3.80257
3.30231
8.33280
8.88177
3.33149
8.30122
3.88102
3.30881
3.88868
3.30854
8.38041
8.38827
3. 30814
3.88314
a. 38814
3.33814
g. 88814
3,33314
3.38814
3.80014
3.22814
3.88814
8.33814
•3.38814
3.83314
'3.38014
3.38314
3.83814
3.32814
3.33014
8. 38314
3.80814
8. 38314
St. U-Nfl*
(NO EX)
Total
Incidence Due
To EDB
3.88353
3.88265
3.80229
0.80197
8.88171
3. 80151
3.30128
8.80104
3.88087
0. 88873
3.32-858
8.00846
8.88835
3.88823
8.88812
0.30812
3.30312
3.88812
8. '30312
0.30812
8. 38812
8.33812
8.38812 •
8.38812
8. 88812
3.30812
8. S0812
8.38812
3.30012
2.30012
0.30812
•3.38812
3.38312
0.38812
St.II-Nfl
(EX)
Total
Incidence Due
To EDB
8.08352
8.30245
5.80288
8.80178
8.88154
3.88136
3.38115
0.80094
3.38873
3.88063
3.30352
3.89842
3.30831
3.80821
3.80810
3.88818
8.00818
3.88818
3.38810
3.00013
0.30818
3.88813
3.SS813
' -3.83013
3. 38813
3.88818
3.38813
3.88810
8.28818
3. 8331 C
3.83910
3.30810
3.38818
3. £091 3
3.38818
St.II-Nfl
(NO EX)
Total
Incidence Due
To EDB
3.80312
0. 00175
3.30144
3. 80121
8.08185
3.28893
. 3.38875
3. 88864
8.80854
3.80843
0.80836
3.38829
3.88821
3.88814
3.88807
3.38887
8. 03807
3. 38987
3. 88887
8.20387
8.88087
8.30087
3.83387
3. 88897
3.80387
3. 88387
3.88337
8.88887
3.88887
3.30887
3.38837
3. 38387
3.38037
3. 88887
3.38387
St. I -Nat ion
(EX)
Total
Incidence Due
To EDB
3.30418
3.30355
a. 38238
3.38253
3.30213
3. 33133
3.38164
3.38134
3.30112
8.33889
3.38874
3.88868
3.38345
3.30038
8.88015
3.30015
8.38015
3.88815
3.38015
3.38815
8.88815
3.38015
3.38315
3.30015
a. 33815
8.88815
3.28015
3. 88815
3.88815
3.38815
3.38815
3.28815
3.38815
3.88815
3.88015
St.I-Nation
(NO. EX)
Total
Incidence Due
To EDB
3.80418
0.30358
8.38232
3.30235
3.88283
3.80179
0.88152
3.30124
3.S8134
8.38883
8.83369
8.88855
3.08041
3.28028
3.38814
8. §3014
3.38814
3.38814
3.88014
3.38814
3.38014
3.38814
3.83014
3. S3314
3. 38814
8.88814
•3.38014
3. 33814
5.38014
3.33814
3.38014
3.33314
3.83014
3. 30814
3.88814
CI:« -
iDu:'; ™
3.3238
3.3193
3.3252
3.5153
2.3217
•3.8143
3.3159
8.3133
3.3144
3.3893
•2.3275
3.3181
3.32S1
3.3173
F-31
-------
T53i2 r-7. -Service Station Incidence Due To EBB And EDC - Theoretical
tear
St. II-N'ation
fEX)
St.II-Nation
flP! (EX)
St.II-Nation St.IJII-Nation St.UII-Nation Onboard-Nation
-------
Tasle F-7. Service Station Incidence Due To EDS find EDC - Theoretical
YH" ™ S';<^ *^* ' «.1^» B>
Totai Totai Total Total Total
Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due
'° EDB '° m To ™ T° EDB To EDB To EDB To EDB
13SS
1987
13S8
1333
1332
1331
.332
1333
1334
1355
15%
1337
1938
1339
2328
2331
2333
2034 .
-•*,/» fj. 17
c«95
2337
£328
2889
2813
"21 '
^Vl A
£312
2813
2314
2315
£315
£317
231S
£319
2023
8. '28353
3.88255
3.38229
3.83197
0.33151
3.88123
8.38104
3.S8387
0.33373
8.38853
8.33346
3.28335
3.30323
3. 88312
3.30814
0. 08315
3.38316
8. 33317
3. 30817
3.33317
3.83817
3.38817
3.33817
8.80317
•> »*rtfiH7
Vm Vi&GlI
3.83317
3.33317
8.33817
3.83817
3.88817
3.38817
8.38017
8.20817
3. 08817
3.28352
8.33246
3.88288
8.88178
3.30154
0.33135
8.88115
8.33394
3. 38879
3.30353
3.38852
3.83842
3.38831
3.38821
8. 33818
3.33313
d. ®SiS
3.38816
2.88817
.3.33317
0.88817
3.33817
8.38017
3.33817
8.88017
3{%fin t -7
.33817
3.J8317
8.80817
8. 38017
8.80017
3.30817.
3.88017
8.08817
8.80017
3. 83017
8.38312
0.38175
3.08144
8.88121
3.88185
3.88893
3.28879
3.80354
8.38854 -
3.88343
8.83836
3.88829
3.08821
3.88814
8.88887
3.88811
3.33(315
0.83815
3.33817
3.38317
3.80317
0.88817
8.23017
8.88317
0.88817
8. 80317
3.38817
3.83817
0.00017
0.33817
3.30017
3.38317
3.33017
3.88817
-3.33317
0.08418
3.88355
3.88298
3.83253
8.38213
0.88193
0.00164
3.33134
0.00112
8.88889
8.38374
3.38358
3.03845
8.08838
8.80815
0.88315
0.88815
3.83815
8.83815
0.88815
8.88815
3.88815
0.88815
8.88315
8.88315
8.38315
0.30815
8.88015
0.38815
8.80815
3.33315
8.38815
8.88815
9.30815
3. £381 5
3.28418
8.30350
3.38282
3.38235
3.33203
8.38179
3.88152
3.88124
3.88855
8.38841
0.30028
3. 88814
3.83814
3.83014
8.38814
3.38314
3.33014
3.38314
3.88814
3.38814
3.83814
0.38814
3.88014
8.30814
8. -£8814
3.33814
3. £3314
3.80814
<3.£0014
3.83814
3.33352
3.83239
3.08191
3.38158
8. 30137
3.33121
3.88182
8.33384
3.38878
3.38856 *
8.38847
3.38837
3.08028
3.88019
3 iMfliW
v* Wvvj
3.80812
3.30014
3.38815
3. 33815
8.80815
3.38815
8.38815
3.88815
8.38815
3.08815
0.88015
3.88815
3.38815
0.38815
8.38815
3.-S3015
3.38815
8.88815
03fflffli^
. vvviu
8.83815 .
3.38312
3.08167
8.38123
3.33897
3.130884
3.88074
-3.38863
3.23851
3.38343
0.88834
8.83828
3.38823
8.S8817
8.38811
3aat-ȣ
o tiwvb
a^iifloq
• tf'QV'Qj
8.88013
3.38014
Sj 30014
V« WVA"
8.38814
3.38314
3.38814
3.38814
8.38814
8.38314
3.80814
3. 38814
3.08014
3.08814
3.33814
3.38814
8.88814
8.03014
SUM =
3.8227
8.3144
3.0211
3.3134
3.3152
3.3138
a. 3276
3.*
3.8261
0.3173
3.3194
0.8125
3.3148
3.30S8
F-33
-------
'Table F-7. Service Station Incidence Due To EDB find EDC - Theoretical
ar
St. II (EX) St.
i Onboard- i
Nation
Total
II (NO EX) St.ISI! (EX) a.IHI(NOEX)
Onboard-
Mat ion
Total
Incidence Due Incidence Sue
IS8S
19S7
1983
1989
199-3
1991
1992
1993
19%
1S95
19%
1997
1958
1999
283$
2831
£232
£?32
2234
2fc«5
&&
£837
£388
sm
5318
2*11
2*212
£013
2814
8215
seis
2817
2®18
S319
£2£3
To EDB
3.80418
9. 20326
0.30218
0.83156
8.90138
0.33115
9.33898
0.03083
0.33067
0.30853
3.88844
8.88035
0.e3027
8.22313
0.33809
8.80339
8.&811
8.23814
8.83016
0.WB17
0.03317
8.28017
3.83017
8.80317
0.82(317
8.83317
0.28917
0.83817
8.33317
8. 38317
0.28917
8.38017
8.^817
8.30317
3.22317
To EDB
3.38418
0.83385
8.30163
3.83098
8.30374
0.03365
3.08855
0.83345
9.30038
0.83038
•3.08025
3.38020
8.00015
3.03310
0.30905
3.88835
8.83003
0.33813
3.38016
0.30317
3.83017
3.33817
3.38017
0.88017
3.08017
0.88317
8.83817
8.33317
3.80017
0.83017
3.00817
8.88017
8.80917
0.88317
8.38817
i Onboard-
Nation
Total
Incidence Due
To EDB
8.38418
0.38317
3.30194
3.83128
8.30106
3.88094
8.880B0
8.83865
8.83354
8.38843
8.88826
0.30329
8.33822
0.30814
0.80387
0.38807
8.30889
8.88812
0.88815
8.33815
8.08015
3.88315
8.80015
8.08015
8.88015
0.88015
0.38015
3.38815
8.38015
8.88015
8.38015
3.80015
3.88015
3.38815
3.38315
J Onboard-
Nation
Total
Incidence Due
To EDB
3.33418
3.80292
3.33130
8.38352
3.80841
8.88036
3.38338
0.812025
3.38321
8.33017
8.80814
3.08811
8.88808.
0.08086
0.88803
8.08883
0.88805
8.80010
0.88813
3.812314
0.39814
8.33014
3.33814
3.33814
3.33814
3.30814
9.S3014
3.83014
0.30014
3.33814
3.30814
8.83814
3. 2001 4
8.88814
0.^3014
Baseline
Total
Incidence Due
To EDC
8.30540
0.08471
3. 3341 6
3.80363
3.00313
8.03277
• 3.30235
0.00192
0.80160
0.28128
3.80107
8.00085
0.30064
3.38843
3.80021
8.00021
8.00021
0.00021
3.80021
8.00021
3.88021
8.00021
8.88321
3.08021
8.88021
8.38821
3.38321
0.83021
3. 83821
3.00821
3.38821
3.88821
3.33821
3.38821
3.83821
St.II-Nft*
(EX)
Total
Incidence Due
To EDC
3.38498
3.883S5
3. £3344
3.38238
3.38256
3.33228
3. '58193
3.33158
3.33131
r. 88185
8.33088
0.88370
3.38353
3.80035
3.33818
8.88818
3.30018
3.33018
0.3801'8
8.88818
3.38818
0.38018
8.38318
8.88818
3.33013
3.S8018
3.38813
3.83818
3.30818
3.33018
8.38818
3.8081B
8.83818
3.33813
3.83318
3.3211
3.3166
3.3113
3.3139
3.3128
3.31.35
8.3397
3.8384
3.3249
3.3322
3.3211
F-34
-------
"able F-7. Service Station Ircidence Due To EDS find EDC - Theoretical
Year
1985
1937
1988
1989
1593
1991
1592
1SS3
13S4
1555
13SS
1397
1998
1999
2383
2881
£382
£283
£334
£385
2306
£-337
338
£313
2311
£312
£813
2314
£315
£31S
£317
£313
£215
£323
m =
IPV = .
St.IJ-Nfi*
(NO EX)
Total
Incidence Due
To EBC
3.83468
3. 33343
8.38295
3.83255
3.802E8
3. 30495
3.30155
3.20135
3.88112
8.33398
8.33875
8. £0368
8.30045
3. 23838
3.00815
0.80315
3. 28015
8.82315
3. 33015
8.3S015
-3. £2815
3.30815
3. 3S015
3.33315
3. 33315
3.33315
8.30815
3. 33815
0. 23815
8.^015
2.33315
:. i'280
3.3135
St.II-Nfl
(EX)
Total
Incidence Due
To EDC
0.08455
3.30317
0.38269
3.30238
8.80199
3. 30175
0.30149
3.38122
8.38181
3.38881
3.88858
8.30054
8.38041
0.30027
0.00014
0. 30814
8.33814
8. 33814
8.30814
3.30814
8.38314
8.88814
8.88814
3.08814
8.88014
3. 80814
3.03314
0.30014
3.38814
3iTAAi ft
.!ffio814
3.3S014
3.33014
3.33314
3.30814
8.38214
.3.3257
3.3171
St. II-Nfl
(NO EX)
Total
Incidence Due
To EDC
8.S3483
3.08225
3.38185
3.38157
0.83136
3.38120
0.38181
3.00883
8.38069
3.33855
8.83046
0.38837
8.888£8
3.08818
8.30339
0.30809
0. 30639
3.30389
3.83809
0.30309
3.82339
8.30099
3. 33389
3.08809
3.8S339
3.38089
3.83339
0. 38389
8. 022129
3. 30809
3.32339
3.38309
3.33339
0.30039
3.33389
3.3166
3.3127
St.I-Nation
(EX)
Total
Incidence Due
To EDC
3.38548
3.08468
3.80385
0.30325
8.38282
0.30253
0.38211
0. S3 173
0.S0144
0.30115
3.88096
0.33377
3.0885S
0.38333
8.03819
8.83319
0.00019
3.33819
8.33019
3.33319
8.33819
8.33813
3.88019
. 0.30813
0.88019
3.33*19
3,33319
3.33319
3.80819
0.38019
8.33819
3. '2031 9
0.3801-9
0,33319
0.30019
3. 3356
3.3233
St.I-Naiion
(NO EX)
Total
Incidence Due
To EDC
2.03548
3.88452
8.30364
3. 33383
3.83262
3.38231
3.38195
3.03163
3.30134
3. 08137
3.80839
3.33071
8.80053
8.08836
8.83818
3.33813
8.83818
3.38818
3.33318
3.30018
6.33818
8. 38818
3.33818
3.33318
8.30018
3.33013
3.33318
3.38318.
2.38318
0.33818
3.38818
3.23318
3.80818
3.00818
3.33818
3.3337
3. 3223
St.Il-Nation
(EX)
Total
Incidence Due
To EDC
3.33540
3.33423
0.38282
3.08231
3.88168
3.08149
8. 33125
3.03183
8.33886
3. '38059
8.30357
3. 38846
8.88034
8.08323
3.88811
3.33011
0.03811
3.03311
8.33811
3.00011
3.33011
8.33811
3.88811
8.30011
3.38311
0.00811
0.33811
0. 3001 1
3.33011
3.33811
3.38811
3.33311
8.30011
3. 03811
3.38311
3.3254
_ - , .,_
St.II-hiation
flPI .(EX)
Total
Incidence Due
To EDC
0.33540
3.33467
3.30375
0.30267
•3.03198
3.38163
3.03134
8.20185
3.38085
3.03069
3.38057
8.33346
8.08034
8.33823
3.38011
3.33011
8.88011
3.33011
3.03811
3.38311
3.68811
8.30011
8.00011
3.33311
3.83011
3. 3001 1
3.38011
3.30311
3.88011
0.30311
3.381311
0.20011
3.83811
3.33811
3.33011
3. 3233
a *3
-------
Taale "-?. Service Station Incidence Due To cuB find EDC - Theoretical
Yiar .St. !I-Natiort St. ISII-Naticn St. 1411-Nation Onboard-Nation
JNC EX) (EX) (NO EX)
Total
Total
Total
Total
St. II-Nfl*
06
C*. 33806
8.3188
3.8143
3.S0540
8.S0489
3.88251
8.80165
3.38137
8.80121
•3.80183
3.33084
3.88878
3.80356
3.08047
8.80037
9.08028
8.88819
3.88009
3.90889
3.30009
9.08889
3.08009
9.33889
0.88009
9.00089
8.08009
3.33889
3.08009
3.3S309
3.00009
8.38089
0.88809
8.30009
3.38809
8.3-3809
8.88809
3.38089
0.88809
2.2225
8.3154
9.88543
0.38377
8.88167
3.38067
9.88052
8.30345
8.08839
3.22032
8.88827 '
9.38821
0.80018
8.30014
0.08911
3.28307
0.80804
3.98384
3.08904
3.88084
0.88004
8.30384
0.08004
3.0S084
9.08804
3.32384
9.08804
3.3-3084
0.00084
8.28084
0.80004
3. £8304
3.03004
8.88884
0.00004
•3.30004
9.S8004
2.3149
0.3122
0.^543
3.08471
^0. 88416
3.558363
9.28313
8.03277
3.38235
8.08192
3.03163
8.00128
0.00187
8.00085
0.IS364
0.2(2043
0.128021
0.80921
0.33821
0.8C021
3.S3321
8.88021
•3. 63821
3.30821
3.80321
3.80321
0.S021
0.08021
0.30321
8.08321
0.83821
3.33321
•3. -331321
8.00821
3.33021
0.0S821
@.§0021
8. -3384
•3.0249
8.28498
8.303%
0.38344
3.30298
0.08258
3.30228
3.30193
3.93158
9.88131
3.00185
8.38088
8.08070
8.88853
3.00035
8.08018
0.30819
9.38021
8.30321
3.39021
3.38021
0.00021
0.150021
0. 20921
•3.20821
0.08021
8.83021
3.80821
0.08921
3.08021
8.08821
3. 00021
3.88821
3. 130021
3.08021
3.88021
3.8338
3.0212
3.30463
3.38343
3.802%
3.88255
0.38220
3.30195
8.38165
3.08135
3.88112
0.30898 "
8.08875
3.08868
3.03845
8.88830
3.00015
8.00818
3. -38021
3.88021
0.08021
8.88021
0.00821
0.80021
9.30021
3.88021
3.08821
3.80021
3.23021
3.00021
0.20021
3.30821
3.33«21
•3.32021
0.08021
9.30021
0. 120021
0.3293
8.0136
8.00455
3.00317
0.33269
0.30230
0.03199
3.88176
3.00149
0.83122
8.38181
0.83881
0.88063
3.88054
8. '20341
3.88027
8.88014
3.00017
0. '38920
3.88021
8.80021
3.08821
0.08321
8.08021
3.08821
3.38021
0.30021
3.30021
3.&J021
0.00021
0.S0021
3.88021
3.30321
0.30021
3.38921
3.38021
0. 80021
3.2272
•3.3173
F-36
-------
Table F-7. Service Station Incidence Due To EDB find EDC - Theoretical
Year
1386
1987
1388
1989
1998
1391
1992
1393
1994
1395
1996
1397
1998
1993
mm
2881
£302
2383
£"234
2885
2886
2307
2008
2089
•313
2811
2812
2813
314
2815
2216
2817
2318
2013
£323
SUM =
NPV =
St.IIHW
(NO EX)
S Critoard
Total
Incidence Due
To -EDC
3.38433
3.38225
0. 03186
0.00157
a. eei3&
3.30120
3.88181
3.38083
3.2-0859
3.82855
3.38846
3.38337
3.33028
a. 08818
3.38889
8. 08014
3. 30020
0.08821
0:08821
3.08821
0. 08021
0.88021
0.80621
2.03021
3. 30021
0.00021
0.38821
•2.08821
3.50021
3. 38021
3.28021
0. 3882!
3. 08021
3.e8321
0.30021
3.0209
0.0132
St. I (EX)
J Onboard-
Nation
Total
Incidence Due
To E33C
0.30549
0.88468
0.00385
0.30325
8.09282
0.30258
3.88211
0.88173
8.08144
0.38115
8.00095
8.08077
0.08858
8.08838
8.88819
0.88813
0. 08013
3.88019
0.88819
3.88019
0. 00819
0.08019
0. 30013
3.88813
8.88813
3.88019
3.30013
8.08319
0,30313
8.08019
8.00019
3.38319
8.03819
3.38019
3.00819
8.5356
3.0223
St. I (NO EX)
I Onboard-
Nation
Total
Incidence Sue
To .EDC
0.83548
8.38452
0.88364
0.38383
3.88262
3.38231
8.881%
3.00163
8.08134
3.88137
8.38889
8.88871
0.88853
3.38836
3.88813
8.90018
0.88013
8.38018
0. 28013
3.-38818
0.88818
8.88818
8. 33813
8.88818
0.80818
0.28318
0. 00813
8.38818
3.38318
3,38318
'3. 30013
3.88818
3.38813
3.38318
3.30318
3.3337
3.3223
J St. I (EX)
4 Onbd-Nation
Total
Incidence Due
To EDC
3.08455
8.88389
0.33247
0.88284
8.88177
0.88156
8.38132
0.33898
3.38872
8.83848
8.03836
0.88824
0.88812
0.88015
•3.03818
0.38319
•3.03819
3.08819
3.38819
3.38819
8.38019
3.33019
3.88819
•3.03013
3.33319
3.33019
3.88019
3.33819
3.83819
0.88819
3.38819
0.00819
3.3251
a. 3162
[-NA (NO EX
,1 (MO EX)
'.fad-Nation
Total
:dence Due
o EDC
8.23433
0.38215
3.83159
8.88125
8.33188
3.03395
3.88381
3.88866
0. 08-355
3.38844
3.03837
8.38829
3.88822
8.08815
8.38887
3. 38012
0.88017
8.38317
0.30818
8.88318
3.38818
0.88318
3.88818
8.88818
3.88813
8.38318
8.08813
0.38013
3.00818
3.38818
3.03013
3.33318
3.83013
3.83318
8.08818
3.8181
0.8115
St. 1 1 (EX)
& Onboard-
Nation
Total
Incidence Due
To EDC
8.88540
3. 38428
3.83282
3.80281
3.83168
3.S8149
3.28125
3.30183
3.38885
3.88869
0.33857
3.38846
0.38834
0.38023
3.08811
3.38811
8.88814
3.38018
3.03821
3.88821
3.30821
0.30821
8.38021
3.88821
8.80821
3.30821
3.38021
8.30321
3.38821
3.38321
0.80021
3.38821
3.38821
3.83821
3.08821 •
3.3272
0. 0131
St. II (NO EX)
4 Onboard-
Nation
Total
Incidence Due
To EDC
0.38548
3.80393
8.03210
0.38117
3.08395
3.38884
3.39871
3.38858
8.38848
8.38039
0.38032
3.80826
3.33819
3.38013
3.38886
0.38885
8.08818
3.08817
3.00321
0.80821
8.38821
8.08821
0.88321
3.83321
3.03821
0.83821
3.38821
8.03021
3.03321
0.33321
8.00021
8.33821
8.83821
3.30021
8.0902!
0.3215
0.8146
F-37
-------
• Table F-7. Service Station Incidence Due To EDB find EC - Theoretical
Year Si. III! (EX) St. IHI (NO EX)
1366
1387
1388
1333
1233
1991
1392
1993
1994
1993
1926
1997
1998
1999
2«08
2831
2002
2283
sew
£895
2836
2837
2e0a
i239
2818
1311
2812
£313
2814
2015
£315
2817
ftift
3®19
£828
&M*
NPV =
S Onboard-
Nation
Total
Incidence Due
ToEDC
3.23543
3.C8A03
3.^51
§.30165
3.00137
3.^121
9.^103
•3. 38884
9.20070
•3.300S&
3.8^47
3. 00837
a.eeesa
3.?^319
a.e-eera
0.03^9
3.8^12
ia.e-2816
3.80019
9.^319
8.89919
8.^819
@.0eei9
8.88819
3.88819
3.83019
8.88819
0.03319
8.33319
3.33819
8.28819
8.*819
8.38819
8.83819
3.28319
0.3244
3.81S5
4 Onboard-
Nation
Total
Incidence Due
To EDC
3.38548
8.83377
3.80167
8.80867
0.88052
0.20846
'3.03039
8.80832
8.08827
8.20821
0.88018
0.80314
8.08011
8.08807
8.08884
8.08834
0.03087
0.00313
0.00817
0.08818
0.08018
0.00018
8.03018
0.00018
0.00018
8.08318
0.08018
0.00018
3.00318
8.00818
0.88018
3.00318
8.88018
0.00318
0.88018
8.3175
3.3125
F-38
-------
Table F-8. Self-Ssrvice Incidence Due To Benzene - Theoretical (Selfsine.wks)
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
20191
2802
2004
2005
2006
2007
2808
2009
2010
2011
2012
2013
2014
2015
2016
2017
2013
2019
2020
Baseline
Incidence
Due To
Benzene
(Theoretical)
4.42
4.24
4.08
3.92
3.80
3.68
3.59
3.50
3.38
3.30
3.23
3.16
3.10
3.03
2.%
2.96
2.%
2.%
2.%
2.9S
2.96
2.95
2.%
2.96
2.%
2.96
2.96
2.96
2.96
2.36
2.96
2.96
2.%
2.36
2.%
St.II-Nfl*
(EX)
Incidence
Due To
Benzene
(Theoretical)
4.25
3.32
3.74
3.59
3.48
3.37
3.29
3.21
3.89
3.02
2.96
2.98
2.83
2.77
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
St.II-Nfl*
(NO EX)
Incidence
Due To
Benzene
(Theoretical)
4.18
3.78
3.61
3.45
3.35
3.25
3.17
3.09
2.98
2.91
2.85
2.79
2.73
2.57
2.61
2.51
2.51
2.61
2.51
2.61
2.61
2.61
2.61
2.61
2.51
2.61
2.61
2.61
2.51
2.61
2.61
2.61
2.61
2.51
2.61
St. II-Nfl
(EX)
Incidence
Due To
Benzene
(Theoretical)
3.98
3.37
3.17
3.01
2.92
2.83
2.76
2.69
2.60
2.54
2.49
2.43
2.38
2.33
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.23
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
St.II-Nfl
(NO EX)
Incidence
Due To
Benzene
(Theoretical)
3.81
3.134
2.85
2.71
2.62
2.54
2.48
2.42
2.33
2.2B
2.23
2.19
2.14
2.89
2.05
2.85
2.35
2.05
2.35
2.35
2.S5
2.35
2/35
£.85
2.85
2.05
2.35
2.05
2.05
2.65
2.35
2.35
2,85
2.35
2.95
SUM =
NPV =
112.61
37.98
133.35
34.95
99.66
33.76
87.33
29.85
73.58
27.36
F-39
-------
Table F-8. Self-Service Incidence Due To Benzene - Theoretical (Selfsine.wks)
Year
sua =
\
198S
1987
1933
1989
1999
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2332
2003
2004
2005
2006
£007
2208
2009
2018
2011
£012
2013
2014
£015
2016
2817
£918
•019
5320
St.II-Nation
(EX)
Incidence
Due To
Benzene
(Theoretical)
4.42
3.57
2.16
1.38
1.23
1.19
1.15
1.13
1.09
1.07
1.35
1.03
1.93
0.98
0.96
0.%
8.96
0.96
0.%
0.%
0.96
0.%
0.96
0.%
0.95
0.9S
0.S5
0.96
0.26
0.95
0.96
0.95
0.95
0.%
0.95
42.63
18.32
St.II-Nation
flPI (EX)
Incidence
Due To
Benzene
(Theoretical)
4.42
4.18
3.58
2.41
1.62
1.47
1.33
1.18
1.09
1.07
1.05
1.03
1.88
0.98
0.%
0.96
0.%
0.95
0.96
0.96
0.96
0.96
0.95
0.95
0.95
0.%
3.95
0.95
0.95
2.55
0.96
0.95
9.95
3.95
0.%
46.47
21.03
St. II-Nation
(NO EX)
Incidence
Due To
Benzene
(Theoretical)
4.42
3.28
1.33
0.29
0.19
0.18
0.18 .
0.17
0.17
0.16
0.16
0.16
0.15
0.15
0.15
0.15
3.15
0.15
0.15
3.15
0.15
0.15
0.15
0.15
3.15
0.15
3.15
0.15
0.15
0.15
0.15
0.15
0.15
3. 15
S.15
14.25
9.37
Onboard
Incidence
Due To
Benzene
(Theoretical)
4.42
4.24
3.59
3.18
2.75
2.35
2.83
1.72
1.41
1.17
0.97
0.79
0.62
0.47
0.34
8.34
3.34
0.34
0.34
•3.34
0.34
0.34
8.34
3.34
3.34
0.34
9.34
0.34
3.34
0.34
0.34
3.34
0.34
3.34
3.34
37.00
22.08
St.II-Nfl*
(EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
4.25
3.92
3.39
2.91
2.52
2.16
i.86
1.58
1.29
1.83
0.89
0.72
8.57
0.44
0.32
0.33
0.34
0.34
3.34
0.34
3.34
3.34
3.34
0.34
0.34
3.34
3.34
0.34
3.34
3.34
3.34
3.34
0.34
3.34
3.34
34.71
20.53
F-40
-------
Table F-8, Self-Service Incidence Due To Benzene - Theoretical (Self sine, wks)
Year
I9B5
1337
19-38
1983
1590
1331
1932
1333
19-34
1335
13%
1337
1398
1939
2800
2001
•2802
2203
2034
2005
283S
2607
2208
2303
2810
2011
2012
2813
2314
2015
2316
2017
2318
2013
SUM =
St. II-Nfi*
(NO EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical}
4. IS
3.78
3.26
2.80
2.43
2.28
1.79
1.52
,1.25
1.34
8.86
8.70
0.55
0.42
0.31
0.32
3.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
3.34
0.34
3.34
0.34
33.73
13.90
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
3.98
3.37
2.87
2.45
2.12
1.82
1.56
1.33
1.09
0.31
0.75
0.61
0.4S
0.37
0.27
0.38
0.33
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
30.75
17.87
' St. II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
3.81
3.04
2.58
2.20
1.91
1.64
1.48
1.20
0.98
0.82
0.68
0.55
0.44
0.34
8.25
0.29
0.33
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
28.58
16.38
St.II-Nation
(EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
4.42
3.57
1.35
1.11
0.30
0.78
0.67
0.57
8.47
0.40
0.33
0.27
0.22
0.17
0.13
0.13
0.18
0.27
0.33
0.34
0.34
0.34
8.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
•3.34
0.34
0.34
0.34
22.34
13.31
St. II-Nation
(NO EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
4.42
3.28
1.20
3.25
0.15
0.13
0.12
0.10
0,09
0.08
0.07
0.06
0.06
0.05
0.04
0.04
0.11
0.25
0.33
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
8.34
3.34
0,34
16.31
9.67
F-41
-------
Tablg F-9. Self-Service Incidence Due To Gasoline Vaqors (Plausible linear Limit) - Theoretical
Year
1985
1987
1988
!989
i2sa
1591
1592
1993
1994
1995
19%
1997
1998
1999
2M3
£031
302
£003
2004
2805
2006
£007
m&
2309
2010
2011
2812
£013
2014
2015
£016
2317
£318
2319
5028
Baseline
Incidence
Due To
Sas Vaoors
{PUL,ratJ
40. £6
38. S3
37. 66
36.40
35.51
34.60
33.56
33.32
32.£9
31.70
31.16
30.62
20.38
£9.54
£9.01
29.91
29.81
29.01
29.31
£9.01
29.01
29.01
29.01
29.01
29.01
29.01
29.01
£9.01
29.01
29.01
29.01
29.01
29. '31
29.01
£9.01
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vaoors
(PUL,rat)
38.80
35.99
34.59
33.32
32.51
31.67
31.38
30.50
£9.56
£9.02
£8.5£
£8.03
27.54
27.04
£6.55
25.55
£5.55
£6.55
£5.55
£6.55
25.55
26.55
25.55
26.55
26.55
£5.55
26.55
£6.55
£6.55
25.55
£6.55
£5.55
26.55
£5.55
26.55
St.II-Nfl*
(NO EX)
Incidence
Due To
Sas Vaoors
(PUL,rat)
38.14
34.71
33.33
32.10
31.32
30.51
£9.95
£9.38
28.48
27.95
£7.48
£7.01
25.53
£6.06
25.58
£5.58
£5.58
£5.58
25.53
£5.58
25.58
25.58
25.58
£5.58
£5.58
25.58
£5.58
£5.58
25.58
£5.58
£5.58
£5.58
25.58
£5.58
25.58
St.II-NA
(EX)
Incidence
Due To
Gas Vaoors
(PULrat)
35.27
30. S0
29. £8
£8.30
£7.31
£5.61
25.12
25.62
• £4.83
24.38
£3.97
£3.55
£3.14
22. 7£
22.31
22.31
22.31
££.31
££.31
£2.31
£2.31
22.31
22.31
22.31
22.31
22.31
£2.31
££.31
22.31
££.31
£2.31
22.31
22.31
£2.31
22.31
St.II-Nfl St.II-Nation St.II-Nation St.II-Nation
(NO EX) (EX) P.PI (EX) . (NO EX)
Incidence
Due To
Sas Vauors
(PULrat)
34.71
27.88
£5.33
£5.14
£4.53
23.90
23.45
23.01
22.30
21.90
21.52
21.15
£0.78
£0.41
£0.84
£0.04
20.04
£0.04
£0.04
£0.04
£0.134
20.04
£0.04
20.04
23.84
£0.04
20.04
20.04
23.84
20.04
20.34
23.04
20.84
£0.04
£0.04
Incidence
Due To
Gas Valors
(PUL,rat)
40. £6
32.88
19.92
12.58
11.51
11.21
• 11.00
10.79
10.45
10.27
10.10
9.92
9.75
9.57
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.4«
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
Incidence
Due To
Gas Vaoors
(PULrat)
40.26
38.40
32.31
2£.3B
15.11
13.78
12. SI
11.25
10.46
10.27
10.10
9.92
9.75
9.57
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.48
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
Incidence
Due To
Gas Vaoors
(PUL,rat)
40. £5
38.15
12.27
2.73
1.75
1.70
1.67
1.64
1.59
1.56
1.53
1.51
1.48
1.45
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
SIM*
1085.14
358.19
995.72
333.23
960.15
318.97
841.16
£81.92
757.75
255.47
407.60
158.66
443.53
196.48
131.30
90.90
F-42
-------
Table F-9. Self-service Incidence Due To Gasoline Vasors (Plausible Unper Li.it) - Theoretical
Year
1385
1387
1988
1389
1S33
1591
1392
1933
1934
1395
1396
1997
1398
1999
--. Tk|>/%
dew
£202
•2233
£834
2335
£306
;307
2388
369
£318
£311
2312
.£•313
2814
£315
£316
£317
2018
£019
2320
Su* =
"IrV =
Onboard
Incidence
Du= To
Sas Vapors
CPUL,rat) '
42.25
33.33
33.32
29.12
25.24
21. S3
13.55
15.81
12.95
13.38
8.92
7.22
5.75
4.^8
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3,24
3.24
3.24
3.24
3.24
3.24
3.24
2.24
3.24
3.24
341.41
£32.53
St.II-Nft*
(EX)
w/ Onboard
Incidence
Due To
Bas Vapors
(POL, rat)
38.83
35.99
31.35
25.65
23.11
19.73
16.39
14.49
11.88
9.91
8.19
S.S3
5.28
4.35
2. '39
3.39
3.21
3.23
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
323.42
188.41
St.II-Nfl»
(NO EX)
w/ Onboard
Incidence
Due ~o
Sas Vaoors
(Pit, rat)
38.14
34.71
29.94
25.53
22.27
19.37
16.38
13.97
11.45
9.55
7.98
5.48
5.10
3. '32
2.89
3.84
3.28
3.23
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
311.92
182.61
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(?UL,rat)
3S.27
33.93
26.30
22.42
19.44
16.65
14.31
12.21
' 10.81
8.35
6.92
5.51
4.48
3.45
2.56
2.34
3.15
3.21
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
28^. 15
153.99
St. II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Gas Vapors
(PUL,rat)
34.71
27.88
23.65
20.14
17.47
14.37
12.87
18.98
9.01
7.53
6.24
5.07
4.05
3.13
2.33
2.71
3.14
3.21
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
264.15
158.35
St.II-Nation
(EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(PUL,rat)
40.26
32.80
17.98
18.28
8.26
7.18
6.13
5.25
4.34
3.66
3.06
2.52
2.34
1.51
1.24
1.24
1.68
2.53
3.14
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
236.84
122.11
St.II-Nation
(NO EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(PUL,rat)
40.26
30.16
11.05
2.26
1.35
1.28
1.38
0.95
0.84
0.75
3.57
8.58
0.54
0.48
8.43
3.43
1 flfi
x. TO
2 35
W« Ww
3.12
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
151.43
88.83
F-43
-------
Taale F-9. Self-Service Incidence Due To Gasoline Vaoors (Plausible Uoser Limit) - Theoretical
Year
1985
1337
13S8
1983
1993
1991
1932
1993
1994
1995
19%
1997
1998
1999
20SS
2201
seas
seas
384
2205
22%
2837
2008
2309
£31®
2311
£812
2313
£014
2915
2016
2317
2018
2819
2820
Baseline
Incidence
Due To
3as Vacors
{pUL.Hice)
24.25
23.55
22.79
22.22
21.49
20.93
20.54
20.16
19.54
19.18
18.85
18.53
18.20
17.88
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vanors
(PUL. Eice)
23.45
21.73
23.88
20.12
13.63
19.12
18.77
18.41
17.34
17.52
17.22
15.92
16.53
15.33
15.03
16.03
16.03
16.03
16.03
16.03
15.33
16.83
16.03
16.03
16.03
16.03
16.03
15.03
16.03
16.03
15.03
16.03
16.03
16.03
15.03
St.II-Nftt
(NO EX)
Incidence
Due To
Sas Vaoors
(PUL, nice)
23.35
20.34
20.11
19.35
18.89
18.40
18.06
17.72
17.18
15.86
15.58
16.29
16.00
15.72
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
St.II-Nfl
(EX)
Incidence
Due To
Sas Vaoors
(PUL. Mice)
£1.33
18.59
17.60
16.82
15.41
15.39
15.63
15.40
' 14.92
14.65
14.40
14.15
13.90
13.55
13.40
13.40
13.40
13.40
13.48
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
St. II-NA St. II-Nation St. II-Nation St. II-Nation
(NO EX) (EX) flPI (EX) . (NO EX)
Incidence
Due To
Sas Vaoors
(PUL. sice)
23.32
16.71
15.77
15.05
14.53
14.31
14.05
13.78
13.36
13.11
12.83
12.57
12.44
12.22
12.00
12.30
12.00
12.30
12.tt
12.00
12.80
12.00
12.00
12.00
12.30
12.80
12.20
12.30
12.^0
12.00
12.08
12.30
12. S3
12.80
12.30
Incidence
Sue To
Gas Vapors
(PUL, mice)
24.35
13.76
11.80
7.34
5.63
6.46
6.34
6.22
5.33
5.32
5.32
5.72
5.52
5.51
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
Incidence
Due To
Eas Vapors
(PUL,!aice)
24.36
23.23
13.48
13.34
8.36
3.05
7.33
6.50
6.33
5.92
5.82
5.72
5.62
5.51
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
Incidence
Due To
Sas Vaoors
(PUL, mice)
24.36'
18.12
7.37
1.18
8.59
3.57
0.55
3.35
3.54
3.53
0.52
8.51
0.53
3.49
3.48
3.48
3.48
3.48
3.48
0.48
3.43
8.48
0.48
3.48
3.43
3.48
3.48
3.48
3.48
0.48
0.48
3.48
3.48
8.48
3.43
SUM
6w&«55
216.72
601.21
199.42
579.20
192.45
535.55
169.51
453.92
153.14
237.20
99.42
259.44
115.63
66.19
51.29
F-44
-------
-able F-9. Self-Service Incidence Due To Baseline Vapors (Plausible Upper Liait) - Theoretical
Year
198S
1937
1988
1939
1952
1991
1992
1393
1994
1995
19%
1997
1998
1999
c'380
£031
£002-
£033
£884
£085
2086
•2007
2S08
£039
£018
2011
2012
2013
2014
2015
2016
£017
2018
2019
2020
SUM =
NPV =
Onboard
Incidence
Bus To
Gas Vapors
(PlL,sice)
24.35
23.55
20,45
17.62
15.27
13.07
' 11.22
9.57
7.34
6.54
5.40
4.37
3.48
2.66
1.96
1,95
1.96
1.X
1.96
1.X
1.X
1.95
1.96
l.X
1.96
1.96
1.96
l.X
1.X
1.96
i.X
1.96
l.X
1.95
1.96
266.57
122.54
St.II-Nfl*
(EX)
w/ Onboard
Incidence
"us To
•Sas Vaoors
(PUL,3iice)
23.45
21.73
18.75
15.10
13.96
11.95
10.26
8.75
7.17
5.98
4.94
4.01
3.19
2.45
1.31
1.87
1.94
1.95
1.95
1.95
1.96
1.%
1.96
1.95
1.X
1.96
1.96
1.96
l.X
1.96
1.96
1.96
1.96
1.95
l.X
193.59
113.80
St. II-Nfl*
(NO EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(POL, mice)
23.05
20.34
18.05
15.50
13.44
11.50
9.88
8.43
6.91
5.76
4.76
3.86
3.08
2.36
1.74
1.34
1.94
1.95
1.X
1.95
l.X
l.X
1.95
1.36
1.96
1.96
1.96
1.96
1.96
1.95
1.95
1.95
1.95
1.X
1.96
188.32
110.21
'St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(PH., nice)
21.89
18.59
15.81
13.47
11.68
10.00
8.50
7.33
' 6.02
5.02
4.16
3.38
2.59
2.07
1.54
1.71
1.91
1.94
1.95
1.96
1.95
i.X
1.96
1.95
1.X
1.96
1.95
1.X
1.95
1.96
l.X
1.96
1.96
1.36
1.96
171. 14
98.69
St.II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Gas Vapors
-------
Table F-10. Self-Service Incidence Due To Sasoline Vaoors (Haxiaua Likelihood Estisate) - Theoretical
Year
NPV
1986
1387
1988
1983
1933
1991
1992
1993
1994
1925
19%
1997
1998
1999
sm
2031
2002
5003
2884
£835
2866
2007
2038
2009
2010
2011
2012
£013
2014
2015
2016
£017
2018
£019
2029
=
a
Baseline
Incidence
Due To
Gas Vapors
(ME, rat)
22.92
22.17
21.44
20.73
20.22
19.70
13.33
18.97
18.39
18.95
17.74
17.44
17.13
16.82
16.52
16.52
16.52
16.52
16.52
16.52
16.52
16.52
16.52
15.52
16.52
15.52
16.52
16.52
16.52
16.52
16.52
16.52
16.52
15.52
16.52
617.88
203.95
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vapors
(ME, rat)
22.89
20.49
19.69
18.97
18.51
18.03
17.70
17.36
16.83
15.52
16.24
15.%
15.68
15.40
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
555.97
188.04
St.II-Nfl*
(NO EX)
Incidence
Due To
Gas Vapors
(ME, rat)
21.72
19.76
18.98
18.28
17.83
17.37
17.05
16.73
16.21
15.92
15.65
15.38
15.11
14.84
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
546.72
181.62
St.II-Nfl
(EX)
Incidence
Due To
Gas Vapors
(ME, rat)
20.65
17.60
16.57
15.94
15.55
15.15
14.87
14.59
• 14. 14
13.88
13.65
13.41
13.17
12.94
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
478.96
160.53
St. II-Nfl St. II-Nation St. II-Nation St. II-Nation
(NO EX)
Incidence
Due To
Gas Vapors
(ME, rat)
19.76
15.37
14.99
14.32
13.97
13.51
13.36
13.10
12.70
12.47
12.25
12.04
11.83
11.52
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
431.47
145.45
(EX)
Incidence
Due To
Gas Vapors
(ME, rat)
22.32
18.68
11.34
7.22
6.55
5.38
• 6.26
6.15
5.96
5.85
5.75
5.55
5,55
5.45
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
232.09
96.24
flPI (EX)
Incidence
Due To
Gas Vapors
(ME, rat)
22.32
21.87
. 18.40
12.74
8.60
7.85
7.18
5.40
5.95
5.85
5.75
5.65
5.55
5.45
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
252.55
111.88
(NO EX)
Incidence
Due To
Sas Vapors
(ME, rat)
22.32
17.17
6.39
1.55
1.00
0.97
8.95
0.93
3.91
0.89
0.87
0.86
0.84
0.83
0.81
0.81
0.81
0.81
8.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
74.75
51.76
F-4b
-------
Table F-10. Self-Service Incidence Due To Basoiine Vauors (Maximum Likelihood Estiaate) - Theoretical
Year
1S86
1987
1988
1989
1998
1991
1992
1993
1994
1995
19%
1997
1998
1999
2089
2081
2082
2883
2024
2085
2086
2087
2088
2889
2818
2911
2312
2013
2014
£015
2016
2017
2318
2019
2820
SUM =
hPV =
Onboard
Incidence
Due To
Gas Vaoors
(ME, rat)
22.32
22.17
19.25
16.58
.14.37
12.38
10.56
9.00
7.38
6.15
5.08
4.11
3.27
2.51
1.34
1.84
1.84
1.34
1.84
1.84
1.34
1.84
1.84
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
194.40
115.32
St. II-N'ft*
(EX)
w/ Onboard
Incidence
Due To
Sas Vacors
(ME, rat)
22.39
20.49
17.69
15.18
13.16
11.27
9.68
8.25
6.76
5.54
4.66
3.78
3.81
2.31
1.70
1.76
1.33
1.84
1.84
1.84
1.84
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
132.45
137.28
St.II-Nfl*
(NO EX)
H/ Onboard
Incidence
Due To
Sas Vapors
, (ME, rat)
21.72
19.76
17.05,
14.63
12.68
10.86
9.33
7.95
5.52
5.44
4.50
3.64
2.93
2.23
1.65
1.73
1.82
1.84
1.84
1.84
1.84
1.84
1.84
1.34
1.34
1.84
1.34
1.84
1.84
1.84
1.84
1.84
1.84
1.84
1.34
177.51
103.98
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Gas Vaoors
(ME, rat)
20. 55
17.63
14.97
12.76
11.07
9.48
3.15
6.95
' 5.78
4.76
3.94
3.20
2.55
1.%
- 1.46
1.61
1.79
1.33
1.84
1.84
1.84
1.84
1.84
1.84
1.34
1.84
1.84
1.84
1.34
1.84
1.84
1.84
1.84
1.84
1.34
161.30
93.33
St.II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(ME, rat)
19.76
15.87
13.47
11.47
9.95
8.52
7.33
5.25
5.13
4.29
3.55
2.89
2.31
1.78
1.32
1.54
1.79
1.83
1.34
1.84
1.84
1.84
1.84
1.34
1.34
1.84
1.84
1.84
1.84
1.84
1.84
1.34
1.84
1.84
1.84
150.41
55.51
St.II-Nation
(EX)
M/ Onboard
Incidence
Due To
Sas Vapors
(ME, rat)
22.92
18.68
10.19
5.81
4.70
4.04
• 3.49
2.99
2.47
2.88
1.74
1.43,
1.16
0.92
3.71
0.71
3.96
1.47
1.79
1.84
1.84
1.84
1.84
1.84
1.34
1.34
1.34
1.34
1.84
1.84
1.84
1.84
1.84
1.84
1.84
117.78
59.53
St.II-Nation
(NO EX)
H/ Onboard
Incidence
Due To
Sas Vapors
(ME, rat)
22.92
17 17
* * • 4 /
S 29
U* ^J
1.2fl
4 • UW
0.77
0.68
0.61
3.55
0.48
0.43
w» ^w
3.38
8.34
0.31
0.27
0.25
3.25
0.60
1.34
1.78
1.84
1.84
1.84
1.84
1.84
1 34
*« l/T
1.84
1.84
1.84
1 S4
i« G^
1 84
A • \fT
1.34
1.34
1.34
.1.84
1.34
36.22
50.56
F-47
-------
Table F-I0. Self-Servics Incidence Due To Gasoline Vaoors (Maximum Likelihood Estisats) - Theoretical
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1995
1997
1998
1999
2003
2081
2002
2803
2884
2085
2806
2807
2008
2889
2318
2811
2012
2013
2814
2015
2016
£017
2018
2019
2020
-
s
Baseline
Incidence
Due To
Sas Vapors
(M.E,aice)
16.42
15.88
15.36
14.85
14.49
14.11
13.85
13.59
13.17
12.93
12.71
12.49
12.27
12.05
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
442.66
146. 12
St. II-Mfl*
(EX)
Incidence
Due To
6as Vapors
(MLE,aice)
15.83
14.68
14.11
13.59
13.26
12.32
12.68
12.44
12.06
11.84
11.64
11.43
11.23
11.03
10.83
18.83
10.83
10.83
10.33
10.83
18.83
10.83
10.83
10.33
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
485.19
134.71
St. II-Nfl*
(NO EX)
Incidence
Due To
Gas Vaoors
(MLE,aice)
15.56
14.15
13.68
13.18
12.78
12.45
12.22
11.39
11.62
11.40
11.21
11.02
10.82
10.63
10.44
18.44
10.44
10.44
18.44
10.44
10.44
18.44
10.44
10.44
10.44
10.44
13.44
10.44
10.44
10.44
18.44
13.44
10.44
10.44
18.44
391.68
138.12
St. II-Nfl
(EX)
Incidence
Due To
Sas Vapors
(MLE,iaice)
14.83
12.61
11.94
11.42
11.14
13.85
18.65
10.45
' 18. 13
9.95
3. 78
9.61
3.44'
9.27
9.10
9.18
9.10
9. 18
9.10
9.1.0
9.10
9.10
9.1.8
3.18
9.10
3.18
9.10
9.10
9.13
9.18
9.18
9.10
9.18
9.10
9.10
343.14
115.88
St. II-Nfl St.II-Nation St.II-Nation St.II-Nation
(NO EX) (EX) AP! (EX) . (NO EX)
Incidence
Due To
6as Vapors
(MLE,3ice)
14.16
11.37
18.74
18.26
18.81
9.75
3.57
9.39
9.13
3.93
8.78
8.63
8.48
8.32
8.17
8.17
8.17
8.17
8.17
3.17
8.17
8.17
3.17
3.17
8.17
8.17
3.17
3.17
8.17
8.17
3.17
3.17
8.17
8.17
8.17
389.11
104.21.
Incidence
Due To
3as Vaoors
(ME,aice)
16.42
13.38
3.12
5.17
4.53
4.57
4.43
4.40
4.27
4.19
4.12
4.85
3.38
3.31
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.33
3.83
3.83
3.83
3.33
3.83
3.83
3.83
3.83
3.83
155.27
68.80
Incidence
Due To
Sas Vaoors
(MLE,wice)
16.42
15.57
13. 18
3.13
6.16
5.62
5.14
4.53
4.27
4.13
4.12
4.85
3.58
3.31
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.33
3.83
3.83
3.33
3.63
3.83
3.33
3.33
3.33
3.33
3.33
3.33
3.83
188.93
58.15
Ineidarice
Due To
Gas VsDors
!MLE,;aic5)
IS. 42
12.33
C ••£«
1.11
3.71
0.S3
3. £3
8.57
8.55
0.54
3.63
8.61
3.60
8.59
8.53
3.58
3.58
'3.58
3.58
0.58
3.5B
3.53
8. 53'
2.58
2.53
8.58
2.53
3.58
3.58
8.58
g.53
8.58
3.53
8.53
8.58
53.56
37.i28
F-48
-------
Table F-10. Self-Service Incidence Due-To Gasoline Vapors (Maxisw Likelihood Estiaate) - Theoretical
Year
1985
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
SUM =
NPV =
Onboard
Incidence
Due To
Sas Vaoors
(HE, Bice)
16.42
15.88
13.79
11.88
10.30
8.81
• 7.57
6.45
5.28
4.41
3.64
2.95
2.34
1.80
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
139.27
82.62
St.II-Nft*
(EX)
H/ Onboard
Incidence
Due To
Sas Vapors
(MLE, sice)
15.83
14.68
12.67
10.88
9.43
8.07
6.93
5.91
4.34
4.04
3.34
2.71
2.15
1.65
1.22
1.25
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
130.71
76.85
St.II-Nft*
(NO EX)
»/ Onboard
Incidence
Due To
Sas Vaoors
(«LE,aice)
15.55
14.16
12.21
10.48
9.09
7.78
5.68
5.70
4.67
3.90
3.22
2.61
2.08
1.60
1.18
1.24
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
127.24
74.49
Sf.II-Nfl
(EX)
«/ Onboard
Incidence
Due To
Sas Vapors
(MLE. slice!
14.80
12.51
10.73
9.14
7.93
6.79
5.84
4.98
' 4.08
3.41
2.82
2.29
1.83
1.41
1.04
1.16
1.29
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
115.91
56.90
St. II-Nft
(W EX)
«/ Onboard
Incidence
Due To
Sas Vapors
fflLE.aice)
14. 16
11.37
9.65
8.22
7.13
5.10
5.25
4.48
3.68
3.07
2.55
2.07
1.65
1.28
0.95
1.11
1.28
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
107.76
51.33
St. II-Nation
- (EX)
K/' Onboard
Incidence
Due To
Sas Vapors
(MLE. sice)
16. 42
12.38
7.33
4.15
3.37
2.89
2.50
2.14
1.77
1.49
1.25
1.83
0.83
0.56
0.51
0.51
0.69
1.06
1.28
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
84.38
49. 81
St. II-Nation
(NO EX)
w/ Onboard
incidence
Due To
Sas Vapors
iME.snce)
15.42
:5. 38
4.51
3.3£
3.55
8.49
3.44
3.39
2.34
3.31
0.27
0.25
0.22
3.20
0.18
9.18
0.43
2.96
1.27
1.32
1.32
1.32-
1.32
1.32
1.32.
1.32
1.32
1.32
1.32
1.32
L32
1.32
1.32
i.32
1.32
61.77
36.22
F-49
-------
Table F-ll. Self-Service Incidence Due To EDB And EDC - Thaoreticai
Year
Baseline St.II-NA*
(EX)
Incidence
Due To
ma
Incidence
Due To
EEB
(Theoretical) (Theoretical)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2900
2001
2982
2993
2994
2895
2906
2097
2999
2919
£911
£912
2913
2914
2915
2916
£917
2918
2919
2829
SUM =
NPV =
9.9296
9.8253
9.0228
0.0199
0.8172
9.9152
0.0128
8.0185
0.0988
0.0070
0.0058
0.0947
9.0935
9.0923
9.9912
9.0912
0.0912
0.8012
0.0012
9.0012
9.0812
0.0012
0.0912
9.8912
0.9912
0.9912
0.0912
0.9912
0.8312
9.3912
9.8012
9.8912
9.3812
0.0912
0.8812
8.218
8.136
3.3284
9.0238
3.0209
0.0181
0.0157
0.0139
9.0117
0.9996
0.8880
9.0864
9.0053
0.9843
0.8832
0.9921
9.0011
0.8911
9.9811
9.0011
0.8911
9.0911
0.8011
8.0911
0.0911
0.0911
9.8911
0.0911
9.0311
9.8911
0.8911
9.8811
9.8311
0.8011
9.8911
0.3911
0.8311
8.194
3.125
St.IHW*
(NO EX)
Incidence
Due To
EDB
(Theoretical)
3.8279
0.0223
9.0281
0.9174
8.9151
3.9133
8.0113
3.8992
9.8977
8.8952
0.8051
0.8041
0.8031
8.8921
8.8910
9.0910
9.8919
9.8919
8.8918
8.0819
9.8810
8.8810
0.8919
9.8910
9.8919
0. '3819
8.8818
3.8810
0.3818
8.0919
9.8810
3.8819
8.8818
9.8919
8.9318
9.187
8.1S2
St.II-Nfl
(EX)
Incidence
Due To
EDB
(Theoretical)
3.8265
9.8293
3.3176
8.9151
8.8131
3.3116
0.8998
3.8888
• 8.8367
3.8353
0.8045
3.8836
0.0827
8.8818
8.8889
3.3039
8.3039
3. £009
8. (3889
3.3889
0.0089
8.8039
0.8803
8.8389
8. (500-3
8.3S99
8.3389
3.8889
3.3889
3.3889
•3.8389
8.8809
9. 8889
3, 0889
0.88(59
3. 165
8.189
St.II-Nfi ot.II-Nation :
!KO EX) (EX)
Incidence
Due To
ESB
(Theoretical)
8.3E54
•3.8183
8.0157
3.3135
8.3117
3.3133
3.8338.
3.3372
9.3868
3.8948
9.3040
3.2332
8.8924
3.0316
8.3988
8.8888
'8.8998.
2.3388
8.0088
3. -2838
3.2369
3. 3368
0. 8883
•8. 3338
3.8388
•3. 8388
8.8828
3.3888
3.3868
8. £388
8.8888
3. mm
9.2088
3. S088
3.0338
8. 149
8.399
Incidence
D-iS "o
E"3
(Theoretical)
•3. 3235
3.8216
8.3117
0.3365
9.8852
3.8846
8. £233
3.3832
3.88£7
3. 2821
0.M13
8.2314
0.8811
3.3837
. 8.9884
3. 3834
8.0884
3.2234
8.3804
8. 2K4
9.0F34
9. 3834
9.2834
3.3804
9.8804
3. C034
3.3834
3.C034
3.3884
8.8884
3.3884
8.23S4
3.8884
3. 2824
8.3884
3. i'33
9.575
St.II-Nsiiori
Ir.cicsr-.c9
Due ":•
EJB
(Tnsorsvlcal)
:.329&
3. C2£5
3.C134
8.3113
3.387®
8. 2858
8. 30i5
3.S323
2. 3827
- 3.3021
3. 3316
8. 8814
3.0811
9. 3387
3.8884
3.2.824
2. -2804
8. 3834
9. 8934
8. 3334
3. 3034
3. 3084
3. 3884
3.3824
3. C924
3. -3884
3. 3804
3. B334
•8. -3334
3. 3084
3. 3834
3. 8834'
8. 3934
8. £384
8. 8884
8.124
8.1335
F-50
-------
Taois F-li. Seif-Ssrvice Incidence Due To EDB find EDC - Theoretical
Year
1986
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2888
2081
£802
2003
2004
2005
2006
2087
2888
2889
2818
2011
2812
2013
2814
2315
2016
2017
2818
2319
2028
SUM =
NPV =
St.II-Nation
(NO EX)
Incidence
Due To
EDB
(Theoretical)
3.82%
0.3198
8.8869
0.3809
0.8083
3.8083
0.8883
0.0802
3.8082
8.0801
0.8001
0.8881
0.8881
.8088
.8888
.0088
.8888
.0008
.0088
.8808
.8888
.8080
.0088
.8000
.8888
.3338
.3888
.3888
.8880
.8388
.0088
.8888
.8883
.8888
.8888
0.859
8.855
Onboard
Incidence
Due To
EDB
(Theoretical)
8.8296
0.3258
8.8228
8.8199
8.0172
0.3152
3.8128
0.3185
8.8888
0.8870
0.8858
8.8047
8.8835
0.8823
8.8812
0.8812
8. 8812
0.0012
8.8812
8.8012
8.8812
0.8812
3.8812
0.8812
8.8812
8.3312
0.8812
3.8812
8.8812
8.3312
8.8012
8.8012
8.8812
0.3312
0.3812
3.218
3. 135
St.II-Nfl*
(EX)
w/ Onboard
Incidence
Due To
EDB
(Theoretical)
8.8234
3.8238
3.8289
3.3181
3.8157
3.0139
3.3117
3.38%
3.3880
8. £064
3.8853
8.8843
3.8332
8.3821
3.8811
8.8811
8.8812
8.8812
3.8812
8.8312
3.8812
3.8812
3.8812
8.3312
3.8812
8.0312
3.3812
3.3012
3.3812
3.8312
8, 3812
3.8012
-3,3812
3. 3312
3.3012
8.195
3. 125
St. II-Nfl*
(NO EX)
w/ Onboard
Incidence
Due To
EDB
(Theoretical)
8.3279
3.8229
8.8281
8.3174
0.3151
3.0133
3.3113
8.8892
' 3.3877
8.0062
8.8851
3.3841
3.3831
0.8821
3.3810
3.3311
8. S812
0.8012
3.3312
0.0812
3.8312
3.0812
. " 3. 3812
0.8812
3.2312
3.8012
3.5312
3.8012
8.3012
0.8812
8.8312
8.8812
3.J312
3. 8812
3.8012
0. 198
3.122
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
EDB
(Theoretical)
3.S255
3. 8283
3.3176
3.3151
3. 3131
8.3116
3.3393-
0. 8888
3. «67
8. 3853
3.8945
3.3835
3.3827
0.3813
0.3889
0. 8818
3. Mil
8.3012
3.3812
0.8012
3. 3812
3.3812
3.8812
3.3012
3.3312
3. 8812
3.3312
•2. 3812
3. 3812
3.3812
3.8312
3.8812
3.3812
3. 3312
3.3912
3.17!
3.139
St.II-Nfl
(NO EX)
w/ jncoara
Incidence
Due To
EDS
(Theoretical i
0.3254
3.3133
3.3157
3.3135
3.3117
0.3133
8. 2833
0.JS7E
0. 3858
0.5048
3. 8348
3. 3332
3. 3824
3.3S215
3.3888
8.3813
8.2811
8.3812
3.3312
3.881E
.3. S3 12
3.S012
3.312
3. £812
3.C312
2. C812
3.8012
3.3812
3.S312
3. £312
3,8812
3.2812
3.3312
2, £3:2
8.2812
3.157
.*.m
St.II-Nation
!EX!
w/ Onboard
Incidence
Due "To
3B
(Theoretical)
*.&%
3.3117
3. 5855
3.335E
8. 2333
3. 2832
0.3027
' 3.3821
3.3818
3.301*
8.3311
3.3887
3.3384
3. 3834
3.3885
3. 3889
3.3311
3. 3812
3. 2012
5.3312
3.3012
3.C312
3. t31?
2. "31?
3. C312
C» ••Jdl.L
3.5-312
T< *ii i J-:
<-« v'oiC
•i *•*«•••
V
-------
Table F-ll. Self-Service Incidence Due To EBB Qnd EDC - ineorsticai
ar
198&
1987
1988
1389
1930
1991
1992
1393
1994
1995
19%
1397
1998
1999
2903
2081
2$02
2803
2084
2995
2006
2007
2828
2809
2919
2911
2012
2313
2914
2915
2016
2917
2918
2919
2029
St.II-Nation
(NO EX)
H/ Onboard
Incidence
Due To
EDB
(Theoretical)
0. '32955
0.91982
0.00595
9.90092
0.80334
0.00039
9. 30925
8.99921
0.90817
9.00814
9.09012
0.99909
0.90307
0.00905
0.99902
0.00002
0.90928
0.00081
0.00112
0.09117
9.90117
8.90117
0.90117
9.00117
9.30117
0.00117
0.30117
0.00117
0.99117
8.80117
8.00117
0.09117
3.30117
0.00117
8.30117
Baseline
Incidence
Due To
EDC
(Theoretical)
8.3381
0.0333
3.3294
0.0256
0.3221
0.9196
3.3166
8.0136
3.3113
0.3890
8.3375
9.3060
8.3345
9.0033
8.3015
0.3015
3.3315
3.3015
3.3815
9.3915
8.3015
8.0315
0.0015
8.0815
3.8815
3.3815
8.3015
8.0915
8.3815
8.3315
3.3815
9.0815
3. '3815
8.8815
9.6315
St.IHM*
(EX)
Incidence
Due To
EBC
(Theoretical)
3. 3367
8.0387
9.3269
8.8234
3.0232
3.3179
3.3151
3.8124
3.8133
8.3833
3.8869
8.3855
8.3841
9.302B
9.3014
8.3814
0.3014
3.0814
8.3314
0.3814
8.3314
9.0014
3.3014
3.8014
9.3814
3.3814
3.3014
3.8814
0.'8014
3.0314
9.S814
0.8914
0. 3014
3. 6014
9.8314
St. II-Nfl*
(NO EX)
Incidence
Due To
EDC
(Theoretical)
3.3351
3.3296
0. 3259
3.8225
8.3195
8.3172
3. 8146
3.3119
3.3899
0.3879
3. 3066
3.5053
8.3840
8.3326
3.3313
9. £013
8.3813
3.3813
3.3813
3.3913
3.3313
3.3813
3.3013
3.3313
3.3313
8.2313
8.8313
0.0813
3.8013
3.13813
3. '2013
9.3813
3. 3013
3.0313
3.3813
St. II-N'fi
(EX)
Incidence
Due To
EDC
(Theoretical)
3.0342
3.3262
0.822S
3.31S5
3.3169
8.3149
8. 8125
0.3103
3. 2066
9.8369
8. 3357
3.3046
3. £334
3. §323
9.3311
8.3811
3. 2311
0.12811
0.3311
0.2811
3.3311
9.C311
8.3311
3.331;
3.3311
3.0311
3.3311
0,3811
8.3-311
3.3311
3.5311
8.S011
8.0311
9.3311
3.8311
3.S327
3.J23S
•3.0293
3.3174
3.3151
•3.3133
3.8113
3.2092
3.2-377
3.3362
0.8351
3.S041
3.3031
2.3021
3.8310
3. '2318
3.8313
3.3813
3. m 13
0.3013
3.3818
3.3310
3.3313
8.6310
3.3312
3.0910
8.3012!
8. S318
s.;
0.980
3.957
8.271
8.176
8. £58
3.163
3.241
3.157
0.213
3.143
8.8310
9.193
3.127
F-52
-------
Table F-ii. Self-Service Incidence Due To EDE And EDC - Theoretics!
Year
1985
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2800
2001
2802
£003
2004
2805
2006
2007
2008
2009
2010
£811
2012
2013
£014
2015
2016
2017
2018
2019
2020
SUM =
NPV =
St.II-Nation
(EX)
Incidence
Due To
EDC
(Theoretical)
0.0381
9.0279
0.0151
0.0084
0.0057
0.3059
•0.0050
0.0041
0.0034
0.8027
0.0023
0.0818
0.0014
0.0009
0.0005
0.0005
0.0005
0.0085
0.0005
0.0005
0.0005
0.8885
' 0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.8085
0.0005
3.S005
0.0005
0.133
0.102
St.II-Nation
API (EX)
Incidence
Due To
EDC
(Theoretical)
0.0381
0.0328
0.0251
0.8154
0.0099
3.®74
9.0058
0.3843
0.0034
0.3027
0.0023
0.0818
0.0014
0.0009
0.0005
0.0005
0.12085
0.0005
0.0005
0.0005
0.0005
0.0005
0.0085
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0305
0.8805
0.0005
0.3005
3.0005
0.0305
0.1£0
0.123
St.II-Nation'
(NO EX)
Incidence
Due To
EDC
(Theoretical)
3.03812
0.82555
3.30895
0.80118
3.30044
0.33339
3.03033
0.38827
0.33322
0.00018
0.38815
3. 03012
0.03089
8.30006
3.83333
8.30803
0.00093
3.00003
0.00083
3.00003
8.38833
0. 83883
0.80803
8.08003
0.08083
3.00003
8.88833
8.30003
0.38833
8.08803
8.03003
8.03803
8.00803
3.89883
0.08003
3.377
0.871
Onboard
Incidence
Due To
EDC
(Theoretical)
0.3381
8.8333
3.0294
3. '3255
0.3221
3.3135
3.8155
8.3136
0.0113
8.0890
0.8375
0.3860
8.3845
0.3038
8.0015
. 3.3315
9.3015
8.3815
0.0815
8.8815
8.8815
9.8015
8.0015
3.8015
3.3015
3.0315
0.3815
3.8815
3.0315
0.@815
3.3015
3.8815
8.8815
3.S815
0.3015
3.271
3. 176
St.II-Nfl*
(EX)
w/ Onboard
Incidence
Sue To
EDC
(Theoretical)
3.3367
3.3337
0.8269
0.3234
3.3202
0.3173
3.0151.
3.3124
3.8103
3.3383
8.8059
8.8855
8.3841
8.2328
3. 3814
3.2314
9. 8015
0.2015
3.3815
3.esi5
0.3815
0.3315
3.3015
3. -3815
3.3015
«. 88 15
3.3315
8.3815
0.2315
8.8315
3.0015
3.2315
0.8815
8.8815
3.0015
3.253
8.163
St.II-Nfl*
(NO EX)
w/ Onboard
Incidence
DUB To
EDC
(Theoretical)
2.3351
3.3296
0.0253
3.3225
8.3155
•3.3172
0.8146
8.3119
3.0099
8. 8879
3.0865
8.3053
0.0343
8.2826
8.6313
0.2314
8.M15
3.3315
• 3.2015
3.3315
3.3015
8.2815
0.3315
3.3315
3.3815
3.8315
3.0815
3.2315
8.8815
3.0315
3.8015
8.3315
0.8815
3.3815
3.8315 .
8.245
3. 158
St.II-NS
(EX)
w/ Onboard
fncic'pncs
Due To
EDC
(Theoretical)
3. 3342
3.3262
3.322S.
3.3195
3.3159
3. 3149
8. 3126
3.3103
3. 2986
8. 8863
3.8057
3.3045
3. 3034
3.3323
0.3911
0.8313
8.3315
3.3315
8. 3015
8.S315
3. 8015
8.8315
3,8315
3. £315
6. 2815
3. -3315
3. 3315
0.3315
3. S3 15
8. 8815
3.3815
3.3315
3. 3015
8.3015
3. 301E
0.223
& 141
F-53
-------
Table F-il. Self-service Incidence Due To EDB ftrid EDC - Theoretical
Year
St.II-Nfl St.II-Nation St.II-Nation
1SBS
1987
1988
1989
1950
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2881
2802
2802
2834
2885
280S
2887
2883
2839
2818
2811
2812
2813
2814
2815
2816
2817
£818
2919
£828
SUM-
m *
(NO EX)
H/ Onboard
Incidence
Due To
EDC
(Theoretical)
8.3327
8.8235
0.8283
8.8174
0.0151
8.0133
0.3113
8.0892
0.6877
0.0862
0.0051
0.0841
0.0831
0.0021
0.0810
0.8012
0.0015
0.0015
8.0015
0.8815
8.8015
0.0015
8.0015
0.0015
0.0015
0.0815
8.3015
0.0815
0.0815
0.0015
0.0015
0.0815
0.0815
0.8915
8.8815
8.282
3.128
(EX)
w/ Onboard
Incidence
Due To
EDC
(Theoretical)
8.0381
0.8279
0.0151
0.8884
3.8867
0.8859
0.8058
8.8041
0.0034
3.8027
8.0023
0.8018
0.0314
0.8089
0.8005
0.0805
0.8007
0.0812
0.0815
0.0315
0.0315
0.8015
0.0315
0.0815
0.8815
0.8315
0.0315
8.0315
3.8815
3.3315
0.8015
0.3315
3.3315
8.8815
8.3815
8.152
3. 133
(NO EX)
«/ Orboard
Incidence
Due To
EDC
(Theoretical)
3.33312
3.02555
9.38896
8.38113
3.03844
0.80039
3.33033
3.8C327
8.28822
0.38018
8.33815
0.83312
3.03339
3.83385
3.39303
3.03333
8.38336
0.38184
3.33145
0.33151
3.30151
0.88151
3.33151
3.03151
9.W151
0.03151
8.23151
3.83151
3.33151
0.09151
8.30151
3.S3151
3.130151
3.08151
3.33151
0.133
3.874
F-54
-------
Table F-12. Service Station Incidence Due To Benzene find 3as Vaoors - in-usa, finnuai Insoections
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2900
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2013
2019
2020
SU« =
i*V =
Baseline
(Ann. Inso. )
Total
Incidence Due
To Benzene
3.2316
0.2255
0.2193
0.2131
0.2090
0.2045
3.2018
0. 1991
0. 1937
0. 1910
0.1883
0.1856
0.1829
0.1803
0. 1776
0. 1776
0. 1776
0.1776
0. 1776
0.1776
0. 1776
0.1776
0. 1776
0.1775
0. 1776
0.1775
0. 1775
3.1776
3. 1776
0.1776
0. 1775
3. 1776
0. 1775
0.1776
0. 1776
6.55
2.13
St. I-Nation
(EX)
(finn. Inso. )
Total '
Incidence Due
To Benzene
0.2316
0.2198
3.2028
0. 1917
0. 1881
0.1840
3. 1815
0. 1791
3. 1743
0. 1719
0.1594
0. 1670
0.1646
0.1622
0.1598
0.1598
3.1598
0. 1598
0.1598
0. 1598
0.1598
0.1598
0. 1598
0. 1598
3. 1598
0. 1598
3. 1598
0. 1598
3. 1598
3. 1598
3.1598
8. 1598
3. 1598
0. 1598
3. 1598
5.94
1.36
St.II-NA*
(EX)
(Ann. Inso. )
Total
Incidence Due
To Benzene
3.2153
3. 1924
3. 1845
0. 1781
0. 1747
0.1709
0. 1586
0.1664
0.1619
8. 1596
0.1574
0. 1551
8*1529
0.1506
0. 1484
0.1484
0. 1484
0. 1484
0.1484
3. 1484
0. 1484
0.1484
0.1484
0.1484
0.1484
3.1484
3. 1434
3.1484
0.1484
0.1484
0. 1434
3.1484
3. 1484
3.1484
0. 1484
5.53
1.33
St.II-Nfl
(EX)
(Ann. InsD. )
Total
Incidence Due
To Benzare
3. 1983
3. 1579
3.1483
3. 1416
0. 1389
0. 1359
3. 1341
3. 1323
3. 1287
3.1259
8. 1251
3. 1234
0.1216
0. 1193
0.1180
0.1180
8. 1130
3. 1180
3. 1130
8.1188
3. 1180
3.1180
0.1180
3.1183
0. 1180
3. 1180
3.1130
3. 11S3
0. 1133
3.1180
3. 1180
3.1188 -
3.1183
3. 1183
3. 1130
4.41
1.47
St.II-Nation
(EX)
(Ann. Inso. )
Total
Incidence Due
To Benzene
3.2315
3.2037
3. 1553
0. 1288
3.1225
3. 1198
0.1182
0.1167
8. 1135
3.1119
8.1134
0.1838
3. 1372
0. 1356
3.1840
8.1340
3. 1340
3.1340
3. 1340
0. 1043
3. 1040
8.1340
0. 1048
3. 1340
3. 1343
3. 1343
3. 1340
3. 1343
3.1040
3. 1348
3.1343
0. 1343
3. 1340
0.1343
-3. 1948
4.24
1.43
St.II-Nation
(NO EX)
(firm. InsD. )
Total
Incidence Due
3.231S
0. 1323
3. 121S
3.8831
3. 3780
3.3763
3.3753
3. 8743
3.3723
0.3713
3. 3733
3.3693
0. 3583
0.3673
3.3563
8.3653
3. 3663
3.0653
3.2663
2.3563
3.3663
%. 3563
3.0653
3.3663
3.3563
3. 2653
3. (3663
8.3663
iV3653
3.3563
3. 3663
3. £663
3. '3553
-0.0563
3.S653
£.74
1.37
Onboard-Nation
(w/ Tasncer. )
(Ann. Inso. )
Incidence Due
To Bsnzsr.a
6. 2316
3.2339
0. 1S4£
2. 1537
2. 1537
8 '317
VI iVj! :
3. 1288
3. 11 IS
3.3977
v* &Jt. 3
0.0655
3. & 13
8. 2813
8.3818
8.2810
3.2318
3. 2318
3. £815
3.2313
2. 2313
3. -:-3ia
3.3813
3.2813
8 2315
3.3318
3.0818
3.3313
3.2313
•3.0818
8.3813
3. 3818
3.77
1.54
F-55
-------
Table F-12. Service Station Ir.cider.ee Due To Benzene find 3as Vanors - In-use. Rnnuai Insoections
Yiar
1586
1987
1588
1989
1930
1991
1592
1993
1994
1995
19%
1997
1998
1999
2000
2001
2003
2084
2085
2006
2007
2008
2009
£010
2011
2012
2013
2014
2015
2015
2017
2018
2019
2020
SUH*
NPV =
Baseline
(Ann. Inso. )
Total
Incidence Due
To Gas Vapors
(PH., rat)
5.842
6.659
5.476
6.293
6.174
6.^39
5.950
5.880
5.721
5.642
5.552
5.483
5.403
5.324
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
St. I-Nation
(EX)
(flnn. Inso. )
Total
Incidence Due
To Sas Vapors
(PUL,rat)
6.842
6.480
5.954
5.617
5.511
5.390
5.319
5.248
5.107
5.036
4.365
4.894
4.823
4.752
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.581
4.681
4.681
4.681
4.581
St.II-Nfi«
(EX)
(flnn. Insn. )
Total
Incidence Due
To 3as Vapors
(PUL,rat)
6.358
5.780
5.469
5.278
5.178
5.365
4.998
4.332
'4.7S9
4.732
4.665
4.599
4.532
4.455
4.399
4.339
4.399
4.399
4.393
4.399
4.393
4.399
4.333
4.399
4.333
4.339
4.333
4.399
4.393
4.339
4.333
4.339
4.393
4.399
4.399
St.il-Nfl
(EX)
(flnn. Insp. )
Total
Incidence Due
To Gas Vapors
(PUL,rat>
5.859
4.589
4.408
4.208
4.129
4.838
3.985
3.932
3.825
3.773
3.719
3.666
3.613
3.553
3.507
3.537
, 3.507
3.587
3.507
3.537
3.507
3.507
3.507
3.507
3.507
3.5S7
3.507
3.507
3.507
3.537
3.507
3.537
3.507
3.507
3.507
St.II-Nation
(EX)
(flnn. Insp. )
Total
Ircidence Due
To Sas Vapors
OIL, rat)
6. 842
5.333
4.645
3.833
3.573
3.590
3.543
3.495
3.401
3.354
3.306
3.259
3.212
3.155
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
'3.117
3.117
3.117
St.II-Nation
(NO EX!
(Ann. Insp. )
Total '
Incidence Due
To Gas Vancrs
(PUL,rat)
£.342
5.535
3.S44
2.517
2.266
2.315
£.284
2.254
2.193
"2.162
£.13£
2.101
2.071
2.341
2.818
2.013
2.013
2.313
2.81-3
2.013
2.013
2.313
2.313
2.013
£.310
£.818
2.313
2.910
£.310
2.313
£.313
2.018
2. 313
2.218
£.013
193.59
62.65
174.24
57.47
153.15
53.48
131.35
43.-60
120.81
42.75
82.82
32.20
F-56
-------
Table F-12. Service Station Incidence Due To Benzene find Sas Vapors - In-use, Annual Inspections
Year
1985
1987
1988
1969
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
£000
2802
£003
2004
d805
2086
2887
2888
2809
2810
£311
£812
2813
2814
2815
2816
2017
2018
2019
£828
SUH =
M3V =
Onboard-Nation
(w/ Tamper. )
(flnn. Insp. )
Total
Incidence Due
To Sas Vaoors
(PH., rat)
6.842
6.659
6.834
5.464
5.814
4.579
' 4.243
3.943
3.600
3.358
3.142
2.952
2.783
2.625
2.488
2.488
2.488
£.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
113.48
45.98
Baseline
(flnn. Insp. )
Total
Incidence Due
To Gas Vapors
(PUMice)
4.15
4.84
3.93
3.82
3.74
3.66
3.61
3.56
3.47
3.42
3.37
3.32
3.28
3.23
3.18
3.18
3.18
3.18
3.18
3.18
3. 18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
117.36
38.19
St.l-Nation
- (EX)
(flnn. Insp. )
Total
Incidence Due
To Sas Vapors
(PUL.raice)
4.15
3.93
3.51
3.41
3.34
3.27
3.££
3.18
' 3.10
3.05
3.31
2.97
2.92
2.88
2.84
2.84
2.84
2*84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
185.63
34.34
St. II-Nfl*
(EX)
(flnn. Insp. )
Total
Incidence Due
To Eas Vapors
(PUL.Hice)
3.66
3.46'
3.32
3.28
3.14
3.07
3.33
2.99
2.91
2.87
£.83
2.79
2.75
2.71
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.57
2.57
2.67
2.67
2.67
2.67
2.57
2.57
2.67
2.S7
2.67
2.67
98.91
32.4?
St. II-Nfl
SEX)
(flnn. In=c. )
Total
Incidence Due
To Sas Vapors
(PL'L, mice)
3.55
2.S4
2.67
2.55
2.53
2.45
2.42
2.38
2.32
2.29
2.25
2.22
2.39
2.16
2.13
2.13
2.13
2.13
2.13
2.13
2.13
£.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
79.45
?A i?
St. II-Nation
(EX)
(finr>. Insp. )
Incidence Due
To Gas Vapors
iPULjMice)
4.15
- ff
u. uu
2. 2°
2.22
2 IS
I~« Aw
2.15
2.1£
2.86
" 2 SR
l»4 OO
£.88
i 9S
1* JO
1 95
*» Jw
1.92
1.89
i.89
1 flq
*. OJ
1.39
1m QJ
1.89
1.89
i 89
A« \JJ
i an
A« U-/
i 39
*• \JJ
< aq
*• Q J
1 flQ
*» 33
i.83
1 39
A* w J
i aq
A* U.7
i.SS
. 1.89
73.24
oe; n<
25.91
F-57
-------
Table F-12. Service Station Incidence Due To Benzene find Gas Vapors - In-use. flnnual Insoections
Year
1986
1987
1988
1989
1999
1991
1992
1993
1994
1995
19%
1997
1998
1999
20®
2091
2902
2083
2884
2835
20%
2807
2808
2809
2818
2811
£812
2913
2814
2815
2816
2817
2818
2819
2828
SUMs
NPV =
St.II-Nation Onboard-Nation
(NO EX)
(Ann. Inso. )
Total
Incidence Due
To Gas Vapors
(PUL,aice)
4.15
3.45
2.21
1.53
1.43
1.48
1.38
1.37
1.33
1.31
1.29
1.27
1.26
1.24
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.2E
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
58.21
19.52
(H/ Tastier.)
(firm. Inso. )
Total
Incidence Due
To Gas Vapors
(PUL,aice)
4.15
4.84
3.66
3.31
3.84
2.78
2.57
2.39
2.18
2.64
1.98
1.79
1.69
1.59
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51 "
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
58.79
27.83
Baseline
(flnn. Insp. >
Total
Incidence Due
To Sas Vaoors
(?LE,rat>
3.895
3.792
3.687
3.583
3.516
3.439
3.393
3.348
'3.258
3.212
3. 167
3.122
3.877
3.831
2.986
2.986
2.986
2.986
2.986
2.986
2.986
2.986
2.385
2.986
2.985
£.986
2.986
2.586
2.985
2.986
2.986
2.986
2.985
2.986
2.986
113.23
35.79
St. I-Nation
(EX)
fftnn. Inso. )
Total
Incidence Due
To Gas Vapors
(ME, rat)
3.896
3.598
3.398
3.198
3.138
3.069
3.029
2.988
2.908
2.367
2.827
2.787
2.745
2.786
2.665
2.655
2.665
2.665
2.665
2.655
2.555
2.665
2.665
£.665
2.665
2.665
2.665
2.665
2.565
2.665
2.565
2.665
2.665
2.665
2.565
99.21
32.72
St. I MB*
(EX)
(flnn. Inso. )
Total'
Incidence Due
To Gas Vapors
OLE, rat)
3.525
3.246
3.114
3.3K
2.949
2.884
2.346
2.808
2.732
2.594
2.655
2.618
2.581
2.543
2.585
2.535
2.505
2.585
2.505
2.535
2.583
2.585
2.585
2.585
2.595
2.585
2.505
2.585
2.585
2.585
2. 585
2.585
£.585
2.535
£.585
92.93
38.45
St.II-Nft
(EX)
(flnn. Insp. )
Total
Incidence Due
To Sas Vaoorc
(MLE,rat)
3.342
2.578
2.513
2.396
2.351
2.299
2.269
2.239
2. 178
'2.148
2. US
2.088
2.857
2.027
1.397
1.997
1.997
1.997
1.997
1.997
1.937
1.997
1.937
1.997
i.937
i.397
i.937
1.397
1.997
1.997
1.997
1.997
1.397
1.397
1.957
74.53
24.33
F-58
-------
Table F-i£. Service Station Incidence Due To Benzene find Sas Vaoors - In-use, Annual Insosctions
Year
1935
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2088
2001
2802
2083
2004
2005
20%
2007
2008
2009
2818
2011
2012
2813
2014
2015
2016
2017
2318
S019
2020
SUM =
!*V =
St. Il-Nation
(EX)
(flnn. InsD. )
Total
Incidence Due
To Sas Vapors
(ME, rat)
3.396
3.433
2.S45
2.182
2.090
2.044
2.817
1.990
1.935
1.910
1.383
1.855
1.829
1.802
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
68.79
24.34
St.II-Nation
(NO EX)
(flnn. Inso. )
Total
Incidence Due
To Sas vapors
(ME, rat)
3.395
3.237
2.875
1.433
1.347
1.318
1.301
1.283
1.249
1.231
1.214
1.197
1.179
1.162
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
47.15
18.33
Onboard-Nation
(w/ Taaper. >
(flnn, Inso. )
Total
Incidence Due
To Sas Vapors
(ME, rat 5
3.395
3.792
3.435
3.111
2.855
2.507
2.415
2.245
2.858
1.912
1.789
1.581
1.585
1.495
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
54.62
26. 14
Baseline
(flnn. Inso. )
Total
Incidence Due
To Bas Vacors
(ME, nice)
2.73
2.72
2.64
2.57
2.52
2.45
2.43
2.49
2.33
2.30
2.27
2.24
2.20
2.17
2.14
2.14
2. 14
2.14
2.14
2.14
2.14
2.14
£.14
2.14
£.14
2.14
2.14
2.14
2.14
2.14
2.14 .
2.14
2.14
2.14
2. 14
78.97
£5.64
St.I-Nation
(EX)
(flnn. Ir.so. }
Total
Incidence Due
To Gas Vapors
(ME. fliics}
2.79
2.64
£.43
2.29'
2.25
2.29
2.17
2.14
2.128
2.35
2.83
2.30
1.97
1.94
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
LSI
1.91
1.31
1.91
1.91
1.91
- 1.91
1.91
1.91
1.91
71.88
£3.44
St. II-Nfl*
(EX)
(flnn. InsD. )
Total
Incidence Due
To Sas Vapors
(ME. a ice)
2.68
£.33
£.23
£.15
2.11
£.37
2.04
£.31
•1.95
" 1.93
1.90
1.S8
1.85
1.82
1.79
1.79
1.79
1.79
1.79
1.79
1.79
1. 79
1.79
1.79
1.79
1.79
1.79
1.73
1.79
1.79
1.79
1.73
1.79
1.73
1.73
66. 5&
£1.82
F-59
-------
Table F-12. Service Station Incidence Due To Benzene find Gas Vaoors - Inruse. fir.nual inspections
ar
1986
1987
1988
1989
1923
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
mi
2302
£833
2384
2005
2006
2007
2008
2039
2010
2011
2812
2013
2014
2015
2016
2017
2018
2019
2020
St.II-Nfl
(EX)
(flnn. Iriso. )
Total
Incidence Due
To 5as Vaoors
(!
-------
Table F-13. Service Station Incidence Cue To Benzene find Sas Vaoors - in-use, No Inscections
/ear
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2200
2301
2002
2003
2004
2005
2336
£307
2338
am
2313
2011
2212
2S13
2314
2315
2216
3317
2S18
2319
£S23
Baseline
(flnn. Inso. )
Total
Incidence Due
To Benzene
3.2315
0.2255
0.2193
0.2131
0.2096
3.2045
•0.2018
0. 1991
8. 1937
0. 1910
0.1883
0.1856
0.1329
0.1803
0. 1776
0. 1776
0. 1776
0.1775
0.1776
0. 1775
8. 1776
3. 1776
0. 1776
0. 1775
3. 1775
3. 1775
3.1775
0. 1775
0.1776
3. 1775
3, 1775
3. 1775
3. 1775
3. 1776
e.1775
St.II-NA*
EX)
(No Inso.)
Total
Incidence Due
To Benzene
3.2247
0.2113
0.2044
0. 1981
0.1944
0. 1901
0. 1876
0. 1851
0.1801
0. 1776
0.1751
0. 1726
0.1701
0. 1675
0.1651
0. 1651
3. 1651
0. 1651
3. 1651
3.1651
3.1651
3. 1551
3. 1651
8. 1651
3.1551
0. 1551
3. 1651
3. 1551
3. 1651
3. 1651
3. 1651
3. 1651
0. 1651
3. 1S51
3. 1551
St.II-NA '
(EX)
(No Insp.)
Total
Incidence Due
To Benzene
8.2123
3.1363
0.1782
0. 1717
0.1684
0. 1647
0.1626
8.1604
0. 1561
0.1539
0.1517
0.1496
0.1474
0.1452
0.1431
0. 1431
0.1431
0.1431
8. 1431
3. 1431
0.1431
0. 1431
0.1431
3. 1431
0. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
8. 1431
St. II-Nation
(EX)
(No Insp.)
Total
Incidence Due
To Benzene
0.2316
3.2128
0. 1825
0.1638
3.1589
3. 1554
. 3. 1534
0. 1513
0.1472
0.1452
0.1431
8. 1411
0.1390
0. 1370
0.1350
8. 1350
8. 1350
3.1350
3.1350
8.1353
3. 1350
8. 1353
3. 1350
0. 1350
3. 1353
0. 1358
3.1353
3. 1353
0. 1353
3,1350
8. 1350
3. 1358
0. 1353
3. 1350
3. 1358
St.II-Nation
(NO EX)
(No Insp. )
Total
Incidence Due
To Benzene
0.2315
3.2345
3. 1584
0. 1319
8. 1272
3.1244
3. 1228
3. 1211
0. 1179
0. 1152
0.1145
0.1129
0.1113
0.1397
0.1088
3.1088
0. 1880
3. 1083
8. 1880
3. 13S8
3.1388
8. 1880
8. 1083
3.1330
3. 13£3
3.1380
3. 1080
3. 1380
0. 1380
3. 1383
3. 1388
3. 1088
3. 1383
3. 1383
Onboard-Nation
(w/ Tasoer. )
(flnn. Inso. )
Total
Incidence Due
To Benzene
3.2316
3.2255
3.2039
3. 1842
3. 1587
3. 1537
3.1421
3. 1317
0.1200
3.1116
8.1342
0.0977
3.9919
8.3655
3.3818
8.0818
0.0818
3.3818
3.3818
0. 2818
3. 3318
3.0813
3.3818
3.3818
8.3813
0.3318
8. 0818
0.3813
3.3818
3.3818
3.3818
8. 3818
3. 3813
3. -3818
3.3818.
5.55
2.13
5.11
1.99
5.31
5.10
1.73
4.17
1.47
3.77
1.54
F-61
-------
"acls F-S3; -Ssrvice Stat::-n Ircidsncs Sue To Benasne find Gas Vaaors - In-usa, Mo InsDections
1S23
* "27
--W/
.538
1339
.5%
1931
1?:2
1S33
122*
«9SS
!9S=
19S7
less
1939
£823
£?*i
£332
cees
ia?4
2M5
28-36
£237
•sea
2^
281S
sen
22!£
£213
sei4
£315
=216
£817
£318
£319
£223
Baseline
ffir.n. Insc. 5
Total
Incidence Due
"o Sas Vapors
-------
Table F-13. Service Station iraidsnce Dus To Benzene find Sas Vapors - In-use. No Inspections
Year
133S
1337
1SS3
195?
1932
19?1
13-32
1533
1934
1935
1996
1997
1993
1393
mm
£081
2832
2803
2M4
2805
2205
2827
£038
2839
2310
2311
2312
2313
2814
£215
£815
2317
2818
2013
2323
Baseline
(Ann. Insp. )
Total'
Incidence Due
To Gas Vapors
«'Pll,iiica)
4.13
4.34
3.93
3.32
3.74
3.S6
3.51
3.55
3.47
3.42
3.37
3.32
3.28
3.23
3.18
3.13
3.13
3.13
3.18
3.13
3.18
3.13
3.18
3.13
3.18
3.18
3.13
3.:8
3.13
3.18
3.18
3.13
3.13
3.18
3.13
St.II-NA*
(EX)
(No InsD.)
Total
Irsciderice Due
To Sas Vapors
(PUL,sice}
4.22
3.73
3.65
3.55
3.48
3.41
3.36
3.32
3.23
3.18
3.14
3.09
3.35
3.30
2.96
2.96
£.36
2.95
2.95
2.96
2.36
2.36
2.36
2.96
2.96
2.36
2.96
2.95
2.95
2.96
2.96
2.95
2.36
2.95
2.96
St.II-Mfl
(EX)
(No Inso.)
Total
Incidence Due
To Sas Vapors
(P'JL,mice)
3. as
3.34
3.19
3.38
3.02
2.95
2.91
2.87
2.S0
2.76
2.72
2.68
2.54
2.60
2.56
2.56
2.55
2.56
2.56
2.56
2.55
2.55
2.56
2.55
2.55
2.56
2.56
2.56
2.56
2.56 .
2.55
2.56
2.55
2.55
2,55
St.II-Nation
(EX)
(Mo Inso.)
Total
Incidence Due
To Sas vapors
(PUL, mice}
4.15
3.81
3.23
2.95
2.86
2.80
2.76
2.72
2.55
2.61
2.53
2.54
2.50
2.47
2.43
2.43
2.43
2.43
2.43
2.43
2,43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
St.II-Nation
(NO EX)
(No -Inso. )
Total
Incidence Due
To Sas Vapors
(PUL, slice)
4.15
3.67
2.85
2.33
2.29
2.24
2.21
2.18
2.12
2.139
2. '37
2.34
2.31
1.98
1.95
1.95
•1.95
1.95
1.95
1.95
1.95
1.95
1.95
1.95
1.35
1.35
1.35
1.35
1.95
1.95
1.95
1.95
1.95
1.55
1.95
Onboard-Nation
(w/ Tamoer, )
(flnn. Inso. )
Total
Incidence Due
To Sas Vanors
(PIL.aice)
4.15
4.04
3.56
3.31
3.134
2.73
2.57
2.39
2.13
2.34
1.30
1.79
1.69
1.53
.1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
117.35
35.12
139.48
35,55
95.19
31,23
91.59
31.31
75.16
26.43
53.79
27.33
F-63
-------
Tafaie ~-:3. Service Station Incidence Due To Berizene find Gas Vaoors - In-use, No Insoections
Yaar
• C3C
« .~C1
.;c<
:ssa
1989
1993
1«S!
i«X
1953
19:4
1S95
1595
1997
:938
1399
s?a>
2231
£222
2233
r?34
2205
226
2M7
•3S8
;eag
2318
2811
1812
2213
£314
£815
3?16
£817
$18
£819
•?28
Sift*
•43V =
Baseline
(flnn. !n=n. 5
Total
Incidence Dae
To Sas Vaacrs
f ME, rat)
3.S96
3.792
3.687
MI wQu
3.516
3.439
3.393
3.343
3.258
3.212
3.167
3.122
3.377
3.831
2.986
2.986
£.986
2.986
2.986
2.986
2.986
2.985
2.986
2.985
2.986
2.985
2.986
2.985
2.986
2.985
£.986
2.985
2. 586
2.985
2. '336
113.23
35.79
St. II-NA*
EX)
(No Inso.)
"otal
Incidence 3ue
To Sas Vaaors
(MLE.rat)
3.779
3.556
3.440
3.334
3.271
3.199
3.157
3.115
3.331
2.989
2.947
2.985
2.853
2.820
2.778
2.778
2.778
2.778
2.778
2.778
£.778
2.778
2.778
2.778
2.778
2.778
2.778
2.778
2.778
2.778
£.778
2.778
2.778
2.778
2.778
102.75
33.49
St. II-Nfl
(EX)
(No Inso. )
Total
Ir-cicenca Due
To Sas Vaoors
(MLE,;«at)
3.572
3.135
2.999
2.889
2.834
2.772
2.736
2.699
2.626
2.590
2.553
2.517
2.489
2.444
2.497
2.487
2.487
2.407
£.407
2.437
2.407
2.407
2.487
2.487
2.407
2.487
2.407
2.487
2.407
2.487
2.407
2.487
2.407
2.487
2.437
89.40
29.38
St. I I -Nat ion
(EX!
(No Inso.)
Total
Ircidencs Due
To Bas Vaoors
CMLE,rat)
3.895
3.583
3.881
2.758
2.686
2.627
2.593
2.558
2.489
2.454
2.420
2.385
2.351
2.316
2.232
2.282
2.282
2.282
2.282
£.282
2.282
2.282
2.282
2.282
2.282
£.282
2.282
2.282
2.282
2.282
2.232
2.282
2.282
2.282
2.282
86.12
29.13
St.II-Nation Onboard-Nation
(NO EX)
(No Inso. )
Total
Incidence Due
To Sas Vaoors
(MLE.rat)
3.896
3.443
2.674
2.232
2.153
2.186
£.878
2.351
1.995
1.367
1.948
1.912
1.884
1.857
1.829
1.329
1.829
1.329
1.329
1.829
1.829
1.829
1.829
1.829
1.329
1.329
1.329
1.829
1.829
1.829
1.329
1.329
1.329
1.829
1.829
78.68
24.82
(w/ Taffioer.)
(flnn. Inso. )
Total ,
Incidence Due
To Sas Vaoors
(ME, rat)
3.396
3.792
3.436
3.111
2.355
£.687
2.416
2.245
2.853
1.912
1.789
1.681
1.585
1.495
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
54.62
26.14
F-64
-------
Table F-13. Service Station Incinence Due To Benzene find Gas Vapors - In-use, No Inspections
Year
1"85
1*87
1588
-.389
1C32
- ~CJ.
1992
1553
139'
;9'r5
19%
1997
1998
1999
2808
2801
2082
2833
2384
£335
£886
£887
2888
2889
£818
£211
2012
313
2814
2315
2816
2817
2318
£319
2823
SUM =
NPV =
Baseline
(flrei. InsD. )
Total
iTiciusree Due
To 035 Vapors
«LE,aice)
£.79
2.72
2.54
2.57
2.52
2.46
. 2.43
2.48
2.33
£.38
2.27
2.24
2.20
2.17
2.14
2.14
2.14
2.14
2.14
2.14
2.14
2.14
£.14
£.14
2.14
2.14
2.14
£.14
2.14
£.14
2.14
2.14
2.14
2.14
£.14
78.97
£5.54
St.II-Nfl*
(EX)
(No Insp.)
Total
Incidence Due
To Sas Vapors
(MLE, fitice)
£.71
£.55,
2.46
2.39
£.34
2. £9
£.26
2.23
2.17
£.14
2.11
2.88
£.05
2.82
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
73.61
23.99
St.II-Nfl •
(EX)
(No Insp.)
Total
Incidence Due
To Sas Vaoors
(MLE, sice)
2.55
£.25
2.15
2.37
2.83
1.99
1.95
1.93
1.88
1.86
1.33
1.38
1.73
1.75
1.72
1.72
1.72
1.72
1.72
1.72
- 1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
64.85
21.35
St.II-Nation
(EX!
(No Inso.)
Total
Incidence Due
To Gas Vaoors
MLE, Bice)
2.79
2.57
2.21
1.98
1.92
1.88
1.85
1.83
1.78
1.76
1.73
1.71
1.68
1.56
1.63
1.53
1.63
1.63
. 1.63
1.53
1.63
1.53
1.63
1.63
1.63
1.63
1.63
1.63
1.53
1.63
1.63
1.53
1.63
1.53
1.53
51.78
20.87
St.II-Nation
(NO EX)
(No Inso.)
Total .
Incidence Due
To Gas Vaoors
(!4LE,3ice)
2.79
£.47
1.92
1.50
1.54
1.51
1.49
1.47
1.43
1.41
1.39
1.37
1.35
1.33
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
i.3i
1.21
1.31
58.58
17.78
Onboard-Nation
(w/ Tanner. )
iflnn. IKSD. )
Total
Incidence Due
To Sas Vaoors
(«LE,raice)
2.79
2.72
2.46
£.£3
2.35
1.87
1.73
i.51
1.47
1.37
1.28
1.28
1.14
1.07
1.81
1.01
1.31
1.31
1.01
1.31
1.81
1.81
1.01
1.31
1.01
1.31
1. 81
1.31
1.31
1.01
1.81
1.81
LSI
1.81
1.81
45. £9
18.72
F-65
-------
Table F-14. Self-Service Incidence Due To Benzene find Gas Vaoors - In-Use
ar
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2336
2901
2002
2303
2204
2S85
2006
2287
2008
2239
2810
2011
2912
2013
2914
2815
2016
2017
2818
2019
2820
Ifil s
PV 2
Baseline
Incidence
Due To
Benzene
(In-use, ftnn. 5
4.39
4.22
4.06
3.91
3.79
3.68
3.59
3.51
3.39
3.31
3.24
3.18
3.11
3.05
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
113.06
37.91
St.II-Nfl*
(EX)
Incidence
Due To
Benzene
(In-use, ftnn. )
4.24
3.93
3.76
3.51
3.50
3.40
3.32
3.24
3.13
3.06
3.00
2.93
2.87
2.31
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
104.60
35.23
St.II-Nfl
(EX)
Incidence
Due To
Benzene
(In-use, flnn.)
3.99
3.43
3.24
3.09
3.00
2.91
2.84
2.77
2.68
2.62
2.56
2.51
2.46
2.41
2.35
2.36
2.36
2.35
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.35
2.36
2.36
2.35
89.98
30.59
St.II-Nation
(EX)
Incidtsnce
Due To
Benzene
(In-use, flnn. )
4.39
3.63
2.35
1.62
1.49
1.45
1.41
1.38
' 1.33
1.30
1.28
1.25
1.23
1.20
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
49.95
20.05
St.II-Nation
(NO EX)
Incidence
Due To
Benzene
(In-use, ftnn. )
4.39
3.37
1.60
0.66
0.56
0.54
0.53
8.51
0.50
0.49
0.48
0.47
0.46
0.45
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
24.17
12.71
St. II-Nft*
(EX)
Incidence
Due To
Benzene
(In-use, No)
4.43
4.16
3.99
3.83
3.72
3.61
3.52
3.44
3.32
3.25
3.18
3.12
3.05
2.99
2.92
. 2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
111.02
37.33
St.II-Nfl
(EX)
Incidence
Due To
Benzene
(In-use, No)
4.27
3.84
3.65
3.50
3.39
3.29
. 3.22
3.14
3.03
-2.96
2.90
2.84
2.79
2.73
2.67
2.67
2.67
2.67
2.67
2.57
2.67
2.67
2.67
2.67
2.67
2.57
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.57
101.58
34.33
St.II-Nationl
(EX) 1
Incidence 1
Due To 1
Benzene 1
(In-use, No) 1
4.52J
3.951
3.861
2.531
2.401
2.33J
2.28J
2.221
2.15J
2.101
2.861
2.0il
1.971
1.931
1.89J
1.89J
1.891
1.391
1.39
1.89
1.89
1.89
1.39
1.39
1.39
1.89
1.89
1.89
1.89
1.39
1.89
1.89
1.89
1.89
1.89
TS.igl
27.37
F-66
-------
Table F-14. Self-Service Incidence Due To Benzene find Sas Vapors - In-Use
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2081
2802
2003
2024
2005
£906
2007
2008
2389
2310
2011
2012
2813
2014
2015
2016
2017
2018
2019
2020
iM =
'V =
St. II-Nation
(NO EX)
Incidence
Due To
Benzene
(In-use.No)
4.52
3.79
2.58
1.90
1.79
1.74
1.70
1.66
1.60
1.56
1.53
1.50
1.47
1.44
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
58.35
22.57
Onboard
W/ Taaoer.
Incidence
Due To
Benzene
(In-use.flnn.)
4.39
4.22
3.69
3.21
2.83
2.48
2.20
1.95
1.70
1.51
1.35
1.21
1.39
0.97
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
3.37
0.87
3.37
0.87
0.87
3.87
0.87
3.87
0.87
51.16
24.63
Baseline
Incidence
Due To
Gas Vapors
(PUL,rat>
(In-use,flnn.)
40.58
39.24
37.%
36.69
35.79
34.87
34.22
33.58
32; 54
31.95
31.41
30.36
30.32
29.78
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
St. II-Nfl*
(EX)
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use,flnn. )
39.24
36.54
35.14
33.87
33.04
32.19
31.59
. 31.00
30.04
29.49
28.99
28.49
27.99
27.49
26.98
25.98
26.98
25.98
25.98
25.98
25.98
26.98
25.98
25.98
25.98
26.98
26.98
26.93
26.98
26.98
26.98
26.96
26.33
26.98
26. SB
St. II-Nfl
(EX)
Incidence
Due To
Sas Vapors
(Pit, rat)
(In-use,flnn. )
35.92
31.88
30.28
28.99
28.28
27.55
•27.04
26.53
25.71
25.24
24.81
24.38
23.%
23.53
23.10
23. 10
23.10
23.10
23.10
23.10
23.10
23.10
23.10
23.10
23. 10
23.10
23.10
23.10
23.10
23.13
23. 10
23. 10 .
23.13
23. 10
23.13
St. II-Nation
(EX)
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use.flnn.)
40.62
33.72
21.92
15.24
14.08
13.72
13.47
13.21
12.81
12.57
12.36
12.14
11.93
11.72
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
.11.50
11.50
11.53
,11.53
11.50
11.50
11.53
St. II-Nation
(NO EX)
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use.flnn.)
40.62
31.33
15.00
6.22
5.24
5.11
5.02
4.92
4.77
4.68
4.60
4.52
4.44
4.36
4.28
4. 28
4.28
4.28
4.28
4.28
4.28
4.28
4.23
4.28
4.23
4.23
4.28
4.28
4.23
4.28
A. 23
4.23
4.23
4, £3
4.23
1093.69
351.01
1011.75
335.40
870.15
231.13
481.27
1SS.73
115.25
F-67
-------
Table F-:4. Self-Service Incidence Due To Benzsrs Snc- Sas Vapors - In-Use
ar
1986
1987
1588
1989
1950
1991
1992
1993
1994
1995
1995
1997
1998
1999
em
£901
3202
2883
2334
mm
£$86
2887
2808
£889
2813
2811
2812
2813
2014
2315
2816
2817
2818
2919
£828
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vaoors
CPUL,rat)
(In-use,No)
40.94
38.58
37.29
35.98
35. IB
34.19
33.56
32.93
31.91
31.33
38.88
30.25
29.73
29.28
28.67
28.67
28.67
28.67
28.67
28.57
£8.67
28.57
28.67
28.67
28.67
28.67
£8.67
28.67
£8.67
£8.67
£8.67
28.67
£6.67
£8.67
£8.67
St.IHB
(EX)
Incidence
Due To
Gas Vapors
iPUL,rat)
(In-use, No)
39.45
35.68
34.15
32.83
32.02
31.20
33.62
30.04
29.12
£8.59
28.10
27.61
27.13
26.64
26.16
26.15
£6.16
26.16
25.16
26.16
25.15
26.16
£5. 16
25.16
£5. 16
25.16
25.16
25.16
26.15
26.16
25.16
25.16
25.16
25.16
26.16
St.II-Nation
(EX)
Incidence
Due To
Sas Vapors
(?'JL,rat)
(In-use, No)
41.31
36.38
28.61
23.77
22.67
£2.09
21.58
21.27
£0.6£
£0.£4
19.98
19.55
19.21
18.86
18.52
18.52
18.52
IB. 52.
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
St.II-Nation
(NO EX)
Incidence
Due To
Gas Vapors
{?UL,rat)
(In-use,'fo)
41.81
35.24
24.89
17.88
16. '98
15.46
16. 15
15.86
- 15.37
15.89
14.83
14.57
14.32
14.06
13.89
13.8&
13.88
13.88
13.88
13.80
13.80
13.88
13.80
13.80
13.88
13.23
13.80
' 13.33
13.80
13.80
13. S0
13.68
13.80
13.88
13.30
Onboard
W/ Tamper.
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use, Arm. i
40. 58
3?. £4
34.23
£3.34
26. 33
23.34
23.44
18. 16
15.38
14.35
12.57
11.29
10. 17
9.12
8..22
8.22
8.22
8.22
3.22
3.22
8.22
8.22
3.2£
3.22
3.22
8.22
3.2E
8.22
8.22
8.22
8.22
8.22
8.22
8.22
3.22
Baseline
Incidence
Due "o
3as Vapors
'(PUL. slice)
24.63
£3.79
£3.81
£2.24
£1.73
21.1*
28.75
28.35
. 19.73
19.37-
19.24
18.71
18.35
13.35
17.72
17.72
17.72
17.72
•7.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17. 7£
17.72
17.72
St.II-Nfi*
(EX)
Ittcitiencs
Due To
Gas Vapors
(PULraicsi
£3.79
22.15
21.33
23.53
2J.03
19. 51
IS. 15
13.79
15.51
17.33
17.57
17.27
16.57
16.66
16.35
16.35
16.35
16.35
16. 35
15.35
iS.35
15.35
15.35
IS. 35
1S.3S
15.36
1S.3S
16.35
15,35
18.35
16. ZS
15.35
15.35
15.35
16.35
SUH «
NPV*
1073.91
355.48
982.47
325.32
725.%
259,81
552.49
£13.81
477.55
223.74
663.33
£18.33
61 -7 ~.r
-3. wO
263.33
F-68
-------
Table F-14. Self-Service Inciderce Due To Benzene Srid Gas Vasors - In-use
Year
1986
1987
1588
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2015
2017
2018
2019
2020
SUM =
NPV =
St.II-Nfl
(EX)
Incidence
Due To
Sas Vapors
(PUL,sice)
(In-use,flnn.)
22.38
19.33
18.35
17.57
17.14
16.70
•16.39
15.03
15.59
15.30
15.04
14.78
14.52
14.26
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.20
14.00
14.00
14.30
14.00
14.00
14.08
14.20
14.00
14.80
14.00
527.51
176. 49
St. II-Nation
(EX)
Incidence
Due To
Sas Vapors
(POL,r8ice)
(In-use,flnn.)
24.52
20.44
13.29
9.24
8.54
8.32
8.16
3.01
7.75
7.62
7.49
7.36
7.23
7.10
5.97
6.97
5.97
5.97
6.97
6.97
6.97
. 6.97
6.97
6.97
5.97
6.97
5.97
6.97
6.97
6.97
5.97
5.97
6.97
6.97
6.97
291.54
115.00
St. II-Nation
(NO EX)
Incidence
Due To
Sas Vapors
(PUL,rsice)
(In-use.Ann. )
24.52
18.99
9.09
3.77
3.18
3.10
3.04
2.98
2.39
2.84
2.79
2.74
2.59
2.65
2.68
2.59
2.60
2.50
2.60
2.60
2.60
2.60
2.60
2.50
2.60
2.50
2.60
2.50
2.50
2.60
2.50
2.60
2.50
2.60
2.50
139.92
72.30
St. II-Nft*
(EX)
Incidence
Due To
Sas Vapors
(PIL, mica)
(In-usBjNo)
24.32
23.45
22.61
21.81
21.28
20.73
20.35
19.55
• 19.35
18.59
18.67
18.35
18.02
17.70
17.38
17.38
17.38
17.33
17.38
17.38
17.38
17.38
17.38
17.33
17.38
17.38
17.38
17.33
17.38
17.33
17.38
17.38
17.38
17.38
17.38
651.04
215.46
St. II-Mfi
(EX)
Incidencs
Due To
Sas Vacors
(PUL,:;,ice)
(In-use,Nc)
23.51
21.53
20.78
19.98
19.41
13.91
18.55
13.21
17.65
17.33
17.04
16.74
16.45
16.15
15.85
15.86
15.36
15.36
15.36
15.86
15.35
15.86
15.36
15.86
15.36
15.86
15.85
15.86
15.86
15.86
15. 3S
15.86
15. 86
15.86
15.35
595.50
153. 13
St.II-Naticn
(EX)
Incidents
. 3u= To
3as Va;-crs
(?UL,5iics)
( la-use, No)
£3.34
22.31
17.34
14.41
13.75
13.39
13.14
12.90
12.53
12.27
12.25
11.85
11.54
11.44
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
440. 10
157.51
St. II-Naticn
(NO EX)
Incidence
Due ""o
Sas Vascrs
(PUL,:r.ica)
25.34
21.35
14.53
IS. 34
13.25
9. S3
9. S3
9.51
9.32
"9.15
3.59
8.83
3. S3
8.52
3.37
3.37
8.37
3.37
8.37
3.37
8.37
3.37
8.37
3.37
3.37
8.37
3.37
8.37
3.37
8.37
3.37
8.37
8.37
8.37
8.37
341.3S
129.52
Onboard
«7 Tanioer.
Incidence
Due TD
Sas Vacc-rs
( In-uss. fif.fi. :
Oi -»
'"2 7^
• 23.75
IS. 33
15.94
13.37
12.35
11.31
5.53
8.52
7.52
5.35
6.17
5.53
4.98
4.38
4.93
4; 93
4.98
4. SB
4.98
4.98
4.98
4.98
4.98
4.98
4.98
4.9S
4.98
4.53
4.98
4.58
4.98
4.98
4.93
289.58
133.57
F-69
-------
Table F-14. Self-service Incidence Due To Benzene find Sas Vanors - In-use
Year
Baseline
Incidence
Due To
Gas Vaoors
(HUE. rat)
St.II-Nfl*
(EX)
Incidence
Due To
Sas Vaoors
«MLE.rat)
St.!I-Nfl St.II-Nation St.II-Nation
(EX)
Incidence
Due To
Sas Vapors
(ME. rat)
tin-use, fern. ) (In-use, flnn. ) (In-use, flnn. )
1986
1987
1388
1989
1990
1991
1932
1993
1994
1995
19%
1997
1998
1999
em
2801
mz
2883
2384
gees
2005
2807
sm
2009
2910
2911
2312
2013
2014
2015
2016
2917
2018
2019
2020
SUM*
NPV =
23.11
22.34
21.61
20.89
20.38
19.85
19.49
19.12
18.53
18.19
17.88
17.57
17.26
16.%
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
522.75
205.55
22.34
20.81
20.01
19.28
18.81
18.33
17.99
17.65
17.10
16.79
16.51
16.22
15.34
15.65
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
576.10
158.98
21.82
18.15
17.24
16.51
16.10
15.69
15.40
15.11
14.54
14.37
14.13
13.88
13.54
13.40
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
455.47
(EX)
Incidence
Due To
Bas Vapors
(MLE,rat)
( In-use, flnn. )
S3. 13
19.20
12.48
8.58
8.02
7.81
7.67
7.52
7.29
7.16
7.04
5.91
6.79
6.67
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
5.55
6.55
6.55
5.55
6.55
6.55
273.92
165.77 108.31
(MI EX)
Incidence
Due To
Sas Vapors
(ME, rat)
( In-use, Ann. )
23.13
17.34
8.54
3.54
2.99
2.31
2.86
2.30
2.72
2.67
2.62
2.58
2.53
2.48
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
. 131.42
67.91
St.II-Nfl*
(EX)
Incidence
Due To
Sas Vaoors
(MLE.rat)
(In-use, No)
£3.31
•22.33
21.23
28.49
19.98
13.47
19.11
18.75
lfi.17
17.84
17.54
17.23
15.33
16.53
16.32
16.32
16.32
15.32
15.32
16.32
16.32
16.32
15.32
16.32
16.32
16.32
15.32
16.32
15.32
16.32
15.32
16.32
15.32
16.32
16.32
611.49
282.37
St.II-Nfl St.II-Nation
(EX)
Incidence
Due To
Sas Vaoors
(MLE,rat)
(In-use, No)
22.45
23.31
13.44
18.59
18.23
17.75
17.44
17.11
16.58
15.28
16.00
15.72
15.45
15.17
14.39
14.39
14.85
14.83
14.39
14.89
14.39
14.89
14,89
14.89
14.89
14.89
14.93
14.89
14. S9
14.83
14.83
14.39
14.89
14.89
14.83
559.42
186.39
(EX)
Incidence
Due To
Sas Vaoors
!XLE,f£t5
(In-use, No)
£3. S3 '
22. S5
IS. 2?
13.53
"} 0«
*C. J x
12.53
12. 34
12.11
11.74
11.52
11.33
11.13
18.94
10.74
13.54
13.54
13.54
10.54
13. 54
18.54
13.54
13.54
18. 54
10.54
13. 54
10.54
13.54
13.54
13.54
13.54
12.54
18.54
IS. 54
10.34
13.5^
413.37
147.94
F-70
-------
Table F-14. Self-Service Incidence Due To Servers find Sas Vaoors - In-uss
Year St.II-Nation Onboard
1986
1987
1988
1989
1998
1991
1992
1993
1994
1935
19%
1997
1998
1999
2000
2001
2002
2203
2004
2005
2006
2807
2098
2009
2010
2011
2012
2013
2014
2015
2015
2017
2018
2013
2020
11 =
'V =
(NO EX)
Incidence
Due To
Sas Vaoors
(ICE, rat)
(In-use,No)
23.80
20.06
13.72
10.18
9.62
9.37
9.20
9.03
8.75
8.59
8.44
8.39
8.15
8.01
7.86
7.86
7. Sfi
7.86
7.85
7.86
7.86
7.86
7.86
7.86
7.86
7.S6
7.86
7.85
7.85
7.86
7.86
7.86
7.86
7.86
7.86
320.29
121.74
NX Taraoer.
Incidence
Due To
Sas Vaoors
(ME, rat)
(In-use,flnn.)
23.11
22.34
19.52
16.99
14.98
13.12
11.64
13.34
9.08
8.01
7.16
6.43
5.79
5.19
4.68
4.68
4.68
4.68
4.68
4.68
4.68
4.58
4.68
4.68
4.68
4.58
4.68
4.68
4.68
4.58
4.58
4.58
4.58
4.68 '
4.68
271.92
139.24
Baseline St.II-*ifi* St.II-Nfl St.II-Nation St.II-Naiion
Incidence
Due To
Sas Vaoors
(In-use,flnn. )
15.55
15.01
15.48
14.37
14.60
14.22
13.96
13.70
13.28"
13.33
12.31
12.59
12.37
12.15
11.33
11.33
11.93
11.93
11.93
11.33
11.33
11.33
11.93
11.93
11.33
11.33
11.33
11.33
11.93
11.93
11.93
11.33
11.93
11.93
11.93
446. 15
147.27
(EX)
Incidence
Due To
Sas Vacors
(Ml£,aice)
(In-use,flnn. )
15.31
14.31
14.34
13.81
13.48
13.13
12.89
12.54
12.25
12.33
11.83
11.62
11.42
11.21
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.31
11.01
11.31
11.01
11.31
11.01
11.01
11.01
412.73
135.82
(EX)
Incidence
Due To
Sas Vapors
(*LE,Bice)
(In-use,firm. )
15.05
13.01
12.35
11.82
11.54
11.24
11.03
10.32
10.43
10.33
18.12
3.35
9.77
3,60
9.42
9.42
9.42
3.42
9.42
9.42
3.42
9.42
9.42
9.42
3.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9. «
9.42
9.42
354.95
113.75
(EX)
Incidence
Dua To
Gas Vapors
(ME, -.nee)
15.57
13.75
3.34
S.c2
5.75
5.50
5.49
5.39
5.22
5.13
5.84
4.95
4.37
4.78
4.63
4.63
4.59
4.69
4.59
4.53
4.69
4.63
4.53
4.55
4.59
4.63
4. S3
4.53
4.59
4.63
4.53
4.69
4.53
4.69
4. S3
135.24
77.33
(NO EX).
Incidence
Due To
Sas Vaoors
(MLE, 12 ice)
(In-use.Pnn. )
1 U« •-! .
12.75
5.12
£.54
£.14
2. S3
2.35
2.31
1.35
*. -•.
1.88
1.84
1.81
1.78
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
i.75
1.75
1.75
1.75
94.15
43.55
F-71
-------
Table F-14. Self-service Incidence Due To Benzene find Gas Vapors - In-use
ar
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
mi
2002
2003
2004
2005
2006
2807
2008
2009
2010
2011
2012
2013
2014
£015
2016
2017
2018
2019
2020
m =
3y =
St.II-Nft*
(EX)
Incidence
DUE To
Gas Vapors
(«LE,aice)
(In-use, No)
16.70
15.78
15.21
14.58
14.32
13.95
13.69
13.43
13.02
12.78
12.55
12.35
12.13
11.91
11.69
11.59
11.69
11.69
11.69
11.59
11.69
11.69
11.69
11.69
11.69
11.59
11.69
11.69
11.69
11.69
11.59
11.69
11.69
11.69
11.69
438.38
144.98
St.II-NA St.II-Nation
(EX)
Incidence
Due To
Gas Vanors
(«LE,nice)
(In-use, No)
16.09
14.55
13.93
13.39
13.06
12.73
12.49
12.25
11.88
11.55
11.46
11.25
11.07
10.87
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
18.67
10.67
10.57
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
480.78
133.32
(EX)
Incidence
Due To
Sas Vaoors
(MLE,aice)
(In-use, No)
17.35
15.31
11.67
9.59
9.25
9.31
8.84
8.58
8.41
8. £5
8.12
7.38
7.fl4
7.69
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
296. J.4-
105.99
St.II-Nation
(NO EX)
Incidence
Due To
Sas Vaoors
(MLE. aice)
tin-use, No)
17.35
14.37
3.83
7.29
6.89
5.72
6.59
6.47
6.27
6.15
6.05
5.94
5.84
5.74
5.63
5.53
5.63
5.63
5.63
5.53
5.63
5. -S3
5.53
5.63
5.63
5.63
5.53
5.53
5.63
5.53
5.63
5.63
5,63
5.63
5.63
223.46
87.22
Onboard
W/ Tamper.
Incidence
Due To
Sas Vaoors
(ME, mice)
(In-use, Ann. !
16.55
16.31
13.38
12.17
13.73
9.40
8.24
7.41
5.45
5.74
5.13
4.51
4.15
3.72
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
134.31
93.31
F-72
-------
TABLE F-15. VEHICLE OPERATIONS INCIDENCE DUE TO BENZENE
Without additional Evaporative Esissions-
flnnual flnnual flnnual
Incidence Incidence Incidence
No Bz Red 62.4* Bz Red 81.3* Bz Red
flnnual
Incidence
No Bz Red 62.4* Bz Red
-With Additional Evaporative EtnisBions-
finnual flnnual flnnual
Incidence Incidence Incidence
81.3* Bz Red Onboard
U/o Tamper.
flnnual
Incidence
Onboard
y/ Tamper.
12.13
11.53
19.99
10.52
10.13
9.76
9.59
9.45
9.36
9.34
9.37
9.37
9.39
9.42
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
11.72
11.18
10.63
13,26
9.91
9.57
'9.41
9.28
9.19
9.1fl
9.21
9.22
9.24
9.27
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
11.60
11.07
ia.se
18.18
9.84
9.50
9.35
9.22
9.14
9.13
9.17
9.17
9.28
9.23
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
338.44
107.52
332.29
105.16
330.39
184.44
12.35
11.74
11.20
10.73
10.34
9.97
9.80
9.55
9.57
9.56
9.59
9.59
9.61
9.64
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
346.16
109.81
11.86
11.32
10.82
10.39
10.04
9.69
9.53
9.40
9.32
9.31
9.34
9.35
9.38
9.40
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
336.91
106.57
11.72
11.19
10.71
10.29
' 9.95
9.61
9.45
9.33
9.25
9.23
9.27
9.28
9.30 .
9.33
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
334.98
105.58 •
12.23
11.63
11.%
10.58
10.17
9.79
9.60
9.45
9.35
9.32
9.34
9.33
9.35
9.37
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
337. S2
107.67
12.23
11.63
11.96
10.58
18. 17
9.79
9.60
9.45
9.35
9.33
9.35
9.34
9.36
9.38
9.39
9.39
9.39
9.39
9.39
9.39
9.39
9.39
3.39
9.39
3.39
9.39
9.39
9.39
3.39
9.39
3.39
9.39
3.33
9.39
9.39
337.91
107.73
F-73
-------
Table F-16. PREDICTED MAXIMUM BENZENE CONCENTRATION (jug/m?)
Facility
1986 1990 1995 2000
Terminals
Controlled
Uncontrolled
Bulk Plants
Controlled
Exempt
Uncontrolled
Service Station Inloading
Uncontrolled (300, -30)a
Uncontrolled (600, -30)*
3.756
22.91
0.3124
0.3151
1.191
0.1548
0.2322
Service Station Refueling (outloading)
Uncontrolled (300, -30)a
Uncontrolled (600, -30)a
0.2787
0.2039
3.406
20.74
0.2822
0.2846
1.076
0.1398
0.2097
0.2518
0.1842
3.120
18.98
0.2575
0.2598
0.9819
0.1276
0.1914
0.2298
0.1681
2.913
17.69
0.2395
0.2416
0.9137
(LI 188
0.1781
0.2139
0.1564
*The two numbers in parantheses are two grid points. The sum of service station
inloading and refueling (outloading) result in a maximum concentration that can
be at either point, depending on the control alternative.
F-74
-------
Table F-17. RISK FROM HIGH EXPOSURE TO BENZENE (x 10~7)
Facility
1986
1990
1995
2000
Total
70-yr. Lifetime
Terminals
Uncontrolled 22.2
Controlled 3.65
Bulk Plants
Uncontrolled 1.156
Controlled with exemption 0.3059
Controlled 0.3033
Service Station Inloading
Uncontrolled 0.229
Stage I 0.0195
Onboard 0.229
Stage II 0.218
Stage I & II 0.0122
Stage I & Onboard 0.0195
Stage II 4 Onboard 0.218
Stage I 5 II S Onooard 0.0122
Service Station Refueling (Outloadi'ng)
Uncontrolled
Stage I
Onboard
Stage II
Stage I &
Stage I 4
II
Onboard
Stage II & Onboard
Stage I 4 II & Onboard
Service Station Total
Uncontrolled
Stage I
Onboard
Stage II
Stage I 4 II
Stage I Si Onboard
Stage IIS Onboard
Stage I 4 II S Onboard
Self Service
0.201
0.275
0.201
0.0238
0.0326
0.275
0.0238
0.0326
0.430
0.294
0.430
0.242
0.0448
0.294
0.242
0.0448
20.1
3.31
1.045
0.2763
0.2740
0.207
0.0176
0.207
0.197
0.0110
0.0176
0.197
0.0110
0.182
0.248
0.131
0.0215
0.0294
0.179
0.0200
0.0273
0.388
0.266
0.338
0.218
0.0405
0.197
0.217
0.0384
18.4
3.03
0.9534
0.2523
0.2500
0.189
0.0161
0.189
0.180
0.0101
0.0161
0.180
0.0101
0.166
0.227
0.0633
0.0197
0.0269
0.0865
0.0165
0.0226
0.354
0.243
0.252
0.199
0.0369
0.103
0.196
0.0327
17.2
2.83
0.8872
0.2346
0.2325
0.176
0.0150
0.176
0.167
0.00937
0.0150
0.167
Ov00937
0.154 •
0.211
0.0280
0.0183
0.0250
0.0383
0.0145
0.0198
0.330
0.226
0.204
0.18S
0.0344
0.0533
0.182
0.0291
l,232a
203a
63.68a
16.84a
16.693
_b
_b
_b
_b
_b
_b
_b
_b
_b
_ij
_b
_b
_b
_b
_b
_b
23.7a
16.2a
15.5a
13.3a
2.473
5.10a
14.2C
3.38C
Uncontrolled"1
Stage Id
Onboard"
Stage IId
Stage I S IId
Stage I 4 Onboard4*
Stage II & Onboardd
Stage I 4 II 4 Onboardd
113
113
0.759
5.09
5.09
0.759
0.759
0.759
aTotal 70-year lifetime risk = 4 [(1986 Risk + 1990 Risk)/2] + 5 [(1990 Risk + 1995 Risk)/2]
+ 5 [(1995 Risk + 2000 Risk}/2] + 56 (2000 Risk)
'Total is not calculated here. Service station inloading and refueling = Service
Station Total.
cTotal 70-year lifetime risk for combinations of Stage II and Onboard
={4 [(1986 Risk +• 1990 Risk)/2] +• 5 [(1990 Risk +'1995 Risk)/2]
+ 5 [(1995 Risk + 2000 Risk)/2] + 2 (2000 Risk}} all with Stage II
+ 4 [(2000 Risk with Stage II + 2000 Risk without Stage ID/2] •
+ 50 [(2000 Risk without Stage II}]
dRisk fron self-service was not calculated for individual years, rather a 70-year lifetime
risk was calculated directly.
F-75
-------
Table F-18. RISK FROM HIGH EXPOSURE TO GASOLINE VAPORS
(PLAUSIBLE UPPER LIMIT UNIT RISK FACTOR) ( x 1C)-')
Facility
Terminals
Uncontrolled*
Controlled*
1986
437 or 704
70.0 or 115
Total
1990 1995 2000 70-year Lifetime
386 or 637 354 or 583 330 or 544 23,640 or 39,000
63.4 or 105 58.1 or 95.9 54.3 or 89.5 3,890 or 6,417
Bulk Plants
Uncontrolled*
Controlled with exemption*
Controlled*
Service Station Inloading
Uncontrolled*
Stage I*
Onboard*
Stage II*
Stage I 4 II*
Stage I S Onboard*
Stage II & Onboard*
Stage I 4 II 4 Onboard*
22.2 or 36.6
5.87 or 9.68
5.82 or 9.60
4.39 or 7.24
0.375 or 0.618
4.39 or 7.24
4.18 or 6.89
0.234 or 0.386
0.375 or 0.618
4.18 or 6.89
0.234 or 0.386
Service Station Refueling (Outloading)
Uncontrolled0
Stage Ij
Onboard0
Stage lie
Stage I 4 IIC
Stage I 4 Onboard0
Stage II 4 Onboard0
Stage I 4 II 4 Onboard6
Service Station Total
3.51 or 5.78
4.79 or 7.90
3.51 or 5.78
0.416 or 0.686
0.568 or 0.937
4.79 or 7.90
0.416 or 0.686
0.568 or 0.937
Uncontrolled
Stage I
Onboard
Stage II
Stage I 4 II
Stage I 4 Onboard
Stage II 4 Onboard
Stage I 4 II 4 Onboard
Self-Service
Uncontrgll6df
Stage I*
Onboard*
Stage IIf .
Stage I 4 II*
Stage I 4 Onboard^
Stage II 4 Onboardf
Stage I S II 4 Onboardr
7.89 or 13.0
5.17 or 8.52
7.89 or 13.0
4.59 or 7.58
0.802 or 1.32
5.17 or 8.52
4.59 or 7.58
0.802 or 1.32
20.1 or 33.1
5.30 or 8.75
5.26 or 8.67
3.97 or 6.54
0.339 or 0.558
3.97 or 6.54
3.77 or 6.23
0.212 or 0.349
0.339 or 0.558
3.77 or 6.23
0.212 or 0.349
3.17 or 5.22
4.33 or 7.14
2.29 or 3.77
0.376 or 0.619
0.513 or 0.847
3.13 or 5.16
0.349 or 0.575
0.477 or 0.787
7.13 or 11.8
4.67 or 7.70
6.25 or 10.3
4.15 or 6.85
0.725 or 1.20
3.47 or 5.72
4.12 or 6.80
0.688 or 1.14
18.3 or 30.2
4.84 or 7.98
4.80 or 7.91
3.62 or 5.97
0.309 or 0.510
3.62 or 5.97
3.45 or 5.68
0.193 or 0.319
0.309 or 0.510
3.45 or 5.68
0.193 or 0.319
2.89 or 4.77
3.95 or 6.52
1.10 or 1.82
0.343 or 0.565
0.469 or 0.773
1.51 or 2.49
0.289 or 0.476
0.394 or 0.651
6.51 or 10.7
4.26 or 7.03
4.72 or 7.79
3.79 or 6.25
0.662 or 1.09
1.82 or 3.00
3.73 or 6.16
0.588 or 0.969
17.0 or 28.1
4.50 or 7.42
4.46 or 7.36
3.37 or 5.56
0.288 or 0.474
3.37 or 5.56
3.21 or 5.29
0.180 or 0.296
0.288 or 0.474
3.21 or 5.29
0.180 or 0.296
2.69 or 4.44
3.68 or 6.06
0.488 or 0.805
0.319 or 0.526
0.436 or 0.719
0.668 or 1.10
0.252 or 0.416
0.345 or 0.569
6.06 or 9.99
3.96 or 6.54
3.86 or 6.36
3.53 or 5.82
0.616 or 1.02
0.955 or 1.58
3.46 or 5.70
0.524 or .865
1,222 or 2,015
323 or 533
320 or 528
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
M.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
M.A.b
H.A.b
N.A.b
435 or 717d
285 or 469d
293 or 484d
253 or 417d J
44.2 or 72.9d
90.9 or 150d
269 or 444e
60.7 or 100e
547 or 903C
547 or 903C
4.39 or 7.24C
24.6 or 40.6C
24.6 or 40.6C
4.39 or 7.24C
4.39 or 7.24°
4.39 or 7.24C
aRisk from high exposure to gasoline vapors = risk from high exposure to benzene x 1.71 or 9.42.
*>Not Applicable. Total is not calculated here. Service Station inloading + refueling = Service Station Total.
°Risk fro* high exposure to gasoline vapors = risk from high exposure to benzene x 1.56 or 8.56.
(fatal 70-year lifetime risk * 4 [(1986 Risk + 1990 Risk)/2] + 5 [(1990 Risk + 1995 Risk)/2]
+ 5 [(1995 Risk + 2000 Risk)/2] + 56 (2000 Risk).
*Total 70-year lifetime risk for options combining Stage II and Onboard
= { 4 [(1986 Risk + 1990 R1sk)/2] + 5 [(1990 Risk + 1995 R1sk)/2]
+ 5 [(1995 Risk + 2000 Risk)/2] + 2 (2000 Risk)} all with Stage II
+ 4 [(2000 Risk with Stage II + 2000 Risk without Stage ID/2]
+ 50 [(2000 Risk without Stage II)].
'Risk fro* self-service was not calculated for individual years, rather a 70-year lifetime risk was calculated directly.
F-76
-------
Table F-19. RISK FROM HIGH EXPOSURE TO GASOLINE VAPORS
(MAXIMUM LIKELIHOOD ESTIMATE UNIT RISK FACTOR) ( x lO'7)
Facility
Terminals
Uncontrolled3
Controlled3
Bulk Plants
Uncontrolled3
Controlled with exemption3
Controlled3
Service Station Inloading
Uncontrolled3
Stage I3
Onboard3
Stage II3
Stage I S II3
Stage I & Onboard3
Stage II S Onboard3
Stage I & II S Onboard3
Service Station Refueling
Uncontrolled0
Stage 1°
Onboard0
Stage 11°
Stage I & 11°
Stage I & Onboard0
Stage II $ Onboard0
Stage I 4 II 4 Onboard0
Service Station Total
Uncontrolled
Stage I
Onboard
Stage II
Stage I * II
Stage I 4 Onboard
Stage II & Onboard
Stage I S II 5 Onboard
Self-Service
Uncontrolled*1
Stage I*1
Onboard*1
Stage IIf
Stage I & IIf
Stage I & Onboard*1
Stage II 4 Onboard*1
Stage I S II 4' Onboard*1
1986
287 or 401
47.1 or 65.7
14.9 or 20.8
3.95 or 5.51
3.92 or 5.47
2.95 or 4.12
0.252 or 0.352
2.95 or 4.12
2.81 or 3.92
0.158 or 0.220
0.252 or 0.352
2.81 or 3.92
0.158 or 0.220
(Outloading)
2.36 or 3.29
3.22 or 4.50
2.36 or 3.29
0.280 or 0.390
0.382 or 0.534
3.22 or 4.50
0.280 or 0.390
0.382 or 0.534
5.31 or 7.41
3.48 or 4.85
5.31 or 7.41
3.09 or 4.32
0.540 or 0.754
3.48 or 4.85
3.09 or 4.32
0.540. or 0.754
_
-
1990
260 or 363
42.7 or 59.6
13.5 or 18.8
3.57 or 4.98
3.54 or 4.94
2.67 or 3.72
0.228 or 0.318
2.67 or 3.72
2.54 or 3.55
0.142 or 0.199
0.228 or 0.318
2.54 or 3.55
0.142 or 0.199
2.13 or 2.97
2.91 or 4.07
1.54 or 2.15
0.253 or 0.353
0.345 or 0.482
2.10 or 2.94
0.235 or 0.328
0.321 or 0.448
4.80 or 6.70
3.14 or 4.38
4.21 or 5.87
2.79 or 3.90
0.488 or 0.681
2.33 or 3.26
2.77 or 3.87
0.463 or 0.647
—
-
1995
238 or 332
39.1 or 54.6
12.3 or 17.2
3.26 or 4.55
3.23 or 4.51
2.44 or 3.40
0.208 or 0.290
2.44 or 3.40
2.32 or 3.24
0.130 or 0.181
0.208 or 0.290
2.32 or 3.24
0.130 or 0.181
1.94 or 2.71
2.66 or 3.71
0.743 or 1.04
0.231 or 0.322
0.315 or 0.440
1.02 or 1.42
0.194 or 0.271
0.265 or 0.370
4.38 or 6.11
2.87 or 4.00
3.18 or 4.44
2.55 or 3.56
0.445 or 0.622
1.22 or 1.71
2.51 or 3.51
0.395 or 0.552
-
2000
222 or 310
36.5 or 51.0
11.5 or 16.0
3.03 or 4.23
3.00 or 4.19
2.27 or 3.16
0.194 or 0.270
2.27 or 3.16
2.16 or 3.01
0.121 or 0.169
0.194 or 0.270
2.16 or 3.01
0.121 or 0.169
1.81 or 2.53
2.47 or 3.45
0.329 or 0.459
0.215 or 0.300
0.293 or 0.410
0.449 or 0.627
0.170 or 0.237
0.232 or 0.324
4.08 or 5.69
2.57 or 3.72
2.59 or 3.62
2.37 or 3.31
0.414 or 0.578
0.643 or 0.897
2.33 or 3.25
0.353 or 0.493
-
Total
70-year Lifetime
15,910 or 22,200
2,618 or 3,654
822 or 1,148
217 or 304
216 or 301
N.A.b
iv .n.
N.A.b
N.A.b
" "• .
N.A.D
'••"••
N.A.D
N.A.b
" "*.
N.A.b
N.A.b
k
N.A.b
l,.0«
N.A.b
U
N.A.b
" •"*«
N.A.b
N.A.b
NiA*.b
293 or 408d
191 or 267 d
197 or 27 5 d
170 or 238d
29.7 or 41 .5d .
61.2 or 85. 4d
181 or 253e
40.9 or 57e
368 or 514°
368 or 514°
2.95 or 4.12°
16.6 or 23.1°
16.6 or 23.1°
2.95 or 4.12°
2.95 or 4.12°
2.95 or 4.12°
aRisk from high exposure to gasoline vapors = risk from high exposure to benzene x 0.821 or 6.28.
bHot Applicable. Total is not calculated here. Service Station inloading + refueling = Service Station Total.
°Risk from high exposure to gasoline vapors = risk from high exposure to benzene x 0.747 or 5.72.
dTotal 70-year lifetime risk = 4 [(1986 Risk + 1990 R1sk)/2] + 5 [(1990 Risk + 1995 R1sk)/2]
+ 5 [(1995 Risk + 2000 Risk)/2] + 56 (2000 Risk).
STotal 70-year lifetime risk for Stage II in combination with Onboard
= | 4 [(1986 Risk + 1990 Risk)/2] + 5 [(1990 Risk + 1995 Risk)/2]
+ 5 [(1995 Risk + 2000 Risk)/2]}a11 with Stage II
+ 4 [(2000 Risk with Stage II + 2000 Risk without Stage ID/2]
+ 50 [(2000 Risk without Stage II)].
Risk from self-service was not calculated for individual years, rather a 70-year lifetime risk was calculated directly.
F-77
-------
Table F-20. RISK FROM HIGH EXPOSURE TO EDB (x 10~9)
Facility
Teralnals
Uncontrolled
Control 1 ed
Sulk Plants
Uncontrolled0
Controlled with exemption0
Controlled0
Service Station Inloadlng
Uncontrolled0
Stage 1°
Onboard0
Stage 11° .
Stage I 4 11°
Stage I 4 Onboard"
Stage II 4 Onboard0
Stage I 4 II 4 Onboard0
Service Station Refueling
Uncontrolled"1
Stage Id
Onboard^*6
Stage IId .
Stage 1.4 11° .
Stage I 4 Onboard'1.6
Stage II 4 Onboard1"-8
Stage I 4 II 4 Onboard"1.6
Service Station Total
Uncontrolled
Stage I
Onboard8
Stage II
Stage I 4 II
Stage I 4 Onboard6
Stage II 4 Onboard6
Stage I 4 II 4 Onboard6
Self-Service
Uncontrolled?
Stage IS
Onboard6-?
Stage 119
Stage I 4 119
Stage I 4 Onboard6. 3
Stage II 4 Onboard5 -9
Stage I 4 II 4 Onboard6. 9
1986
39.7
6.61
2.07
0.547
0.543
0.409
0.0349
0.409
0.389
0.0218
0.0349
0.389
0.0218
(Outloadlng)
0.369
0.504
0.369
0.0437
0.0598
0.504
0.0437
0.0598
0.778
0.539
0.778
0.433
0.0816
0.539
0.433
0.0816
.
-
-
-
-
1990
23.4
3.95
1.20
0.318
0.315
0.238
0.0203
0.238
0.226
0.0127
0.0203
0.226
0.0127
0.214
0.293
0.214
0.0254 '
0.0347
0:293
0.0254
0.0347
0.452
0.313
0.452
0.252
0.0474
0.313
0.252
0.0474
.
-
-
-
-
1995
9.59
1.74
0.493
0.131
0.129
0.0964
0.008?.
0.0964
0.0913
0.0052
0.0082
0.0092
0.0052
0.0869
0.119
0.0869
0.0103
0.0141
0.119
0.0103
0.0141
0.183
0.127
0.183
0.102
0.0192
0.127
0.102
0.0192
~
-
-
-
2000
1.95
0.498
0.081
0.021
0.021
0.0166
0.0014
0.0166
0.0158
0.0009
0.0014
0.0016
0.0009
0.0150
0.0205
0.0150
0.0018
0.0024
0.0205
0.0018
0.0024
0.0316
0.0219
0.0316
0.0176
0.0033
0.0219
0.0176
0.0033
-
—
-
-
-
Total
70-yr. Lifetime
347
68.8
16.8
4.43
4.39
.c
.c
.c
-c
.c
.c
.c
' _c
_c
_c
_c
_c
_c
_c
_c
6.36a
4.40a
6.36a
3.54«
0.667a
4.40a
4.27?
1.63?
196"
i QKn
iy o
1^6
3*33
8 23^
. 196"
8.83"
8.83"
Total 70-year lifetime risk - 4 [(1986 Risk * 1990 Rislc)/2] * 5 CU990 R1s* * 199S
* 5 CI1995 Risk + 2000 R1sk)/2] + 56 (2000 Risk)
Nlsk fro* high exposure to EDB - (risk fron high exposure to benzene) x (7.56 x 10"2) x (leaded gasoline
throughput 1n year/total gasoline throughput 1n year).
«Total Is not calculated here. Service Stations Inloadlng * refueling =• Service Station Total.
d!l1sk fron high exposure to EDB • (Risk from high exposure to benzene) x (7.76 x lO"2) x (leaded gasoline
throughput In year/total gasoline throughput In year).
*S1sk froa high exposure to EDB not affected by Onboard controls, because Onboard controls will be Installed only on
cars using unleaded gasoline only and because unleaded gasoline does not contain EDB.
Total 70-year Hfetlne risk for combinations of Stage II and Onboard
- 14 [(1986 Risk * 1990 R1sk)/2] + 5 [(1990 Risk * 1995 R1sk)/2]
+ 5 [(1995 Risk + 2000 R1sk)/2] * 2 (2000 Risk)} all with Stage II
». +4 [(2000 Risk with Stage II +.2000 Risk without Stage ID/2]
* 50 (2000 Risk without Stage II)
3R1SK from self service was not calculated for individual years, rather a 70-year lifetime risk was calculated directly.
"-otal risk fro. nigh exposure to EDB = total risk fron high exposure to benzene x 7.7,6 x ID'2, since assumed Individual
using only all leaaed gasoline.
F-78
-------
Table F-21. RISK FROM HIGH EXPOSURE TO EDC (x 10~9)
Facility
Terminals
Uncontrolled3
Controlled3
Bulk Plants
Uncontrolled3
Controlled with exemption3
Controlled3
Service Station Inloading
Uncontrolled3
Stage I3
Onboard3
Stage II3
Stage I 4 II3
Stage I 4 Onboard3
Stage I! 4 Onboard3
Stage I 4 II 4 Onboard3
Service Station Refueling
Uncontrolled1-
Stage Ic
Onboard0*1*
Stage IIC
Stage I 4 lie.
Stage I 4 Onboardc-d
Stage II 4 OnhoardC.d
Stage I 4 II 4 Onboardc'd
Service Station Total
Uncontrolled
Stage I
Onboard"
Stage II
Stage I 4 II
Stage I 4 Onboard1*
Stage II 4 Onboardd
Stage I 4 II 4 Onboardd
Self-Service
UncontrolledS
Stage 19
Onboarddi9
Stage 119
Stage I 4 119
Stage I 4 Onboardd-9
Stage II 4 Onboard11 -3
Stage I 4 II 4 Onboardd-9
1986
51.3
8.55
2.68
0.703
0.702
0.522
0.0446
0.522
0.497
0.0279
0.0446
0.497
0.0279
(Outloading)
0.476
0.651
0.476
O.OS65
0.0772
0.651
0.0565
0.0772
0.999
0.695
0.999
0.554
0.105
0.695
0.554
0.105
-
_
.
_
-
-
1990
30.2
5.11
1.55
0.411
0.408
0.304
0.0259
0.304
0.289
0.0162
0.0259
0.289
0.0162
0.277
0.378 '
• 0.277
0.0328
0.0449
0.378
0.0328
0.0449
0.580
0.404
0.580
0.322
0.0611
0.404
0.322
0.0611
.
_
_
_
-
-
1995
12.4
2.24
0.638
0.169
0.167
0.123
0.0105
0.123
0.117
0.0066
0.0105
0.117
0.0066
0.112
0.153
0.112
0.0133
0.0182
0.153
0.0133
0.0182
0.235
0.164
0.235
0.131
0.0248
0.164
0.131
0.0248
_
_
_
_
-
-
2000
2.53
0.644
0.105
0.028
0.028
0.0212
0.0018
0.0212
0.0202
0.0011
0.0018
0.0202
0.0011
0.0194
0.0265
0.0194
0.0023
0.0031
0.0265
0.0023
0.0031
0.0406
0.0283
. 0.0406
0.0225
0.0043
0.0283
0.0225
0.0043
_
_
_
„
•
-
Total
70-yr. Lifetime
448
89.0
21.7
5.73
5.68
_b
_b
_b
_b
_b
_b
_b
_b
_b
_b
_b
-b
_b
.b
.b
.0
8.16e
5.68e
8.166
4.536
0.8596
5.68«
5.47f
2.11f
253C
253<=
253C
11. 4=
11. 4C
253<=
11. 4C
11. 4C
Risk from high exposure to EDC » risk from high exposure to EDB x 0.992.
b
Total is not calculated here. Service station inloading and refueling = Service
Station Total.
Risk from high exposure to EDC = risk from nigh exposure to EDB x 0.989.
d
Risk from high exposure to EDC not affected by Onboard controls, because Onboard
controls will be installed only on cars using unleaded gasoline only and because unleaded
gasoline does not contain EDC.
STotal 70-year lifetime risk = 4 [(1986 Risk + 1990 Risk)/2] + 5 [(1990 Risk * 1995 Risk!/2]
5 [(1995 Risk * 2000 Risk)/2] + 56 (2000 Risk)
'Total 70-year lifetime risk for combinations of Stage II and Onboard
= |4 [(1986 Risk + 1990 Risk)/2] » 5 [(1990 Risk +• 1995 Risk 1/2]
+ 5 [(1995 Risk + 2000 Risk)/2] + 2 (2000 Risk)|all with Stage II
<• 4 [(2000 Risk with Stage :i + 2000 Risk without Stage ID/2]
i- 50 (2000 Risk without Stage II!
Risk from self-service was not calculated for individual years, rather a 70-year lifetime
risk was calculated directly.
F-79
-------
-------
APPENDIX G
CUMULATIVE VALUES OF CAPITAL AND ANNUALIZED COSTS
(1986-2020)
G-l
-------
-------
Table S-i. Baseline Marketing Ootion Costs
Tersiinal
Tensinal
Year
1982
1906
!937
1983
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2002
2083
2004
2885
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2320
Total Costs
NPV of Costs
Annual ized
Costs
($ Million)
0.0
5.8
17.9
24.6
25.0
25.6
25.9
26.2
26.8
27.1
27.4
22.4
22.7
23.0
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
Capital
Costs
($ Million!
0.0
85.4
85.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
71.3
71.3
0.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
71.3
71.3
0.0
0.0
9.0
0.0
0.0
0.0
0.0
0.0
71.3
71.3
0.0
3.0
Storage Tank Storage Tank Bulk Plant
Annualized Capital
Costs Costs
t$ Million) ($ Million)
Bulk Plant
789
214
598
221
$0
($1.9)
($5.6)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($7.5)
($246.6)
($65.1)
$0
$8.2
$8.2
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$8.2
$8.2
$0
$0
$0
$0
$0
$0
$0
$0
$0
$0
$32.7
$16.3
Annual i zed
Costs
($ Million)
(No ex.)
0.0
9.6
28.9
38.7
38.9
39.0
39.1
39.2
39.4
39.5
39.6
39.5
39.7
39.8
39.9
39.9
37.1
37.1
37.1
37.1
37.1
37.1
37.1
37.1
37,1
37.1
37.1
37.1
37.1
37.1
37.1
37.1
37. 1
37.1
37.1
1255
337
Capital
Costs
($ Million)
(No ex.)
0.0
113.6
113.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
106.5
106.5
0.8
0.0
0.0
0.0
0.0
0.0
0.8
0.0
0.9
0.0
0.0
0.0
0.0
136.5
186.5
0.0
0.0
653
, 252
G-3
-------
Table G-l. Gasoline Marketina Ootion Costs
r
1982
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
28W
2001
2882
2003
2004
20d5
2806
2007
2888
2389
2810
2811
2812
2813
2814
2315
2816
2817
2818
2019
2828
flnnualized
Costs
(S Million)
(Ex.)
0.0
5.7
17.4
23.5
23.6
23.8
23.9
24.0
24.3
24.4
24. S
24.6
24.7
24.8
24.9
24.9
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
Capital
Costs
($ Million)
(Ex.)
8.8
79.7
79.7
0.0
0.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.8
0.0
75.6
75.6
0.0
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
75.6
75.6
0.0
0.0
Total Costs
NPV of Costs
781
208
462
177
0.0
2.8
8.3
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
10.8
10.8
10.8
10.8
10.8
.10.6
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
361
96
0.0
18.3
10.3
0.8
0.0
0.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
9.2
9.2
0.8
8.0
0.0
0.0
8.0
8.8
0.0
0.8
0.8
0.8
0.0
0.0
0.0
9.2
9.2
8.0
0.8
57
23
Bulk Plant
Trucks
Annual i2ed
Costs
($ Million)
(No Ex.)
31.1
8. a
7.8
23.4
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
30.4
39.4
30.4
39.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
38.4
30.4
Bulk Plant
Trucks
Capital
Costs
<* Million)
(No Ex.)
0.0
28. 5
28. 5
0.0
0.0
0.0
0.0
- 0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.9
25.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.9
25.9
0.0
0.0
1014
278
161
&3
6-4
-------
Table 6-1. Gasoline Marketing (lotion Costs
Year
1982
1985
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2880
2001
£002
2083
2004
2305
2008
£087
2008
£389
2810
2011
2012
2813
2014
2015
2016
2017
2018
2019
2020
Total Costs
NPV of Costs
8.0
4.5
13.5
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
17.6
17.6
17.6
17.6
17.5
17.6
17.6
17.6
17.6
17.6
17.5
17.5
17.6
17.5
17.6
17.6
17.6
17.6
17.6
585
156
0.0
16.5
16.5
0.0
8.0
0.0
8.0
8.0
0.0
8.0
8.0
0.0
0.0
8.0
0.0
0.8
14.9
14.9
0.0
8.0
0.0
8.0
0.0
0.0
0.0
8.0
0.8
0.8
0.8
0.8
0.0
14.9
14.9
8.8
0.0
93
36
0.8
16.9
50.8
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
57.7
67.7
57.7
67.7
67.7
67.7
67.7
67.7
57.7
67.7
67.7
57.7
67.7
67.7
67.7
67.7
67.7
2234
598
8.0
168.8
168.0
0.8
0.0
8.8
0.0
8.0
8.0
0.0
0.0
0.0
0.0
8.0
8.0
8.0
168.8
168.8
0.0
0.0
0.0
0.0
0.0
8.0
0.0
8.8
0.0
8.8
0.8
0.8
0.0
168:8
158.8
8.8
8.8
1088
378
0.0
7. 1
21.3
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
938
248
0.0
78.6
70.6
8.0
8.8
0.8
0.0
8.8
0.0
8.0
0.9
0.8
0.0
8.8
0.0
0.0
78.6
70.6
0.0
8.8
0.0
0.0
0.0
8.8
0.0
0.0
8.0
0.0
0.8
0.0
8.0
. 78.5
78.6
8.0
0.0
423
159
G-5
-------
Table G-i. Gasoline Marketing Ootion Costs
tr
1982
1985
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2082
2003
2904
2005
20%
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Ser. Sta.
Stage MB
ftnnualized
Costs
($ Million)
(No Ex.)
2.4
1.2,
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
wSi • t3 6 a*
Stage I-Nfl
Capital
Costs
W Billion)
(No Ex.)
5.9
5.9
0.0
0.0
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.9
5.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.9
5.9
0.0
0.0
0.0
Ser. Sta.
Stage I-Nft
ftnnualized
Costs
($ Million)
(Ex.)
1.0
0.5
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
1.0
1.0
1.0
'l.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
1.0
1.0
1.0
1.0
Ser. Sta.
Stage I-Nft
Capital
Costs
<$ Million)
(Ex.)
2.5
2.5
0.8
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.5
2.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.5
2.5
0.0
0.0
0.8
Total Costs
NPV of Costs
81
24
35
14
34
10
15
&
G-6
-------
Table S-2. Stage II and Onboard Ootion Costs
YEAR
1986
1937
1988
1989
1998
1991
1992
1993
1994
.1595
1996
1997
1998
1999
2008
2081
sm
2803
2884
2885
2886
2887
2888
2889
2810
2811
2812
2813
2814
2815
2816
2817
2818
2019
2020
Total Costs
NPV of Costs
St. II
Nation.
(No Ex. )
Net firm.
Cost
($ Si 11 ion)
8
129
388
537
554
557
558
568
563
565
566
568
569
571
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
$18,718
$4,869
St. II
Nation.
(No Ex. )
Cap. Cost
Total
($ Million)
0
1,156
1,155
134
8
8
8
8
8
113
118
14
0
0
0
8
1,039
1,156
238
14
0
0
0
0
0
118
118
14
8
0
0
1,039
1,039
238
118
$7,823
$2.847
St. II
Nation.
(Ex.)
Net Ann.
Cost
($ Million)
8
42
128
175
182
184
185
186
188
189
190
192
193
194
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
$6,323
$1,628
St. II
Nation.
(Ex.)
Can. Cost
Total
($ Million)
0
444
444
47
0
0
8
0
0
41
41
4
0
0
0
0
403
444
85
4
0
0
0
0
0
41
41
4
0
8
0
403
403
85
41
$2,977
$1.086
St. II
fill m
(No Ex.)
Net flnn.
Cost
($ Million)
72
150
168
155
166
157
167
168
169
169
170
178
171
171
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
$5,841
$1,672
St. II
All m
(No Ex.)
Cap. Cost
Total
($ Million)
653
48
48
8
8
0
0
0
66
4
4
8
0
0
0
587
102
48
4
8
0
0
0
8
66
4
4
0
0
0
587
35
102
4
4
$2,350
$977
St. II
All Nfl
(Ex.)
Net Ann.
Cost
($ Million)
24
49
52
54
54
55
55
56
55
57
57
58
58
- 58
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
$1,373
$558
St. II
All Nfl
(Ex. )
Cao. Cost
Total
($ Million)
253
14
14
0
0
0
0
8
24
1
1
0
0
0
0
229
37
14
i
0
0
8
0
0
24
i
1
8
0
0
229
13
37
i
1
$895
$373
G-7
-------
Table G-2. Stage II and Onboard Ootion Costs
YEflR
($
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
19%
1999
2888
2882
£083
2884
2085
2886
2007
2883
2009
2010
2011
2012
2013
2014
2015
2016
2817
2018
2019
2028
Total Costs
HPV of Costs
St. II
Sel. NA
(No Ex.)
Net ftnn.
Cost
Million)
27
57
61
63
63
63
64
64
64
64
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
$2,220
$636
St. II
Sel. Nfl
(No Ex.)
Cao. Cost
Total
($ Million)
248
15
15
0
0
8
0
8
25
2
2
0
8
0
0
223
39
15
2
0
0
0
0
8
25
2
2
8
8
8
223
14
39
2
2
$893
$371
St. II
Sel. Nfl
(Ex.)
Net Ann.
Cost
($ Million)
9
19
20
21
21
21
21
21
21
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
$758
$212
St. II ONBOflRD ONBOARD
Sel. Nfl CAPITAL
(Ex.)
Cap. Cost ($
Total
($ Million)
%.8
5.4
!i.4
0.0
0.0
0.0
0.0
8.0
9.8
8.4
0.4
0
0.0
19.0
8.0
8S. 9
14.8
5.4
8.4
8.0
0.0
8.0
0.8
0.0
9.8
8.4
0.4
0.0
0.8
8.0
86.9
5.8
14.0
8.4
0.4
$348
$142
COSTS
Million) ($
0
8
281
281
198
283
287
287
286
288
288
288
288
288
288
288
288
288
208
288
268
288
288
288
288
288
288
288
288
288
288
288
288
288
288
$6,836
$1.787
ANNUAL
COSTS
Million)
0
0
32
64
%
129
162
195
228
261
294
328
338
339
341
342
342
342
343
343
343
343
343
343
343
343
343
343
343
343
343
343
343
343
343
$9,666
$1.921
St. II Coai.
Nation.
(No Ex. )
Net Ann.
Cost
($ Million)
8
129
399
563
592
604
615
624
634
641
648
654
659
663
667
669
519
211
28
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
$9,521
$4,261
St. II Coa.
Nation.
(No Ex.)
Cap. Cost
Total
($ Million)
8
1,156
1,156
134
8
8
ei
0
8
118
118
14
8
8
0
O
&
0
0
0
8
8
8
0
0
8
8
0
0
0
8
8
0
8
8
$2,695
$2,287
6-3
-------
Table G-2. Stage II and Onboard Ootion Costs
3R
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Costs
Costs
St. II COH.
Nation.
(Ex.)
Net flnn.
Cost
($ Billion)
0
42
135
195
209
218
226
. 232
239
245
250
254
257
261
263
265
206
83
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$3,589
$1,575
St. II COM.
Nation.
(Ex.)
Can.' Cost
Total
($ Million)
0
444
444
47
0
0
0
0
0
41
41
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$1,022
$842
St. II Cora.
fill Nfl
(No Ex. 5
Net flnn.
Cost
($ Million)
72
150
164
173
177
181
184
187
190
192
194
1%
198
199
200
116
21
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$2,803
$1,429
St. II Cos.
fill Nfl
(No Ex. )
Cap. Cost
Total
($ Million)
653
40
40
0
0
0
0
0
66
4
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$808
$757
St. II Com.
911 Nfl
(Ex.)
Net flnn.
Cost
($ Million)
24
49
55
60
63
65
68
70
72
73
75
76
77
78
79
45
8
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$1,040
$518
St. II Coo.
fill Nfl
(Ex.)
Cao. Cost
Total
($ Million)
253
14
14
0
0
0
0
0
24
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$307
$289
St. II COIB.
Sel. Nfl
(No Ex. )
Net flnn.
Cost
($ Million)
27
57
62
66
67
69
70
71
72
73
74
75
75
76
76
44
8
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$1,065
$543
St. II Cos.
Sel. Nfl
(No Ex.)
Can. Cost
Total
($ Million)
248
15
15
0
0
0
0
0
25
2
2
0
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$387
$288
G-9
-------
Table 6-2. Stage II and Onboard (lotion Costs
YEAR
St. II Cos.
Sal. m
(Ex.)
Net ftnn.
Cost
($ Million)
St. II Cca.
Sel. Nfl
(Ex.)
Cap. Cost
Total
($ Million)
1986
1987
1988
1989
1998
1991
1992
1993
1994
1995
19%
1997
19%
1999
2808
2801
2802
2883
2884
2885
2086
2887
2888
2818
2811
2812
2813
2814
2815
2816
2817
2818
2819
2828
9
19
21
23
24
25
26
26
27
28
28
29
29
38
38
17
3
1
8
8
8
8
0
8
0
0
0
0
0
0
8
8
8
8
%.0
5.4
5.4
8.0
8.0
0.8
0.0
0.0
9.0
8.4
0.4
8.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
0
8
0
8
0
0
0
0
0
0
0
Total Costs
(PV of Costs
$395
$197
$117
$110
G-10
-------
TfiBLE
IiM-U5E STAGE II
YEflR
TOTflL NET flNNUftL
COSTS($MM)
NOTION (BI-flNNUflL)
(EX.)
TOTflL NET ftNNUfiL
COSTS($MK)
NflTIQN !BI-flNNUflL)
(NO EX!
YEAR
TOTflL NET flNNUftL
COSTS($MM)
NOTION (MINIMflL)
(EX.)
TOTftL NET fiNNUftL
COSTS($M!"i)
NATION (MINIMflL)
(NO EX)
198£
1966
1987
1988
1939
1998
1991
1992
1993
1994
1995
19%
1997
1998
1999
mm
2001
2882
2083
2884
2085
2886
2087
2888
2889
2810
2811
2012
2013
2014
2815
2816
2817
2818
2819
2828
—
46
139
192
198
199
288
281
283
284
285
206
207
208
289
209
289
209
289
209
289
209
209
289
289
289
289
289
289
289
289
289
289
289
289
135
485
558
576
578
588
581
584
585
586
^flfl
WWW
589
590
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
591
198£
1986
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2888
2081
2082
2803
2884
2885
2886
2087
2888
2089
2810
2811
2812
2013
2014
2815
2016
2017
2018
2819
2820
29
92
139
158
165
166
167
168
169
169
170
171
171
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
82
261
396
451
469
472
473
475
476
477
477
478
479
480
480
488
480
488
480
488
480
480
488
488
488
438
488
488
488
488
488
488
488
488
TOTflL COSTS
Wi
6.790
1,757
19357
5848
TOTflL COSTS
(*»))
5,546
1,486
15553
3966
G-ll
-------
YEflR
TOTAL NET ANNUflL
COSTSHMi)
NflTION (QUflRTERLY)
(EX.)
TOTflL NET flNNUflL
COSTS($MM)
NOTION (QUflRTERLY)
(NO EX)
YEAS
ma
1986
1987
1S88
1983
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
em
2891
2882
2093
2934
2935
2986
2037
2098
2099
2019
2911
2912
2913
2014
2015
2016
2017
2018
2919
2029
43
131
181
186
IBS
189
199
193
194
195
1%
197
198
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
199
131
393
543
561
563
565
566
569
571
572
574
575
577
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
198£
1986
19S7
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2001
2902
2093
2004
2005
2006
2007
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
TOTAL NET flNNUflL
COSTS (*!*!)
NflTION (flNNUflL)
(EX.)
TOTAL NET ANNUAL
COSTS («"!«)
NATION (ANNUAL)
(NO EX)
44
134
185
190
192
193
194
196
197
198
199
200
201
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
132
397
548
566
568
570
571
574
576
577
578
580
581
583
583
583
583
533
583
583
583
583
583
583
583
583
583
583
583
583
583
583
583
583
TOTftL COSTS
t«w<)
NPV
6,462
1,667
18902
4922
TOTflL COSTS
(«MM)
NPVCSMH)
6,571
1.697
19053
4964
G-12
-------
YEflR
TDTflL NET flNMJfiL
COSTS(*MM)
flLL Nfl (MIN)
(EX.)
1982
1956
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2(981
20192
2804
2885
2007
2008
2010
2011
2012
2313
2014
2015
2016
2017
2018
2019
2820
22
45
47
49
49
50
50
50
50
51
51
51
51
51
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
TQTfiL NET fiNNUflL
COSTS («ffl)
ftLL Nfi (MIN)
(NO EX)
YEflR
61
128
136
140
141
141
141
142
142
143
143
143
143
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
144
TOTfiL NET flNMJflL
COSTS ($WI)
flLL Nft (BI-flNNUflL)
(EX.)
1982
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2081
2003
2004
2005
2006
2087
2008
2009
2310
2011
2012
2013
2814
2015
2016
2017
2018
2019
26
53
56
58
58
59
59
59
60
60
60
61
61
61
62
62
52
62
62
62
62
62
62
62
62
62
62
62
62
62
62
62
62
62
62
TOTflL NET flNNLML
COSTS<$MM>
ALL m (BI-flNNUflL)
(NO EX)
75
155
165
171
171
172
172
173
174
174
174
175
175
176
176
176
176
176
176
176
176
176
176
176
176
176
176
175
176
176
176
176
176
176
176
TOTflL COSTS
($M«)
1,730
4850
TOTflL COSTS
2,068
5923
NPV (JW)
500
1411
($»»
593
1721
6-13
-------
YEflR
TQTflL NET ftNNUftL
COSTS <«*«
fiLL m (WML)
(EX.)
TQTftL NET ftNNUflL
COSTS($HM)
ai N
(NO EX)
YEflR
1982
1986
1987
1963
1989
1993
1991
1993
1993
1994
1995
199S
1997
1998
1999
2008
2802
2§83
2884
2805
2006
2908
2811
2912
£813
2014
2015
2916
2817
2918
2019
2929
25
51
54
56
56
57
57
58
58
59
59
59
60
60
60
60
60
60
60
60
69
60
60
60
60
60
60
60
60
60
60
60
60
60
60
74
153
163
168
169
173
178
170
171
172
172
173
173
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
174
TOTfiL NET flNNUAL
COSTS($MM)
flLL Nfl (QUART.)
(EX.)
TOTfiL NET flNNUAL
COSTS($MM)
fill Nfl (QUfiRT.)
(NO EX)
198E
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2001
2002
2003
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
24
50
53
55
55
56
56
57
57
58
58
58
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
73
151
161
167
167
168
169
169
178
178
171
171
172
172
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
TOTftL COSTS
NPV ($SM)
2,009
577
5852
1699
TOTftL COSTS
1,975
566
5805
1684
G-14
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-84-012a
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Evaluation of Air Pollution Regulatory Strategies
for the Gasoline Marketing Industry
5. REPORT DATE
July 1984
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS "
Director, Office of Air Quality Planning and Standards,
Director, Office of Mobile Sources
U.S. Environmental Protection Agency
Washington, D.C. 20460
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3060
12. SPONSORING AGENCY NAME AND ADDRESS
Assistant Administrator for Air and Radiation
U.S. Environmental Protection Agency
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
I.6S1
The gasoline marketing industry (bulk terminals, bulk plants, service station
storage tanks, and service station vehicle refueling operations) emit to the atmos-
phere several organic compounds of concern. These include: volatile organic
compounds (VOC)s>which contribute to ozone formation; benzene, which has been listed
as a hazardous air pollutant based on human evidence of carcinogencity; and ethylene
dichloride (EDC), ethylene dibromide (EDB), and gasoline vapors, for which there is
animal evidence of carcinogencity. This report contains an analysis of the health,
emission, cost, and economic impacts of several regulatory strategies for addressing
organic compound emissions from gasoline marketing sources. The regulatory strategic:
considered are: (1) service station controls (Stage II) for vehicle refueling
emissions only in areas requiring additional VOC control to attain the national ozone
ambient standard; (2) service station controls (Stage II) for vehicle refueling
emissions on a nationwide basis; (3) Onboard vehicle controls for vehicle refueling
emissions on a nationwide basis; (4) bulk terminal, bulk plant, and service station
storage tank controls on a nationwide basis; and (5) various permutations and
combinations of these alternatives.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPSN ENDED TERMS
c. COS AT I Field/Group
Gasoline
Air Pollution
Pollution Control
Stationary Sources
Mobile Sources
Volatile Organic Compounds Emissions
Benzene Emissions
Air Pollution Control
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS I This Report)
21. NO. OF PAGES
Unclassified
644
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rav. 4-77) PREVIOUS EDITION is OBSOLETE
-------
------- |